Initial model upload

README.md

@@ -0,0 +1,202 @@
+---
+base_model: openai/whisper-small
+library_name: peft
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.15.1

adapter_config.json

@@ -0,0 +1,37 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": {
+    "base_model_class": "WhisperForConditionalGeneration",
+    "parent_library": "transformers.models.whisper.modeling_whisper"
+  },
+  "base_model_name_or_path": "openai/whisper-small",
+  "bias": "none",
+  "corda_config": null,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "q_proj",
+    "v_proj"
+  ],
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_rslora": false
+}

adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:05c62ba82b49d5530cb59a6660f691542857584997a011ed9991d3fb8d3a93e9
+size 14176064

checkpoint-1282/README.md

@@ -0,0 +1,202 @@
+---
+base_model: openai/whisper-small
+library_name: peft
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.15.1

checkpoint-1282/adapter_config.json

@@ -0,0 +1,37 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": {
+    "base_model_class": "WhisperForConditionalGeneration",
+    "parent_library": "transformers.models.whisper.modeling_whisper"
+  },
+  "base_model_name_or_path": "openai/whisper-small",
+  "bias": "none",
+  "corda_config": null,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "q_proj",
+    "v_proj"
+  ],
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_rslora": false
+}

checkpoint-1282/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:82b2e92985d3234a43c270820c04a56fd7d33c101571a985789bfcc3d72940e7
+size 14176064

checkpoint-1282/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4406a14dda35216d56130bd385ed45523e6ef9311405f7a871124af1d2639a61
+size 28432570

checkpoint-1282/preprocessor_config.json

@@ -0,0 +1,14 @@
+{
+  "chunk_length": 30,
+  "feature_extractor_type": "WhisperFeatureExtractor",
+  "feature_size": 80,
+  "hop_length": 160,
+  "n_fft": 400,
+  "n_samples": 480000,
+  "nb_max_frames": 3000,
+  "padding_side": "right",
+  "padding_value": 0.0,
+  "processor_class": "WhisperProcessor",
+  "return_attention_mask": false,
+  "sampling_rate": 16000
+}

checkpoint-1282/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:47d11c8794792252f10f0d90106a215a71081e57908e27af545a6ca3bffac4e3
+size 14244

checkpoint-1282/scaler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:83d6806c71e0146991e30f44773eeb503604e704766e0fd5e522689319ba0acf
+size 988

checkpoint-1282/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4bc09cf0b7de1c8eefda4c397cba706f60486de08de54f5db5e1b8f67f3d4295
+size 1064

checkpoint-1282/trainer_state.json

@@ -0,0 +1,938 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 1.0,
+  "eval_steps": 500,
+  "global_step": 1282,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.0078003120124804995,
+      "grad_norm": 2.144841432571411,
+      "learning_rate": 0.0009989971027412525,
+      "loss": 1.1174,
+      "step": 10
+    },
+    {
+      "epoch": 0.015600624024960999,
+      "grad_norm": 1.0625462532043457,
+      "learning_rate": 0.0009978827724537553,
+      "loss": 0.5035,
+      "step": 20
+    },
+    {
+      "epoch": 0.0234009360374415,
+      "grad_norm": 1.619573950767517,
+      "learning_rate": 0.0009967684421662581,
+      "loss": 0.3841,
+      "step": 30
+    },
+    {
+      "epoch": 0.031201248049921998,
+      "grad_norm": 3.2681403160095215,
+      "learning_rate": 0.000995654111878761,
+      "loss": 0.4974,
+      "step": 40
+    },
+    {
+      "epoch": 0.0390015600624025,
+      "grad_norm": 1.5944786071777344,
+      "learning_rate": 0.0009945397815912637,
+      "loss": 0.4797,
+      "step": 50
+    },
+    {
+      "epoch": 0.046801872074883,
+      "grad_norm": 1.8234444856643677,
+      "learning_rate": 0.0009934254513037665,
+      "loss": 0.637,
+      "step": 60
+    },
+    {
+      "epoch": 0.054602184087363496,
+      "grad_norm": 3.790844440460205,
+      "learning_rate": 0.0009923111210162691,
+      "loss": 0.4479,
+      "step": 70
+    },
+    {
+      "epoch": 0.062402496099843996,
+      "grad_norm": 2.9351046085357666,
+      "learning_rate": 0.000991196790728772,
+      "loss": 0.4332,
+      "step": 80
+    },
+    {
+      "epoch": 0.07020280811232449,
+      "grad_norm": 2.2770025730133057,
+      "learning_rate": 0.0009900824604412747,
+      "loss": 0.4399,
+      "step": 90
+    },
+    {
+      "epoch": 0.078003120124805,
+      "grad_norm": 1.86591637134552,
+      "learning_rate": 0.0009889681301537775,
+      "loss": 0.407,
+      "step": 100
+    },
+    {
+      "epoch": 0.08580343213728549,
+      "grad_norm": 1.6851640939712524,
+      "learning_rate": 0.0009878537998662805,
+      "loss": 0.3862,
+      "step": 110
+    },
+    {
+      "epoch": 0.093603744149766,
+      "grad_norm": 2.5469818115234375,
+      "learning_rate": 0.0009867394695787831,
+      "loss": 0.4166,
+      "step": 120
+    },
+    {
+      "epoch": 0.10140405616224649,
+      "grad_norm": 1.8210259675979614,
+      "learning_rate": 0.000985625139291286,
+      "loss": 0.3785,
+      "step": 130
+    },
+    {
+      "epoch": 0.10920436817472699,
+      "grad_norm": 2.031057119369507,
+      "learning_rate": 0.0009845108090037887,
+      "loss": 0.4177,
+      "step": 140
+    },
+    {
+      "epoch": 0.11700468018720749,
+      "grad_norm": 1.6646612882614136,
+      "learning_rate": 0.0009833964787162915,
+      "loss": 0.402,
+      "step": 150
+    },
+    {
+      "epoch": 0.12480499219968799,
+      "grad_norm": 1.8680285215377808,
+      "learning_rate": 0.0009822821484287943,
+      "loss": 0.3282,
+      "step": 160
+    },
+    {
+      "epoch": 0.13260530421216848,
+      "grad_norm": 1.8039604425430298,
+      "learning_rate": 0.0009811678181412971,
+      "loss": 0.3536,
+      "step": 170
+    },
+    {
+      "epoch": 0.14040561622464898,
+      "grad_norm": 3.3018901348114014,
+      "learning_rate": 0.0009800534878538,
+      "loss": 0.4595,
+      "step": 180
+    },
+    {
+      "epoch": 0.1482059282371295,
+      "grad_norm": 3.684013843536377,
+      "learning_rate": 0.0009789391575663027,
+      "loss": 0.4288,
+      "step": 190
+    },
+    {
+      "epoch": 0.15600624024961,
+      "grad_norm": 1.4512592554092407,
+      "learning_rate": 0.0009778248272788055,
+      "loss": 0.5086,
+      "step": 200
+    },
+    {
+      "epoch": 0.16380655226209048,
+      "grad_norm": 2.3981761932373047,
+      "learning_rate": 0.0009767104969913081,
+      "loss": 0.4084,
+      "step": 210
+    },
+    {
+      "epoch": 0.17160686427457097,
+      "grad_norm": 3.7943010330200195,
+      "learning_rate": 0.000975596166703811,
+      "loss": 0.4524,
+      "step": 220
+    },
+    {
+      "epoch": 0.1794071762870515,
+      "grad_norm": 2.657606840133667,
+      "learning_rate": 0.0009744818364163138,
+      "loss": 0.3592,
+      "step": 230
+    },
+    {
+      "epoch": 0.187207488299532,
+      "grad_norm": 2.7629363536834717,
+      "learning_rate": 0.0009733675061288166,
+      "loss": 0.4263,
+      "step": 240
+    },
+    {
+      "epoch": 0.19500780031201248,
+      "grad_norm": 1.3749983310699463,
+      "learning_rate": 0.0009722531758413193,
+      "loss": 0.48,
+      "step": 250
+    },
+    {
+      "epoch": 0.20280811232449297,
+      "grad_norm": 2.648716449737549,
+      "learning_rate": 0.0009711388455538221,
+      "loss": 0.416,
+      "step": 260
+    },
+    {
+      "epoch": 0.21060842433697347,
+      "grad_norm": 1.5672308206558228,
+      "learning_rate": 0.0009700245152663249,
+      "loss": 0.4223,
+      "step": 270
+    },
+    {
+      "epoch": 0.21840873634945399,
+      "grad_norm": 2.618163585662842,
+      "learning_rate": 0.0009689101849788277,
+      "loss": 0.4172,
+      "step": 280
+    },
+    {
+      "epoch": 0.22620904836193448,
+      "grad_norm": 3.6365268230438232,
+      "learning_rate": 0.0009677958546913305,
+      "loss": 0.5501,
+      "step": 290
+    },
+    {
+      "epoch": 0.23400936037441497,
+      "grad_norm": 2.740039825439453,
+      "learning_rate": 0.0009666815244038332,
+      "loss": 0.3553,
+      "step": 300
+    },
+    {
+      "epoch": 0.24180967238689546,
+      "grad_norm": 3.406210422515869,
+      "learning_rate": 0.000965567194116336,
+      "loss": 0.3518,
+      "step": 310
+    },
+    {
+      "epoch": 0.24960998439937598,
+      "grad_norm": 1.4707075357437134,
+      "learning_rate": 0.000964452863828839,
+      "loss": 0.3452,
+      "step": 320
+    },
+    {
+      "epoch": 0.2574102964118565,
+      "grad_norm": 1.608324408531189,
+      "learning_rate": 0.0009633385335413417,
+      "loss": 0.4908,
+      "step": 330
+    },
+    {
+      "epoch": 0.26521060842433697,
+      "grad_norm": 4.090480327606201,
+      "learning_rate": 0.0009622242032538444,
+      "loss": 0.4597,
+      "step": 340
+    },
+    {
+      "epoch": 0.27301092043681746,
+      "grad_norm": 2.2214395999908447,
+      "learning_rate": 0.0009611098729663472,
+      "loss": 0.4552,
+      "step": 350
+    },
+    {
+      "epoch": 0.28081123244929795,
+      "grad_norm": 1.9134166240692139,
+      "learning_rate": 0.00095999554267885,
+      "loss": 0.3571,
+      "step": 360
+    },
+    {
+      "epoch": 0.28861154446177845,
+      "grad_norm": 1.8127851486206055,
+      "learning_rate": 0.0009588812123913528,
+      "loss": 0.3808,
+      "step": 370
+    },
+    {
+      "epoch": 0.296411856474259,
+      "grad_norm": 2.2262885570526123,
+      "learning_rate": 0.0009577668821038556,
+      "loss": 0.4099,
+      "step": 380
+    },
+    {
+      "epoch": 0.3042121684867395,
+      "grad_norm": 2.8041303157806396,
+      "learning_rate": 0.0009566525518163583,
+      "loss": 0.3988,
+      "step": 390
+    },
+    {
+      "epoch": 0.31201248049922,
+      "grad_norm": 6.797432899475098,
+      "learning_rate": 0.0009555382215288611,
+      "loss": 0.4728,
+      "step": 400
+    },
+    {
+      "epoch": 0.31981279251170047,
+      "grad_norm": 3.1861369609832764,
+      "learning_rate": 0.000954423891241364,
+      "loss": 0.3502,
+      "step": 410
+    },
+    {
+      "epoch": 0.32761310452418096,
+      "grad_norm": 2.9223642349243164,
+      "learning_rate": 0.0009533095609538667,
+      "loss": 0.4215,
+      "step": 420
+    },
+    {
+      "epoch": 0.33541341653666146,
+      "grad_norm": 6.848895072937012,
+      "learning_rate": 0.0009521952306663694,
+      "loss": 0.415,
+      "step": 430
+    },
+    {
+      "epoch": 0.34321372854914195,
+      "grad_norm": 9.054282188415527,
+      "learning_rate": 0.0009510809003788722,
+      "loss": 0.4667,
+      "step": 440
+    },
+    {
+      "epoch": 0.35101404056162244,
+      "grad_norm": 2.3005900382995605,
+      "learning_rate": 0.0009499665700913752,
+      "loss": 0.6565,
+      "step": 450
+    },
+    {
+      "epoch": 0.358814352574103,
+      "grad_norm": 2.9467573165893555,
+      "learning_rate": 0.000948852239803878,
+      "loss": 0.3837,
+      "step": 460
+    },
+    {
+      "epoch": 0.3666146645865835,
+      "grad_norm": 1.5977652072906494,
+      "learning_rate": 0.0009477379095163808,
+      "loss": 0.407,
+      "step": 470
+    },
+    {
+      "epoch": 0.374414976599064,
+      "grad_norm": 2.8274600505828857,
+      "learning_rate": 0.0009466235792288835,
+      "loss": 0.4501,
+      "step": 480
+    },
+    {
+      "epoch": 0.38221528861154447,
+      "grad_norm": 1.8566502332687378,
+      "learning_rate": 0.0009455092489413863,
+      "loss": 0.3146,
+      "step": 490
+    },
+    {
+      "epoch": 0.39001560062402496,
+      "grad_norm": 2.5871951580047607,
+      "learning_rate": 0.0009443949186538891,
+      "loss": 0.3335,
+      "step": 500
+    },
+    {
+      "epoch": 0.39781591263650545,
+      "grad_norm": 2.0552711486816406,
+      "learning_rate": 0.0009432805883663919,
+      "loss": 0.3698,
+      "step": 510
+    },
+    {
+      "epoch": 0.40561622464898595,
+      "grad_norm": 1.5244548320770264,
+      "learning_rate": 0.0009421662580788946,
+      "loss": 0.581,
+      "step": 520
+    },
+    {
+      "epoch": 0.41341653666146644,
+      "grad_norm": 1.5146633386611938,
+      "learning_rate": 0.0009410519277913974,
+      "loss": 0.3634,
+      "step": 530
+    },
+    {
+      "epoch": 0.42121684867394693,
+      "grad_norm": 2.394819736480713,
+      "learning_rate": 0.0009399375975039002,
+      "loss": 0.338,
+      "step": 540
+    },
+    {
+      "epoch": 0.4290171606864275,
+      "grad_norm": 2.3500325679779053,
+      "learning_rate": 0.000938823267216403,
+      "loss": 0.4025,
+      "step": 550
+    },
+    {
+      "epoch": 0.43681747269890797,
+      "grad_norm": 2.4186370372772217,
+      "learning_rate": 0.0009377089369289058,
+      "loss": 0.4095,
+      "step": 560
+    },
+    {
+      "epoch": 0.44461778471138846,
+      "grad_norm": 1.9770065546035767,
+      "learning_rate": 0.0009365946066414085,
+      "loss": 0.4275,
+      "step": 570
+    },
+    {
+      "epoch": 0.45241809672386896,
+      "grad_norm": 1.7679632902145386,
+      "learning_rate": 0.0009354802763539114,
+      "loss": 0.3949,
+      "step": 580
+    },
+    {
+      "epoch": 0.46021840873634945,
+      "grad_norm": 2.0794620513916016,
+      "learning_rate": 0.0009343659460664142,
+      "loss": 0.4208,
+      "step": 590
+    },
+    {
+      "epoch": 0.46801872074882994,
+      "grad_norm": 2.677424192428589,
+      "learning_rate": 0.000933251615778917,
+      "loss": 0.4089,
+      "step": 600
+    },
+    {
+      "epoch": 0.47581903276131043,
+      "grad_norm": 1.526112675666809,
+      "learning_rate": 0.0009321372854914197,
+      "loss": 0.4196,
+      "step": 610
+    },
+    {
+      "epoch": 0.4836193447737909,
+      "grad_norm": 1.8656370639801025,
+      "learning_rate": 0.0009310229552039225,
+      "loss": 0.3732,
+      "step": 620
+    },
+    {
+      "epoch": 0.4914196567862715,
+      "grad_norm": 3.3338847160339355,
+      "learning_rate": 0.0009299086249164253,
+      "loss": 0.4231,
+      "step": 630
+    },
+    {
+      "epoch": 0.49921996879875197,
+      "grad_norm": 2.1057350635528564,
+      "learning_rate": 0.0009287942946289281,
+      "loss": 0.3921,
+      "step": 640
+    },
+    {
+      "epoch": 0.5070202808112324,
+      "grad_norm": 1.544977068901062,
+      "learning_rate": 0.0009276799643414309,
+      "loss": 0.3748,
+      "step": 650
+    },
+    {
+      "epoch": 0.514820592823713,
+      "grad_norm": 3.4070258140563965,
+      "learning_rate": 0.0009265656340539336,
+      "loss": 0.4027,
+      "step": 660
+    },
+    {
+      "epoch": 0.5226209048361935,
+      "grad_norm": 5.5486931800842285,
+      "learning_rate": 0.0009254513037664364,
+      "loss": 0.4326,
+      "step": 670
+    },
+    {
+      "epoch": 0.5304212168486739,
+      "grad_norm": 3.6824769973754883,
+      "learning_rate": 0.0009243369734789392,
+      "loss": 0.4809,
+      "step": 680
+    },
+    {
+      "epoch": 0.5382215288611545,
+      "grad_norm": 6.3154778480529785,
+      "learning_rate": 0.000923222643191442,
+      "loss": 0.3588,
+      "step": 690
+    },
+    {
+      "epoch": 0.5460218408736349,
+      "grad_norm": 3.133465528488159,
+      "learning_rate": 0.0009221083129039447,
+      "loss": 0.4712,
+      "step": 700
+    },
+    {
+      "epoch": 0.5538221528861155,
+      "grad_norm": 4.222598552703857,
+      "learning_rate": 0.0009209939826164475,
+      "loss": 0.4135,
+      "step": 710
+    },
+    {
+      "epoch": 0.5616224648985959,
+      "grad_norm": 4.6125078201293945,
+      "learning_rate": 0.0009198796523289504,
+      "loss": 0.393,
+      "step": 720
+    },
+    {
+      "epoch": 0.5694227769110765,
+      "grad_norm": 6.543318748474121,
+      "learning_rate": 0.0009187653220414532,
+      "loss": 0.6138,
+      "step": 730
+    },
+    {
+      "epoch": 0.5772230889235569,
+      "grad_norm": 2.596463680267334,
+      "learning_rate": 0.000917650991753956,
+      "loss": 0.3694,
+      "step": 740
+    },
+    {
+      "epoch": 0.5850234009360374,
+      "grad_norm": 2.428490161895752,
+      "learning_rate": 0.0009165366614664587,
+      "loss": 0.4535,
+      "step": 750
+    },
+    {
+      "epoch": 0.592823712948518,
+      "grad_norm": 1.8790688514709473,
+      "learning_rate": 0.0009154223311789615,
+      "loss": 0.3986,
+      "step": 760
+    },
+    {
+      "epoch": 0.6006240249609984,
+      "grad_norm": 3.141587734222412,
+      "learning_rate": 0.0009143080008914643,
+      "loss": 0.3881,
+      "step": 770
+    },
+    {
+      "epoch": 0.608424336973479,
+      "grad_norm": 2.125810146331787,
+      "learning_rate": 0.0009131936706039671,
+      "loss": 0.4225,
+      "step": 780
+    },
+    {
+      "epoch": 0.6162246489859594,
+      "grad_norm": 2.532404661178589,
+      "learning_rate": 0.0009120793403164698,
+      "loss": 0.3423,
+      "step": 790
+    },
+    {
+      "epoch": 0.62402496099844,
+      "grad_norm": 3.6324350833892822,
+      "learning_rate": 0.0009109650100289726,
+      "loss": 0.4205,
+      "step": 800
+    },
+    {
+      "epoch": 0.6318252730109204,
+      "grad_norm": 1.4804415702819824,
+      "learning_rate": 0.0009098506797414754,
+      "loss": 0.4042,
+      "step": 810
+    },
+    {
+      "epoch": 0.6396255850234009,
+      "grad_norm": 1.5140562057495117,
+      "learning_rate": 0.0009087363494539782,
+      "loss": 0.3056,
+      "step": 820
+    },
+    {
+      "epoch": 0.6474258970358814,
+      "grad_norm": 2.470576047897339,
+      "learning_rate": 0.000907622019166481,
+      "loss": 0.391,
+      "step": 830
+    },
+    {
+      "epoch": 0.6552262090483619,
+      "grad_norm": 3.4496209621429443,
+      "learning_rate": 0.0009065076888789837,
+      "loss": 0.4163,
+      "step": 840
+    },
+    {
+      "epoch": 0.6630265210608425,
+      "grad_norm": 1.8823250532150269,
+      "learning_rate": 0.0009053933585914866,
+      "loss": 0.3877,
+      "step": 850
+    },
+    {
+      "epoch": 0.6708268330733229,
+      "grad_norm": 2.492297410964966,
+      "learning_rate": 0.0009042790283039894,
+      "loss": 0.3542,
+      "step": 860
+    },
+    {
+      "epoch": 0.6786271450858035,
+      "grad_norm": 3.977569341659546,
+      "learning_rate": 0.0009031646980164922,
+      "loss": 0.4168,
+      "step": 870
+    },
+    {
+      "epoch": 0.6864274570982839,
+      "grad_norm": 3.938462495803833,
+      "learning_rate": 0.0009020503677289949,
+      "loss": 0.4894,
+      "step": 880
+    },
+    {
+      "epoch": 0.6942277691107644,
+      "grad_norm": 1.3457701206207275,
+      "learning_rate": 0.0009009360374414977,
+      "loss": 0.4903,
+      "step": 890
+    },
+    {
+      "epoch": 0.7020280811232449,
+      "grad_norm": 6.3473124504089355,
+      "learning_rate": 0.0008998217071540005,
+      "loss": 0.4766,
+      "step": 900
+    },
+    {
+      "epoch": 0.7098283931357254,
+      "grad_norm": 3.145792245864868,
+      "learning_rate": 0.0008987073768665033,
+      "loss": 0.4119,
+      "step": 910
+    },
+    {
+      "epoch": 0.717628705148206,
+      "grad_norm": 1.809446930885315,
+      "learning_rate": 0.0008975930465790061,
+      "loss": 0.3991,
+      "step": 920
+    },
+    {
+      "epoch": 0.7254290171606864,
+      "grad_norm": 1.7960044145584106,
+      "learning_rate": 0.0008964787162915088,
+      "loss": 0.3095,
+      "step": 930
+    },
+    {
+      "epoch": 0.733229329173167,
+      "grad_norm": 2.9710285663604736,
+      "learning_rate": 0.0008953643860040116,
+      "loss": 0.5104,
+      "step": 940
+    },
+    {
+      "epoch": 0.7410296411856474,
+      "grad_norm": 2.460524797439575,
+      "learning_rate": 0.0008942500557165144,
+      "loss": 0.4332,
+      "step": 950
+    },
+    {
+      "epoch": 0.748829953198128,
+      "grad_norm": 1.6166704893112183,
+      "learning_rate": 0.0008931357254290172,
+      "loss": 0.3856,
+      "step": 960
+    },
+    {
+      "epoch": 0.7566302652106084,
+      "grad_norm": 1.747750163078308,
+      "learning_rate": 0.0008920213951415199,
+      "loss": 0.3973,
+      "step": 970
+    },
+    {
+      "epoch": 0.7644305772230889,
+      "grad_norm": 1.4469414949417114,
+      "learning_rate": 0.0008909070648540227,
+      "loss": 0.3344,
+      "step": 980
+    },
+    {
+      "epoch": 0.7722308892355694,
+      "grad_norm": 4.661273956298828,
+      "learning_rate": 0.0008897927345665256,
+      "loss": 0.6003,
+      "step": 990
+    },
+    {
+      "epoch": 0.7800312012480499,
+      "grad_norm": 3.6588950157165527,
+      "learning_rate": 0.0008886784042790284,
+      "loss": 0.473,
+      "step": 1000
+    },
+    {
+      "epoch": 0.7878315132605305,
+      "grad_norm": 2.208383798599243,
+      "learning_rate": 0.0008876755070202809,
+      "loss": 0.3899,
+      "step": 1010
+    },
+    {
+      "epoch": 0.7956318252730109,
+      "grad_norm": 2.3569576740264893,
+      "learning_rate": 0.0008865611767327836,
+      "loss": 0.3871,
+      "step": 1020
+    },
+    {
+      "epoch": 0.8034321372854915,
+      "grad_norm": 2.7071454524993896,
+      "learning_rate": 0.0008854468464452864,
+      "loss": 0.3149,
+      "step": 1030
+    },
+    {
+      "epoch": 0.8112324492979719,
+      "grad_norm": 2.8024532794952393,
+      "learning_rate": 0.0008843325161577892,
+      "loss": 0.3328,
+      "step": 1040
+    },
+    {
+      "epoch": 0.8190327613104524,
+      "grad_norm": 3.0969290733337402,
+      "learning_rate": 0.000883218185870292,
+      "loss": 0.3948,
+      "step": 1050
+    },
+    {
+      "epoch": 0.8268330733229329,
+      "grad_norm": 2.982484817504883,
+      "learning_rate": 0.0008821038555827947,
+      "loss": 0.4323,
+      "step": 1060
+    },
+    {
+      "epoch": 0.8346333853354134,
+      "grad_norm": 3.133814573287964,
+      "learning_rate": 0.0008809895252952975,
+      "loss": 0.4393,
+      "step": 1070
+    },
+    {
+      "epoch": 0.8424336973478939,
+      "grad_norm": 3.3123364448547363,
+      "learning_rate": 0.0008798751950078003,
+      "loss": 0.3244,
+      "step": 1080
+    },
+    {
+      "epoch": 0.8502340093603744,
+      "grad_norm": 2.308555841445923,
+      "learning_rate": 0.0008787608647203032,
+      "loss": 0.424,
+      "step": 1090
+    },
+    {
+      "epoch": 0.858034321372855,
+      "grad_norm": 3.654137134552002,
+      "learning_rate": 0.000877646534432806,
+      "loss": 0.3445,
+      "step": 1100
+    },
+    {
+      "epoch": 0.8658346333853354,
+      "grad_norm": 2.149843692779541,
+      "learning_rate": 0.0008765322041453087,
+      "loss": 0.3398,
+      "step": 1110
+    },
+    {
+      "epoch": 0.8736349453978159,
+      "grad_norm": 3.1334431171417236,
+      "learning_rate": 0.0008754178738578115,
+      "loss": 0.3333,
+      "step": 1120
+    },
+    {
+      "epoch": 0.8814352574102964,
+      "grad_norm": 2.2942090034484863,
+      "learning_rate": 0.0008743035435703143,
+      "loss": 0.4188,
+      "step": 1130
+    },
+    {
+      "epoch": 0.8892355694227769,
+      "grad_norm": 2.0195343494415283,
+      "learning_rate": 0.0008731892132828171,
+      "loss": 0.4047,
+      "step": 1140
+    },
+    {
+      "epoch": 0.8970358814352574,
+      "grad_norm": 2.3850839138031006,
+      "learning_rate": 0.0008720748829953198,
+      "loss": 0.3931,
+      "step": 1150
+    },
+    {
+      "epoch": 0.9048361934477379,
+      "grad_norm": 1.6200228929519653,
+      "learning_rate": 0.0008709605527078226,
+      "loss": 0.409,
+      "step": 1160
+    },
+    {
+      "epoch": 0.9126365054602185,
+      "grad_norm": 2.9001989364624023,
+      "learning_rate": 0.0008698462224203254,
+      "loss": 0.4289,
+      "step": 1170
+    },
+    {
+      "epoch": 0.9204368174726989,
+      "grad_norm": 1.52889883518219,
+      "learning_rate": 0.0008687318921328282,
+      "loss": 0.3575,
+      "step": 1180
+    },
+    {
+      "epoch": 0.9282371294851794,
+      "grad_norm": 2.359733819961548,
+      "learning_rate": 0.000867617561845331,
+      "loss": 0.3837,
+      "step": 1190
+    },
+    {
+      "epoch": 0.9360374414976599,
+      "grad_norm": 2.3807597160339355,
+      "learning_rate": 0.0008665032315578337,
+      "loss": 0.4206,
+      "step": 1200
+    },
+    {
+      "epoch": 0.9438377535101404,
+      "grad_norm": 1.8366179466247559,
+      "learning_rate": 0.0008653889012703365,
+      "loss": 0.3101,
+      "step": 1210
+    },
+    {
+      "epoch": 0.9516380655226209,
+      "grad_norm": 3.1048014163970947,
+      "learning_rate": 0.0008642745709828393,
+      "loss": 0.4104,
+      "step": 1220
+    },
+    {
+      "epoch": 0.9594383775351014,
+      "grad_norm": 1.5314342975616455,
+      "learning_rate": 0.0008631602406953422,
+      "loss": 0.3539,
+      "step": 1230
+    },
+    {
+      "epoch": 0.9672386895475819,
+      "grad_norm": 2.8501791954040527,
+      "learning_rate": 0.0008620459104078449,
+      "loss": 0.4104,
+      "step": 1240
+    },
+    {
+      "epoch": 0.9750390015600624,
+      "grad_norm": 5.708191394805908,
+      "learning_rate": 0.0008609315801203477,
+      "loss": 0.4038,
+      "step": 1250
+    },
+    {
+      "epoch": 0.982839313572543,
+      "grad_norm": 2.173867702484131,
+      "learning_rate": 0.0008598172498328505,
+      "loss": 0.338,
+      "step": 1260
+    },
+    {
+      "epoch": 0.9906396255850234,
+      "grad_norm": 2.7057418823242188,
+      "learning_rate": 0.0008587029195453533,
+      "loss": 0.3941,
+      "step": 1270
+    },
+    {
+      "epoch": 0.9984399375975039,
+      "grad_norm": 1.3492989540100098,
+      "learning_rate": 0.0008575885892578561,
+      "loss": 0.3696,
+      "step": 1280
+    },
+    {
+      "epoch": 1.0,
+      "eval_loss": 0.4044143855571747,
+      "eval_runtime": 635.8286,
+      "eval_samples_per_second": 0.865,
+      "eval_steps_per_second": 0.865,
+      "eval_wer": 26.665379416875844,
+      "step": 1282
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 8974,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 7,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 1.12949897084928e+18,
+  "train_batch_size": 3,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-1282/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:87bdde76802c6f6975c7a62016872baaa0d2d042f00a6968d4a9fadade014aa7
+size 5432

checkpoint-2564/README.md

@@ -0,0 +1,202 @@
+---
+base_model: openai/whisper-small
+library_name: peft
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.15.1

checkpoint-2564/adapter_config.json

@@ -0,0 +1,37 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": {
+    "base_model_class": "WhisperForConditionalGeneration",
+    "parent_library": "transformers.models.whisper.modeling_whisper"
+  },
+  "base_model_name_or_path": "openai/whisper-small",
+  "bias": "none",
+  "corda_config": null,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "q_proj",
+    "v_proj"
+  ],
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_rslora": false
+}

checkpoint-2564/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0f17d5065d8eda1b950862c985a0732f25bc23d36f055704bf9b1b3f6b721105
+size 14176064

checkpoint-2564/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:72752c6151e44f5621b4b0bd7ff5123dc5aaa03c691cdc294f7a04474ba89f46
+size 28432570

checkpoint-2564/preprocessor_config.json

@@ -0,0 +1,14 @@
+{
+  "chunk_length": 30,
+  "feature_extractor_type": "WhisperFeatureExtractor",
+  "feature_size": 80,
+  "hop_length": 160,
+  "n_fft": 400,
+  "n_samples": 480000,
+  "nb_max_frames": 3000,
+  "padding_side": "right",
+  "padding_value": 0.0,
+  "processor_class": "WhisperProcessor",
+  "return_attention_mask": false,
+  "sampling_rate": 16000
+}

checkpoint-2564/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:368f5ba3422484c4c03ee3552e25e2e96767e49b156f4bad0bf0eb89a7f32c1a
+size 14244

checkpoint-2564/scaler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:650a92621c02c276880616f57763587540819bc02f9bc2826dfa7c6bf9546b4a
+size 988

checkpoint-2564/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c6976c8664acd082a8fdcda4af5691a792f52438851c190626d6ba42568d3ee3
+size 1064

checkpoint-2564/trainer_state.json

@@ -0,0 +1,1843 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 2.0,
+  "eval_steps": 500,
+  "global_step": 2564,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.0078003120124804995,
+      "grad_norm": 2.144841432571411,
+      "learning_rate": 0.0009989971027412525,
+      "loss": 1.1174,
+      "step": 10
+    },
+    {
+      "epoch": 0.015600624024960999,
+      "grad_norm": 1.0625462532043457,
+      "learning_rate": 0.0009978827724537553,
+      "loss": 0.5035,
+      "step": 20
+    },
+    {
+      "epoch": 0.0234009360374415,
+      "grad_norm": 1.619573950767517,
+      "learning_rate": 0.0009967684421662581,
+      "loss": 0.3841,
+      "step": 30
+    },
+    {
+      "epoch": 0.031201248049921998,
+      "grad_norm": 3.2681403160095215,
+      "learning_rate": 0.000995654111878761,
+      "loss": 0.4974,
+      "step": 40
+    },
+    {
+      "epoch": 0.0390015600624025,
+      "grad_norm": 1.5944786071777344,
+      "learning_rate": 0.0009945397815912637,
+      "loss": 0.4797,
+      "step": 50
+    },
+    {
+      "epoch": 0.046801872074883,
+      "grad_norm": 1.8234444856643677,
+      "learning_rate": 0.0009934254513037665,
+      "loss": 0.637,
+      "step": 60
+    },
+    {
+      "epoch": 0.054602184087363496,
+      "grad_norm": 3.790844440460205,
+      "learning_rate": 0.0009923111210162691,
+      "loss": 0.4479,
+      "step": 70
+    },
+    {
+      "epoch": 0.062402496099843996,
+      "grad_norm": 2.9351046085357666,
+      "learning_rate": 0.000991196790728772,
+      "loss": 0.4332,
+      "step": 80
+    },
+    {
+      "epoch": 0.07020280811232449,
+      "grad_norm": 2.2770025730133057,
+      "learning_rate": 0.0009900824604412747,
+      "loss": 0.4399,
+      "step": 90
+    },
+    {
+      "epoch": 0.078003120124805,
+      "grad_norm": 1.86591637134552,
+      "learning_rate": 0.0009889681301537775,
+      "loss": 0.407,
+      "step": 100
+    },
+    {
+      "epoch": 0.08580343213728549,
+      "grad_norm": 1.6851640939712524,
+      "learning_rate": 0.0009878537998662805,
+      "loss": 0.3862,
+      "step": 110
+    },
+    {
+      "epoch": 0.093603744149766,
+      "grad_norm": 2.5469818115234375,
+      "learning_rate": 0.0009867394695787831,
+      "loss": 0.4166,
+      "step": 120
+    },
+    {
+      "epoch": 0.10140405616224649,
+      "grad_norm": 1.8210259675979614,
+      "learning_rate": 0.000985625139291286,
+      "loss": 0.3785,
+      "step": 130
+    },
+    {
+      "epoch": 0.10920436817472699,
+      "grad_norm": 2.031057119369507,
+      "learning_rate": 0.0009845108090037887,
+      "loss": 0.4177,
+      "step": 140
+    },
+    {
+      "epoch": 0.11700468018720749,
+      "grad_norm": 1.6646612882614136,
+      "learning_rate": 0.0009833964787162915,
+      "loss": 0.402,
+      "step": 150
+    },
+    {
+      "epoch": 0.12480499219968799,
+      "grad_norm": 1.8680285215377808,
+      "learning_rate": 0.0009822821484287943,
+      "loss": 0.3282,
+      "step": 160
+    },
+    {
+      "epoch": 0.13260530421216848,
+      "grad_norm": 1.8039604425430298,
+      "learning_rate": 0.0009811678181412971,
+      "loss": 0.3536,
+      "step": 170
+    },
+    {
+      "epoch": 0.14040561622464898,
+      "grad_norm": 3.3018901348114014,
+      "learning_rate": 0.0009800534878538,
+      "loss": 0.4595,
+      "step": 180
+    },
+    {
+      "epoch": 0.1482059282371295,
+      "grad_norm": 3.684013843536377,
+      "learning_rate": 0.0009789391575663027,
+      "loss": 0.4288,
+      "step": 190
+    },
+    {
+      "epoch": 0.15600624024961,
+      "grad_norm": 1.4512592554092407,
+      "learning_rate": 0.0009778248272788055,
+      "loss": 0.5086,
+      "step": 200
+    },
+    {
+      "epoch": 0.16380655226209048,
+      "grad_norm": 2.3981761932373047,
+      "learning_rate": 0.0009767104969913081,
+      "loss": 0.4084,
+      "step": 210
+    },
+    {
+      "epoch": 0.17160686427457097,
+      "grad_norm": 3.7943010330200195,
+      "learning_rate": 0.000975596166703811,
+      "loss": 0.4524,
+      "step": 220
+    },
+    {
+      "epoch": 0.1794071762870515,
+      "grad_norm": 2.657606840133667,
+      "learning_rate": 0.0009744818364163138,
+      "loss": 0.3592,
+      "step": 230
+    },
+    {
+      "epoch": 0.187207488299532,
+      "grad_norm": 2.7629363536834717,
+      "learning_rate": 0.0009733675061288166,
+      "loss": 0.4263,
+      "step": 240
+    },
+    {
+      "epoch": 0.19500780031201248,
+      "grad_norm": 1.3749983310699463,
+      "learning_rate": 0.0009722531758413193,
+      "loss": 0.48,
+      "step": 250
+    },
+    {
+      "epoch": 0.20280811232449297,
+      "grad_norm": 2.648716449737549,
+      "learning_rate": 0.0009711388455538221,
+      "loss": 0.416,
+      "step": 260
+    },
+    {
+      "epoch": 0.21060842433697347,
+      "grad_norm": 1.5672308206558228,
+      "learning_rate": 0.0009700245152663249,
+      "loss": 0.4223,
+      "step": 270
+    },
+    {
+      "epoch": 0.21840873634945399,
+      "grad_norm": 2.618163585662842,
+      "learning_rate": 0.0009689101849788277,
+      "loss": 0.4172,
+      "step": 280
+    },
+    {
+      "epoch": 0.22620904836193448,
+      "grad_norm": 3.6365268230438232,
+      "learning_rate": 0.0009677958546913305,
+      "loss": 0.5501,
+      "step": 290
+    },
+    {
+      "epoch": 0.23400936037441497,
+      "grad_norm": 2.740039825439453,
+      "learning_rate": 0.0009666815244038332,
+      "loss": 0.3553,
+      "step": 300
+    },
+    {
+      "epoch": 0.24180967238689546,
+      "grad_norm": 3.406210422515869,
+      "learning_rate": 0.000965567194116336,
+      "loss": 0.3518,
+      "step": 310
+    },
+    {
+      "epoch": 0.24960998439937598,
+      "grad_norm": 1.4707075357437134,
+      "learning_rate": 0.000964452863828839,
+      "loss": 0.3452,
+      "step": 320
+    },
+    {
+      "epoch": 0.2574102964118565,
+      "grad_norm": 1.608324408531189,
+      "learning_rate": 0.0009633385335413417,
+      "loss": 0.4908,
+      "step": 330
+    },
+    {
+      "epoch": 0.26521060842433697,
+      "grad_norm": 4.090480327606201,
+      "learning_rate": 0.0009622242032538444,
+      "loss": 0.4597,
+      "step": 340
+    },
+    {
+      "epoch": 0.27301092043681746,
+      "grad_norm": 2.2214395999908447,
+      "learning_rate": 0.0009611098729663472,
+      "loss": 0.4552,
+      "step": 350
+    },
+    {
+      "epoch": 0.28081123244929795,
+      "grad_norm": 1.9134166240692139,
+      "learning_rate": 0.00095999554267885,
+      "loss": 0.3571,
+      "step": 360
+    },
+    {
+      "epoch": 0.28861154446177845,
+      "grad_norm": 1.8127851486206055,
+      "learning_rate": 0.0009588812123913528,
+      "loss": 0.3808,
+      "step": 370
+    },
+    {
+      "epoch": 0.296411856474259,
+      "grad_norm": 2.2262885570526123,
+      "learning_rate": 0.0009577668821038556,
+      "loss": 0.4099,
+      "step": 380
+    },
+    {
+      "epoch": 0.3042121684867395,
+      "grad_norm": 2.8041303157806396,
+      "learning_rate": 0.0009566525518163583,
+      "loss": 0.3988,
+      "step": 390
+    },
+    {
+      "epoch": 0.31201248049922,
+      "grad_norm": 6.797432899475098,
+      "learning_rate": 0.0009555382215288611,
+      "loss": 0.4728,
+      "step": 400
+    },
+    {
+      "epoch": 0.31981279251170047,
+      "grad_norm": 3.1861369609832764,
+      "learning_rate": 0.000954423891241364,
+      "loss": 0.3502,
+      "step": 410
+    },
+    {
+      "epoch": 0.32761310452418096,
+      "grad_norm": 2.9223642349243164,
+      "learning_rate": 0.0009533095609538667,
+      "loss": 0.4215,
+      "step": 420
+    },
+    {
+      "epoch": 0.33541341653666146,
+      "grad_norm": 6.848895072937012,
+      "learning_rate": 0.0009521952306663694,
+      "loss": 0.415,
+      "step": 430
+    },
+    {
+      "epoch": 0.34321372854914195,
+      "grad_norm": 9.054282188415527,
+      "learning_rate": 0.0009510809003788722,
+      "loss": 0.4667,
+      "step": 440
+    },
+    {
+      "epoch": 0.35101404056162244,
+      "grad_norm": 2.3005900382995605,
+      "learning_rate": 0.0009499665700913752,
+      "loss": 0.6565,
+      "step": 450
+    },
+    {
+      "epoch": 0.358814352574103,
+      "grad_norm": 2.9467573165893555,
+      "learning_rate": 0.000948852239803878,
+      "loss": 0.3837,
+      "step": 460
+    },
+    {
+      "epoch": 0.3666146645865835,
+      "grad_norm": 1.5977652072906494,
+      "learning_rate": 0.0009477379095163808,
+      "loss": 0.407,
+      "step": 470
+    },
+    {
+      "epoch": 0.374414976599064,
+      "grad_norm": 2.8274600505828857,
+      "learning_rate": 0.0009466235792288835,
+      "loss": 0.4501,
+      "step": 480
+    },
+    {
+      "epoch": 0.38221528861154447,
+      "grad_norm": 1.8566502332687378,
+      "learning_rate": 0.0009455092489413863,
+      "loss": 0.3146,
+      "step": 490
+    },
+    {
+      "epoch": 0.39001560062402496,
+      "grad_norm": 2.5871951580047607,
+      "learning_rate": 0.0009443949186538891,
+      "loss": 0.3335,
+      "step": 500
+    },
+    {
+      "epoch": 0.39781591263650545,
+      "grad_norm": 2.0552711486816406,
+      "learning_rate": 0.0009432805883663919,
+      "loss": 0.3698,
+      "step": 510
+    },
+    {
+      "epoch": 0.40561622464898595,
+      "grad_norm": 1.5244548320770264,
+      "learning_rate": 0.0009421662580788946,
+      "loss": 0.581,
+      "step": 520
+    },
+    {
+      "epoch": 0.41341653666146644,
+      "grad_norm": 1.5146633386611938,
+      "learning_rate": 0.0009410519277913974,
+      "loss": 0.3634,
+      "step": 530
+    },
+    {
+      "epoch": 0.42121684867394693,
+      "grad_norm": 2.394819736480713,
+      "learning_rate": 0.0009399375975039002,
+      "loss": 0.338,
+      "step": 540
+    },
+    {
+      "epoch": 0.4290171606864275,
+      "grad_norm": 2.3500325679779053,
+      "learning_rate": 0.000938823267216403,
+      "loss": 0.4025,
+      "step": 550
+    },
+    {
+      "epoch": 0.43681747269890797,
+      "grad_norm": 2.4186370372772217,
+      "learning_rate": 0.0009377089369289058,
+      "loss": 0.4095,
+      "step": 560
+    },
+    {
+      "epoch": 0.44461778471138846,
+      "grad_norm": 1.9770065546035767,
+      "learning_rate": 0.0009365946066414085,
+      "loss": 0.4275,
+      "step": 570
+    },
+    {
+      "epoch": 0.45241809672386896,
+      "grad_norm": 1.7679632902145386,
+      "learning_rate": 0.0009354802763539114,
+      "loss": 0.3949,
+      "step": 580
+    },
+    {
+      "epoch": 0.46021840873634945,
+      "grad_norm": 2.0794620513916016,
+      "learning_rate": 0.0009343659460664142,
+      "loss": 0.4208,
+      "step": 590
+    },
+    {
+      "epoch": 0.46801872074882994,
+      "grad_norm": 2.677424192428589,
+      "learning_rate": 0.000933251615778917,
+      "loss": 0.4089,
+      "step": 600
+    },
+    {
+      "epoch": 0.47581903276131043,
+      "grad_norm": 1.526112675666809,
+      "learning_rate": 0.0009321372854914197,
+      "loss": 0.4196,
+      "step": 610
+    },
+    {
+      "epoch": 0.4836193447737909,
+      "grad_norm": 1.8656370639801025,
+      "learning_rate": 0.0009310229552039225,
+      "loss": 0.3732,
+      "step": 620
+    },
+    {
+      "epoch": 0.4914196567862715,
+      "grad_norm": 3.3338847160339355,
+      "learning_rate": 0.0009299086249164253,
+      "loss": 0.4231,
+      "step": 630
+    },
+    {
+      "epoch": 0.49921996879875197,
+      "grad_norm": 2.1057350635528564,
+      "learning_rate": 0.0009287942946289281,
+      "loss": 0.3921,
+      "step": 640
+    },
+    {
+      "epoch": 0.5070202808112324,
+      "grad_norm": 1.544977068901062,
+      "learning_rate": 0.0009276799643414309,
+      "loss": 0.3748,
+      "step": 650
+    },
+    {
+      "epoch": 0.514820592823713,
+      "grad_norm": 3.4070258140563965,
+      "learning_rate": 0.0009265656340539336,
+      "loss": 0.4027,
+      "step": 660
+    },
+    {
+      "epoch": 0.5226209048361935,
+      "grad_norm": 5.5486931800842285,
+      "learning_rate": 0.0009254513037664364,
+      "loss": 0.4326,
+      "step": 670
+    },
+    {
+      "epoch": 0.5304212168486739,
+      "grad_norm": 3.6824769973754883,
+      "learning_rate": 0.0009243369734789392,
+      "loss": 0.4809,
+      "step": 680
+    },
+    {
+      "epoch": 0.5382215288611545,
+      "grad_norm": 6.3154778480529785,
+      "learning_rate": 0.000923222643191442,
+      "loss": 0.3588,
+      "step": 690
+    },
+    {
+      "epoch": 0.5460218408736349,
+      "grad_norm": 3.133465528488159,
+      "learning_rate": 0.0009221083129039447,
+      "loss": 0.4712,
+      "step": 700
+    },
+    {
+      "epoch": 0.5538221528861155,
+      "grad_norm": 4.222598552703857,
+      "learning_rate": 0.0009209939826164475,
+      "loss": 0.4135,
+      "step": 710
+    },
+    {
+      "epoch": 0.5616224648985959,
+      "grad_norm": 4.6125078201293945,
+      "learning_rate": 0.0009198796523289504,
+      "loss": 0.393,
+      "step": 720
+    },
+    {
+      "epoch": 0.5694227769110765,
+      "grad_norm": 6.543318748474121,
+      "learning_rate": 0.0009187653220414532,
+      "loss": 0.6138,
+      "step": 730
+    },
+    {
+      "epoch": 0.5772230889235569,
+      "grad_norm": 2.596463680267334,
+      "learning_rate": 0.000917650991753956,
+      "loss": 0.3694,
+      "step": 740
+    },
+    {
+      "epoch": 0.5850234009360374,
+      "grad_norm": 2.428490161895752,
+      "learning_rate": 0.0009165366614664587,
+      "loss": 0.4535,
+      "step": 750
+    },
+    {
+      "epoch": 0.592823712948518,
+      "grad_norm": 1.8790688514709473,
+      "learning_rate": 0.0009154223311789615,
+      "loss": 0.3986,
+      "step": 760
+    },
+    {
+      "epoch": 0.6006240249609984,
+      "grad_norm": 3.141587734222412,
+      "learning_rate": 0.0009143080008914643,
+      "loss": 0.3881,
+      "step": 770
+    },
+    {
+      "epoch": 0.608424336973479,
+      "grad_norm": 2.125810146331787,
+      "learning_rate": 0.0009131936706039671,
+      "loss": 0.4225,
+      "step": 780
+    },
+    {
+      "epoch": 0.6162246489859594,
+      "grad_norm": 2.532404661178589,
+      "learning_rate": 0.0009120793403164698,
+      "loss": 0.3423,
+      "step": 790
+    },
+    {
+      "epoch": 0.62402496099844,
+      "grad_norm": 3.6324350833892822,
+      "learning_rate": 0.0009109650100289726,
+      "loss": 0.4205,
+      "step": 800
+    },
+    {
+      "epoch": 0.6318252730109204,
+      "grad_norm": 1.4804415702819824,
+      "learning_rate": 0.0009098506797414754,
+      "loss": 0.4042,
+      "step": 810
+    },
+    {
+      "epoch": 0.6396255850234009,
+      "grad_norm": 1.5140562057495117,
+      "learning_rate": 0.0009087363494539782,
+      "loss": 0.3056,
+      "step": 820
+    },
+    {
+      "epoch": 0.6474258970358814,
+      "grad_norm": 2.470576047897339,
+      "learning_rate": 0.000907622019166481,
+      "loss": 0.391,
+      "step": 830
+    },
+    {
+      "epoch": 0.6552262090483619,
+      "grad_norm": 3.4496209621429443,
+      "learning_rate": 0.0009065076888789837,
+      "loss": 0.4163,
+      "step": 840
+    },
+    {
+      "epoch": 0.6630265210608425,
+      "grad_norm": 1.8823250532150269,
+      "learning_rate": 0.0009053933585914866,
+      "loss": 0.3877,
+      "step": 850
+    },
+    {
+      "epoch": 0.6708268330733229,
+      "grad_norm": 2.492297410964966,
+      "learning_rate": 0.0009042790283039894,
+      "loss": 0.3542,
+      "step": 860
+    },
+    {
+      "epoch": 0.6786271450858035,
+      "grad_norm": 3.977569341659546,
+      "learning_rate": 0.0009031646980164922,
+      "loss": 0.4168,
+      "step": 870
+    },
+    {
+      "epoch": 0.6864274570982839,
+      "grad_norm": 3.938462495803833,
+      "learning_rate": 0.0009020503677289949,
+      "loss": 0.4894,
+      "step": 880
+    },
+    {
+      "epoch": 0.6942277691107644,
+      "grad_norm": 1.3457701206207275,
+      "learning_rate": 0.0009009360374414977,
+      "loss": 0.4903,
+      "step": 890
+    },
+    {
+      "epoch": 0.7020280811232449,
+      "grad_norm": 6.3473124504089355,
+      "learning_rate": 0.0008998217071540005,
+      "loss": 0.4766,
+      "step": 900
+    },
+    {
+      "epoch": 0.7098283931357254,
+      "grad_norm": 3.145792245864868,
+      "learning_rate": 0.0008987073768665033,
+      "loss": 0.4119,
+      "step": 910
+    },
+    {
+      "epoch": 0.717628705148206,
+      "grad_norm": 1.809446930885315,
+      "learning_rate": 0.0008975930465790061,
+      "loss": 0.3991,
+      "step": 920
+    },
+    {
+      "epoch": 0.7254290171606864,
+      "grad_norm": 1.7960044145584106,
+      "learning_rate": 0.0008964787162915088,
+      "loss": 0.3095,
+      "step": 930
+    },
+    {
+      "epoch": 0.733229329173167,
+      "grad_norm": 2.9710285663604736,
+      "learning_rate": 0.0008953643860040116,
+      "loss": 0.5104,
+      "step": 940
+    },
+    {
+      "epoch": 0.7410296411856474,
+      "grad_norm": 2.460524797439575,
+      "learning_rate": 0.0008942500557165144,
+      "loss": 0.4332,
+      "step": 950
+    },
+    {
+      "epoch": 0.748829953198128,
+      "grad_norm": 1.6166704893112183,
+      "learning_rate": 0.0008931357254290172,
+      "loss": 0.3856,
+      "step": 960
+    },
+    {
+      "epoch": 0.7566302652106084,
+      "grad_norm": 1.747750163078308,
+      "learning_rate": 0.0008920213951415199,
+      "loss": 0.3973,
+      "step": 970
+    },
+    {
+      "epoch": 0.7644305772230889,
+      "grad_norm": 1.4469414949417114,
+      "learning_rate": 0.0008909070648540227,
+      "loss": 0.3344,
+      "step": 980
+    },
+    {
+      "epoch": 0.7722308892355694,
+      "grad_norm": 4.661273956298828,
+      "learning_rate": 0.0008897927345665256,
+      "loss": 0.6003,
+      "step": 990
+    },
+    {
+      "epoch": 0.7800312012480499,
+      "grad_norm": 3.6588950157165527,
+      "learning_rate": 0.0008886784042790284,
+      "loss": 0.473,
+      "step": 1000
+    },
+    {
+      "epoch": 0.7878315132605305,
+      "grad_norm": 2.208383798599243,
+      "learning_rate": 0.0008876755070202809,
+      "loss": 0.3899,
+      "step": 1010
+    },
+    {
+      "epoch": 0.7956318252730109,
+      "grad_norm": 2.3569576740264893,
+      "learning_rate": 0.0008865611767327836,
+      "loss": 0.3871,
+      "step": 1020
+    },
+    {
+      "epoch": 0.8034321372854915,
+      "grad_norm": 2.7071454524993896,
+      "learning_rate": 0.0008854468464452864,
+      "loss": 0.3149,
+      "step": 1030
+    },
+    {
+      "epoch": 0.8112324492979719,
+      "grad_norm": 2.8024532794952393,
+      "learning_rate": 0.0008843325161577892,
+      "loss": 0.3328,
+      "step": 1040
+    },
+    {
+      "epoch": 0.8190327613104524,
+      "grad_norm": 3.0969290733337402,
+      "learning_rate": 0.000883218185870292,
+      "loss": 0.3948,
+      "step": 1050
+    },
+    {
+      "epoch": 0.8268330733229329,
+      "grad_norm": 2.982484817504883,
+      "learning_rate": 0.0008821038555827947,
+      "loss": 0.4323,
+      "step": 1060
+    },
+    {
+      "epoch": 0.8346333853354134,
+      "grad_norm": 3.133814573287964,
+      "learning_rate": 0.0008809895252952975,
+      "loss": 0.4393,
+      "step": 1070
+    },
+    {
+      "epoch": 0.8424336973478939,
+      "grad_norm": 3.3123364448547363,
+      "learning_rate": 0.0008798751950078003,
+      "loss": 0.3244,
+      "step": 1080
+    },
+    {
+      "epoch": 0.8502340093603744,
+      "grad_norm": 2.308555841445923,
+      "learning_rate": 0.0008787608647203032,
+      "loss": 0.424,
+      "step": 1090
+    },
+    {
+      "epoch": 0.858034321372855,
+      "grad_norm": 3.654137134552002,
+      "learning_rate": 0.000877646534432806,
+      "loss": 0.3445,
+      "step": 1100
+    },
+    {
+      "epoch": 0.8658346333853354,
+      "grad_norm": 2.149843692779541,
+      "learning_rate": 0.0008765322041453087,
+      "loss": 0.3398,
+      "step": 1110
+    },
+    {
+      "epoch": 0.8736349453978159,
+      "grad_norm": 3.1334431171417236,
+      "learning_rate": 0.0008754178738578115,
+      "loss": 0.3333,
+      "step": 1120
+    },
+    {
+      "epoch": 0.8814352574102964,
+      "grad_norm": 2.2942090034484863,
+      "learning_rate": 0.0008743035435703143,
+      "loss": 0.4188,
+      "step": 1130
+    },
+    {
+      "epoch": 0.8892355694227769,
+      "grad_norm": 2.0195343494415283,
+      "learning_rate": 0.0008731892132828171,
+      "loss": 0.4047,
+      "step": 1140
+    },
+    {
+      "epoch": 0.8970358814352574,
+      "grad_norm": 2.3850839138031006,
+      "learning_rate": 0.0008720748829953198,
+      "loss": 0.3931,
+      "step": 1150
+    },
+    {
+      "epoch": 0.9048361934477379,
+      "grad_norm": 1.6200228929519653,
+      "learning_rate": 0.0008709605527078226,
+      "loss": 0.409,
+      "step": 1160
+    },
+    {
+      "epoch": 0.9126365054602185,
+      "grad_norm": 2.9001989364624023,
+      "learning_rate": 0.0008698462224203254,
+      "loss": 0.4289,
+      "step": 1170
+    },
+    {
+      "epoch": 0.9204368174726989,
+      "grad_norm": 1.52889883518219,
+      "learning_rate": 0.0008687318921328282,
+      "loss": 0.3575,
+      "step": 1180
+    },
+    {
+      "epoch": 0.9282371294851794,
+      "grad_norm": 2.359733819961548,
+      "learning_rate": 0.000867617561845331,
+      "loss": 0.3837,
+      "step": 1190
+    },
+    {
+      "epoch": 0.9360374414976599,
+      "grad_norm": 2.3807597160339355,
+      "learning_rate": 0.0008665032315578337,
+      "loss": 0.4206,
+      "step": 1200
+    },
+    {
+      "epoch": 0.9438377535101404,
+      "grad_norm": 1.8366179466247559,
+      "learning_rate": 0.0008653889012703365,
+      "loss": 0.3101,
+      "step": 1210
+    },
+    {
+      "epoch": 0.9516380655226209,
+      "grad_norm": 3.1048014163970947,
+      "learning_rate": 0.0008642745709828393,
+      "loss": 0.4104,
+      "step": 1220
+    },
+    {
+      "epoch": 0.9594383775351014,
+      "grad_norm": 1.5314342975616455,
+      "learning_rate": 0.0008631602406953422,
+      "loss": 0.3539,
+      "step": 1230
+    },
+    {
+      "epoch": 0.9672386895475819,
+      "grad_norm": 2.8501791954040527,
+      "learning_rate": 0.0008620459104078449,
+      "loss": 0.4104,
+      "step": 1240
+    },
+    {
+      "epoch": 0.9750390015600624,
+      "grad_norm": 5.708191394805908,
+      "learning_rate": 0.0008609315801203477,
+      "loss": 0.4038,
+      "step": 1250
+    },
+    {
+      "epoch": 0.982839313572543,
+      "grad_norm": 2.173867702484131,
+      "learning_rate": 0.0008598172498328505,
+      "loss": 0.338,
+      "step": 1260
+    },
+    {
+      "epoch": 0.9906396255850234,
+      "grad_norm": 2.7057418823242188,
+      "learning_rate": 0.0008587029195453533,
+      "loss": 0.3941,
+      "step": 1270
+    },
+    {
+      "epoch": 0.9984399375975039,
+      "grad_norm": 1.3492989540100098,
+      "learning_rate": 0.0008575885892578561,
+      "loss": 0.3696,
+      "step": 1280
+    },
+    {
+      "epoch": 1.0,
+      "eval_loss": 0.4044143855571747,
+      "eval_runtime": 635.8286,
+      "eval_samples_per_second": 0.865,
+      "eval_steps_per_second": 0.865,
+      "eval_wer": 26.665379416875844,
+      "step": 1282
+    },
+    {
+      "epoch": 1.0062402496099845,
+      "grad_norm": 2.5727291107177734,
+      "learning_rate": 0.0008564742589703588,
+      "loss": 0.3369,
+      "step": 1290
+    },
+    {
+      "epoch": 1.0140405616224648,
+      "grad_norm": 2.681490182876587,
+      "learning_rate": 0.0008553599286828616,
+      "loss": 0.2434,
+      "step": 1300
+    },
+    {
+      "epoch": 1.0218408736349454,
+      "grad_norm": 2.0296504497528076,
+      "learning_rate": 0.0008542455983953644,
+      "loss": 0.2122,
+      "step": 1310
+    },
+    {
+      "epoch": 1.029641185647426,
+      "grad_norm": 2.277512311935425,
+      "learning_rate": 0.0008531312681078672,
+      "loss": 0.3932,
+      "step": 1320
+    },
+    {
+      "epoch": 1.0374414976599065,
+      "grad_norm": 4.77215576171875,
+      "learning_rate": 0.0008520169378203699,
+      "loss": 0.2658,
+      "step": 1330
+    },
+    {
+      "epoch": 1.045241809672387,
+      "grad_norm": 1.4027091264724731,
+      "learning_rate": 0.0008509026075328727,
+      "loss": 0.3524,
+      "step": 1340
+    },
+    {
+      "epoch": 1.0530421216848673,
+      "grad_norm": 2.2849514484405518,
+      "learning_rate": 0.0008497882772453755,
+      "loss": 0.2705,
+      "step": 1350
+    },
+    {
+      "epoch": 1.0608424336973479,
+      "grad_norm": 1.2896777391433716,
+      "learning_rate": 0.0008486739469578784,
+      "loss": 0.264,
+      "step": 1360
+    },
+    {
+      "epoch": 1.0686427457098284,
+      "grad_norm": 2.4552128314971924,
+      "learning_rate": 0.0008475596166703812,
+      "loss": 0.2833,
+      "step": 1370
+    },
+    {
+      "epoch": 1.076443057722309,
+      "grad_norm": 2.044693946838379,
+      "learning_rate": 0.0008464452863828839,
+      "loss": 0.2116,
+      "step": 1380
+    },
+    {
+      "epoch": 1.0842433697347893,
+      "grad_norm": 1.3727463483810425,
+      "learning_rate": 0.0008453309560953867,
+      "loss": 0.2533,
+      "step": 1390
+    },
+    {
+      "epoch": 1.0920436817472698,
+      "grad_norm": 1.6917822360992432,
+      "learning_rate": 0.0008442166258078895,
+      "loss": 0.4259,
+      "step": 1400
+    },
+    {
+      "epoch": 1.0998439937597504,
+      "grad_norm": 2.198549747467041,
+      "learning_rate": 0.0008431022955203923,
+      "loss": 0.3161,
+      "step": 1410
+    },
+    {
+      "epoch": 1.107644305772231,
+      "grad_norm": 1.7467869520187378,
+      "learning_rate": 0.000841987965232895,
+      "loss": 0.2377,
+      "step": 1420
+    },
+    {
+      "epoch": 1.1154446177847115,
+      "grad_norm": 2.5347695350646973,
+      "learning_rate": 0.0008408736349453978,
+      "loss": 0.3258,
+      "step": 1430
+    },
+    {
+      "epoch": 1.1232449297971918,
+      "grad_norm": 1.9081774950027466,
+      "learning_rate": 0.0008397593046579006,
+      "loss": 0.2462,
+      "step": 1440
+    },
+    {
+      "epoch": 1.1310452418096724,
+      "grad_norm": 1.5889848470687866,
+      "learning_rate": 0.0008386449743704034,
+      "loss": 0.2404,
+      "step": 1450
+    },
+    {
+      "epoch": 1.138845553822153,
+      "grad_norm": 1.8944768905639648,
+      "learning_rate": 0.0008375306440829062,
+      "loss": 0.2707,
+      "step": 1460
+    },
+    {
+      "epoch": 1.1466458658346335,
+      "grad_norm": 2.5448453426361084,
+      "learning_rate": 0.0008364163137954089,
+      "loss": 0.3342,
+      "step": 1470
+    },
+    {
+      "epoch": 1.154446177847114,
+      "grad_norm": 2.0936005115509033,
+      "learning_rate": 0.0008353019835079117,
+      "loss": 0.385,
+      "step": 1480
+    },
+    {
+      "epoch": 1.1622464898595943,
+      "grad_norm": 2.614129066467285,
+      "learning_rate": 0.0008341876532204145,
+      "loss": 0.2817,
+      "step": 1490
+    },
+    {
+      "epoch": 1.1700468018720749,
+      "grad_norm": 1.6156001091003418,
+      "learning_rate": 0.0008330733229329174,
+      "loss": 0.3527,
+      "step": 1500
+    },
+    {
+      "epoch": 1.1778471138845554,
+      "grad_norm": 1.4294220209121704,
+      "learning_rate": 0.0008319589926454201,
+      "loss": 0.2469,
+      "step": 1510
+    },
+    {
+      "epoch": 1.185647425897036,
+      "grad_norm": 3.197176456451416,
+      "learning_rate": 0.0008308446623579229,
+      "loss": 0.2753,
+      "step": 1520
+    },
+    {
+      "epoch": 1.1934477379095163,
+      "grad_norm": 2.1629223823547363,
+      "learning_rate": 0.0008297303320704257,
+      "loss": 0.3951,
+      "step": 1530
+    },
+    {
+      "epoch": 1.2012480499219969,
+      "grad_norm": 2.9824419021606445,
+      "learning_rate": 0.0008286160017829285,
+      "loss": 0.3278,
+      "step": 1540
+    },
+    {
+      "epoch": 1.2090483619344774,
+      "grad_norm": 2.866138219833374,
+      "learning_rate": 0.0008275016714954313,
+      "loss": 0.4267,
+      "step": 1550
+    },
+    {
+      "epoch": 1.216848673946958,
+      "grad_norm": 2.36781644821167,
+      "learning_rate": 0.000826387341207934,
+      "loss": 0.3414,
+      "step": 1560
+    },
+    {
+      "epoch": 1.2246489859594383,
+      "grad_norm": 1.8305447101593018,
+      "learning_rate": 0.0008252730109204368,
+      "loss": 0.2925,
+      "step": 1570
+    },
+    {
+      "epoch": 1.2324492979719188,
+      "grad_norm": 1.9879776239395142,
+      "learning_rate": 0.0008241586806329396,
+      "loss": 0.357,
+      "step": 1580
+    },
+    {
+      "epoch": 1.2402496099843994,
+      "grad_norm": 2.183350086212158,
+      "learning_rate": 0.0008230443503454424,
+      "loss": 0.3409,
+      "step": 1590
+    },
+    {
+      "epoch": 1.24804992199688,
+      "grad_norm": 2.197072744369507,
+      "learning_rate": 0.0008219300200579451,
+      "loss": 0.33,
+      "step": 1600
+    },
+    {
+      "epoch": 1.2558502340093605,
+      "grad_norm": 3.2065696716308594,
+      "learning_rate": 0.0008208156897704479,
+      "loss": 0.2853,
+      "step": 1610
+    },
+    {
+      "epoch": 1.2636505460218408,
+      "grad_norm": 2.0581350326538086,
+      "learning_rate": 0.0008197013594829507,
+      "loss": 0.3647,
+      "step": 1620
+    },
+    {
+      "epoch": 1.2714508580343213,
+      "grad_norm": 3.149153232574463,
+      "learning_rate": 0.0008185870291954536,
+      "loss": 0.3921,
+      "step": 1630
+    },
+    {
+      "epoch": 1.2792511700468019,
+      "grad_norm": 2.5097105503082275,
+      "learning_rate": 0.0008174726989079563,
+      "loss": 0.3302,
+      "step": 1640
+    },
+    {
+      "epoch": 1.2870514820592824,
+      "grad_norm": 2.7537474632263184,
+      "learning_rate": 0.0008163583686204591,
+      "loss": 0.3286,
+      "step": 1650
+    },
+    {
+      "epoch": 1.294851794071763,
+      "grad_norm": 1.966965675354004,
+      "learning_rate": 0.0008152440383329619,
+      "loss": 0.2785,
+      "step": 1660
+    },
+    {
+      "epoch": 1.3026521060842433,
+      "grad_norm": 1.9159988164901733,
+      "learning_rate": 0.0008141297080454647,
+      "loss": 0.3299,
+      "step": 1670
+    },
+    {
+      "epoch": 1.3104524180967239,
+      "grad_norm": 2.2212252616882324,
+      "learning_rate": 0.0008130153777579675,
+      "loss": 0.2807,
+      "step": 1680
+    },
+    {
+      "epoch": 1.3182527301092044,
+      "grad_norm": 4.194318771362305,
+      "learning_rate": 0.0008119010474704702,
+      "loss": 0.3161,
+      "step": 1690
+    },
+    {
+      "epoch": 1.3260530421216847,
+      "grad_norm": 1.7189604043960571,
+      "learning_rate": 0.000810786717182973,
+      "loss": 0.3356,
+      "step": 1700
+    },
+    {
+      "epoch": 1.3338533541341655,
+      "grad_norm": 1.5196418762207031,
+      "learning_rate": 0.0008096723868954758,
+      "loss": 0.2616,
+      "step": 1710
+    },
+    {
+      "epoch": 1.3416536661466458,
+      "grad_norm": 1.497450351715088,
+      "learning_rate": 0.0008085580566079786,
+      "loss": 0.2946,
+      "step": 1720
+    },
+    {
+      "epoch": 1.3494539781591264,
+      "grad_norm": 1.74885892868042,
+      "learning_rate": 0.0008074437263204813,
+      "loss": 0.2801,
+      "step": 1730
+    },
+    {
+      "epoch": 1.357254290171607,
+      "grad_norm": 2.040701389312744,
+      "learning_rate": 0.0008063293960329841,
+      "loss": 0.3203,
+      "step": 1740
+    },
+    {
+      "epoch": 1.3650546021840873,
+      "grad_norm": 3.760457754135132,
+      "learning_rate": 0.0008052150657454869,
+      "loss": 0.4569,
+      "step": 1750
+    },
+    {
+      "epoch": 1.3728549141965678,
+      "grad_norm": 2.92971134185791,
+      "learning_rate": 0.0008041007354579897,
+      "loss": 0.3321,
+      "step": 1760
+    },
+    {
+      "epoch": 1.3806552262090483,
+      "grad_norm": 1.9461047649383545,
+      "learning_rate": 0.0008029864051704926,
+      "loss": 0.2696,
+      "step": 1770
+    },
+    {
+      "epoch": 1.388455538221529,
+      "grad_norm": 3.2626147270202637,
+      "learning_rate": 0.0008018720748829953,
+      "loss": 0.3322,
+      "step": 1780
+    },
+    {
+      "epoch": 1.3962558502340094,
+      "grad_norm": 2.1270642280578613,
+      "learning_rate": 0.0008007577445954981,
+      "loss": 0.2965,
+      "step": 1790
+    },
+    {
+      "epoch": 1.4040561622464898,
+      "grad_norm": 2.3174221515655518,
+      "learning_rate": 0.0007996434143080009,
+      "loss": 0.3425,
+      "step": 1800
+    },
+    {
+      "epoch": 1.4118564742589703,
+      "grad_norm": 2.5749576091766357,
+      "learning_rate": 0.0007985290840205037,
+      "loss": 0.3622,
+      "step": 1810
+    },
+    {
+      "epoch": 1.4196567862714509,
+      "grad_norm": 1.873813509941101,
+      "learning_rate": 0.0007974147537330064,
+      "loss": 0.2497,
+      "step": 1820
+    },
+    {
+      "epoch": 1.4274570982839314,
+      "grad_norm": 3.633928060531616,
+      "learning_rate": 0.0007963004234455092,
+      "loss": 0.4058,
+      "step": 1830
+    },
+    {
+      "epoch": 1.435257410296412,
+      "grad_norm": 2.356269598007202,
+      "learning_rate": 0.000795186093158012,
+      "loss": 0.2635,
+      "step": 1840
+    },
+    {
+      "epoch": 1.4430577223088923,
+      "grad_norm": 1.9108752012252808,
+      "learning_rate": 0.0007940717628705148,
+      "loss": 0.336,
+      "step": 1850
+    },
+    {
+      "epoch": 1.4508580343213728,
+      "grad_norm": 1.5505330562591553,
+      "learning_rate": 0.0007929574325830176,
+      "loss": 0.3493,
+      "step": 1860
+    },
+    {
+      "epoch": 1.4586583463338534,
+      "grad_norm": 1.9970422983169556,
+      "learning_rate": 0.0007918431022955203,
+      "loss": 0.273,
+      "step": 1870
+    },
+    {
+      "epoch": 1.466458658346334,
+      "grad_norm": 2.753758192062378,
+      "learning_rate": 0.0007907287720080231,
+      "loss": 0.2845,
+      "step": 1880
+    },
+    {
+      "epoch": 1.4742589703588145,
+      "grad_norm": NaN,
+      "learning_rate": 0.0007897258747492757,
+      "loss": 0.2544,
+      "step": 1890
+    },
+    {
+      "epoch": 1.4820592823712948,
+      "grad_norm": 3.0995099544525146,
+      "learning_rate": 0.0007886115444617785,
+      "loss": 0.2884,
+      "step": 1900
+    },
+    {
+      "epoch": 1.4898595943837754,
+      "grad_norm": 5.728559970855713,
+      "learning_rate": 0.0007874972141742812,
+      "loss": 0.2681,
+      "step": 1910
+    },
+    {
+      "epoch": 1.497659906396256,
+      "grad_norm": 1.492622971534729,
+      "learning_rate": 0.000786382883886784,
+      "loss": 0.2891,
+      "step": 1920
+    },
+    {
+      "epoch": 1.5054602184087362,
+      "grad_norm": 1.7419252395629883,
+      "learning_rate": 0.0007852685535992868,
+      "loss": 0.4089,
+      "step": 1930
+    },
+    {
+      "epoch": 1.513260530421217,
+      "grad_norm": 6.814690589904785,
+      "learning_rate": 0.0007841542233117896,
+      "loss": 0.3372,
+      "step": 1940
+    },
+    {
+      "epoch": 1.5210608424336973,
+      "grad_norm": 2.380725860595703,
+      "learning_rate": 0.0007830398930242924,
+      "loss": 0.3189,
+      "step": 1950
+    },
+    {
+      "epoch": 1.5288611544461779,
+      "grad_norm": 5.004116058349609,
+      "learning_rate": 0.0007819255627367951,
+      "loss": 0.3018,
+      "step": 1960
+    },
+    {
+      "epoch": 1.5366614664586584,
+      "grad_norm": 2.604365825653076,
+      "learning_rate": 0.0007808112324492979,
+      "loss": 0.3054,
+      "step": 1970
+    },
+    {
+      "epoch": 1.5444617784711387,
+      "grad_norm": 1.585584044456482,
+      "learning_rate": 0.0007796969021618007,
+      "loss": 0.3477,
+      "step": 1980
+    },
+    {
+      "epoch": 1.5522620904836193,
+      "grad_norm": 1.8678693771362305,
+      "learning_rate": 0.0007785825718743035,
+      "loss": 0.3577,
+      "step": 1990
+    },
+    {
+      "epoch": 1.5600624024960998,
+      "grad_norm": 1.654689073562622,
+      "learning_rate": 0.0007774682415868062,
+      "loss": 0.2625,
+      "step": 2000
+    },
+    {
+      "epoch": 1.5678627145085804,
+      "grad_norm": 2.108919858932495,
+      "learning_rate": 0.0007763539112993092,
+      "loss": 0.2497,
+      "step": 2010
+    },
+    {
+      "epoch": 1.575663026521061,
+      "grad_norm": 7.198604106903076,
+      "learning_rate": 0.000775239581011812,
+      "loss": 0.3382,
+      "step": 2020
+    },
+    {
+      "epoch": 1.5834633385335413,
+      "grad_norm": 2.2285892963409424,
+      "learning_rate": 0.0007741252507243148,
+      "loss": 0.2598,
+      "step": 2030
+    },
+    {
+      "epoch": 1.5912636505460218,
+      "grad_norm": 1.7743014097213745,
+      "learning_rate": 0.0007730109204368176,
+      "loss": 0.2757,
+      "step": 2040
+    },
+    {
+      "epoch": 1.5990639625585024,
+      "grad_norm": 1.7763789892196655,
+      "learning_rate": 0.0007718965901493203,
+      "loss": 0.2703,
+      "step": 2050
+    },
+    {
+      "epoch": 1.6068642745709827,
+      "grad_norm": 2.159956693649292,
+      "learning_rate": 0.000770782259861823,
+      "loss": 0.2824,
+      "step": 2060
+    },
+    {
+      "epoch": 1.6146645865834635,
+      "grad_norm": 1.4845560789108276,
+      "learning_rate": 0.0007696679295743259,
+      "loss": 0.2528,
+      "step": 2070
+    },
+    {
+      "epoch": 1.6224648985959438,
+      "grad_norm": 3.627887010574341,
+      "learning_rate": 0.0007685535992868287,
+      "loss": 0.3197,
+      "step": 2080
+    },
+    {
+      "epoch": 1.6302652106084243,
+      "grad_norm": 2.2174973487854004,
+      "learning_rate": 0.0007674392689993314,
+      "loss": 0.2994,
+      "step": 2090
+    },
+    {
+      "epoch": 1.6380655226209049,
+      "grad_norm": 2.5977325439453125,
+      "learning_rate": 0.0007663249387118342,
+      "loss": 0.2991,
+      "step": 2100
+    },
+    {
+      "epoch": 1.6458658346333852,
+      "grad_norm": 1.9066824913024902,
+      "learning_rate": 0.000765210608424337,
+      "loss": 0.2166,
+      "step": 2110
+    },
+    {
+      "epoch": 1.653666146645866,
+      "grad_norm": 1.7197297811508179,
+      "learning_rate": 0.0007640962781368398,
+      "loss": 0.2948,
+      "step": 2120
+    },
+    {
+      "epoch": 1.6614664586583463,
+      "grad_norm": 2.054304361343384,
+      "learning_rate": 0.0007629819478493426,
+      "loss": 0.3073,
+      "step": 2130
+    },
+    {
+      "epoch": 1.6692667706708268,
+      "grad_norm": 1.7934963703155518,
+      "learning_rate": 0.0007618676175618453,
+      "loss": 0.2667,
+      "step": 2140
+    },
+    {
+      "epoch": 1.6770670826833074,
+      "grad_norm": 2.5259838104248047,
+      "learning_rate": 0.0007607532872743482,
+      "loss": 0.3322,
+      "step": 2150
+    },
+    {
+      "epoch": 1.6848673946957877,
+      "grad_norm": 3.6354122161865234,
+      "learning_rate": 0.000759638956986851,
+      "loss": 0.3909,
+      "step": 2160
+    },
+    {
+      "epoch": 1.6926677067082685,
+      "grad_norm": 1.6722809076309204,
+      "learning_rate": 0.0007585246266993538,
+      "loss": 0.2987,
+      "step": 2170
+    },
+    {
+      "epoch": 1.7004680187207488,
+      "grad_norm": 4.3235015869140625,
+      "learning_rate": 0.0007574102964118565,
+      "loss": 0.3112,
+      "step": 2180
+    },
+    {
+      "epoch": 1.7082683307332294,
+      "grad_norm": 2.236316442489624,
+      "learning_rate": 0.0007562959661243593,
+      "loss": 0.3174,
+      "step": 2190
+    },
+    {
+      "epoch": 1.71606864274571,
+      "grad_norm": 21.32891273498535,
+      "learning_rate": 0.0007551816358368621,
+      "loss": 0.3696,
+      "step": 2200
+    },
+    {
+      "epoch": 1.7238689547581902,
+      "grad_norm": 2.4251410961151123,
+      "learning_rate": 0.0007540673055493649,
+      "loss": 0.3171,
+      "step": 2210
+    },
+    {
+      "epoch": 1.7316692667706708,
+      "grad_norm": 2.4152424335479736,
+      "learning_rate": 0.0007529529752618677,
+      "loss": 0.328,
+      "step": 2220
+    },
+    {
+      "epoch": 1.7394695787831513,
+      "grad_norm": 2.0988574028015137,
+      "learning_rate": 0.0007518386449743704,
+      "loss": 0.3061,
+      "step": 2230
+    },
+    {
+      "epoch": 1.7472698907956317,
+      "grad_norm": 4.469291687011719,
+      "learning_rate": 0.0007507243146868732,
+      "loss": 0.3315,
+      "step": 2240
+    },
+    {
+      "epoch": 1.7550702028081124,
+      "grad_norm": 2.4917778968811035,
+      "learning_rate": 0.000749609984399376,
+      "loss": 0.4296,
+      "step": 2250
+    },
+    {
+      "epoch": 1.7628705148205928,
+      "grad_norm": 3.073840379714966,
+      "learning_rate": 0.0007484956541118788,
+      "loss": 0.3685,
+      "step": 2260
+    },
+    {
+      "epoch": 1.7706708268330733,
+      "grad_norm": 2.205733299255371,
+      "learning_rate": 0.0007473813238243815,
+      "loss": 0.2691,
+      "step": 2270
+    },
+    {
+      "epoch": 1.7784711388455539,
+      "grad_norm": 2.3948941230773926,
+      "learning_rate": 0.0007462669935368844,
+      "loss": 0.3065,
+      "step": 2280
+    },
+    {
+      "epoch": 1.7862714508580342,
+      "grad_norm": 2.6060824394226074,
+      "learning_rate": 0.0007451526632493872,
+      "loss": 0.3354,
+      "step": 2290
+    },
+    {
+      "epoch": 1.794071762870515,
+      "grad_norm": 3.2586774826049805,
+      "learning_rate": 0.00074403833296189,
+      "loss": 0.261,
+      "step": 2300
+    },
+    {
+      "epoch": 1.8018720748829953,
+      "grad_norm": 1.6417285203933716,
+      "learning_rate": 0.0007429240026743928,
+      "loss": 0.3828,
+      "step": 2310
+    },
+    {
+      "epoch": 1.8096723868954758,
+      "grad_norm": 4.006927967071533,
+      "learning_rate": 0.0007418096723868955,
+      "loss": 0.4228,
+      "step": 2320
+    },
+    {
+      "epoch": 1.8174726989079564,
+      "grad_norm": 2.5880374908447266,
+      "learning_rate": 0.0007406953420993983,
+      "loss": 0.2766,
+      "step": 2330
+    },
+    {
+      "epoch": 1.8252730109204367,
+      "grad_norm": 1.602337121963501,
+      "learning_rate": 0.0007395810118119011,
+      "loss": 0.2794,
+      "step": 2340
+    },
+    {
+      "epoch": 1.8330733229329175,
+      "grad_norm": 5.932153224945068,
+      "learning_rate": 0.0007384666815244039,
+      "loss": 0.3471,
+      "step": 2350
+    },
+    {
+      "epoch": 1.8408736349453978,
+      "grad_norm": 4.076808452606201,
+      "learning_rate": 0.0007373523512369066,
+      "loss": 0.2649,
+      "step": 2360
+    },
+    {
+      "epoch": 1.8486739469578783,
+      "grad_norm": 4.666397571563721,
+      "learning_rate": 0.0007362380209494094,
+      "loss": 0.2751,
+      "step": 2370
+    },
+    {
+      "epoch": 1.856474258970359,
+      "grad_norm": 3.792745590209961,
+      "learning_rate": 0.0007351236906619122,
+      "loss": 0.2606,
+      "step": 2380
+    },
+    {
+      "epoch": 1.8642745709828392,
+      "grad_norm": 2.5275423526763916,
+      "learning_rate": 0.000734009360374415,
+      "loss": 0.3469,
+      "step": 2390
+    },
+    {
+      "epoch": 1.8720748829953198,
+      "grad_norm": 1.59649658203125,
+      "learning_rate": 0.0007328950300869178,
+      "loss": 0.2724,
+      "step": 2400
+    },
+    {
+      "epoch": 1.8798751950078003,
+      "grad_norm": 3.7428267002105713,
+      "learning_rate": 0.0007317806997994206,
+      "loss": 0.4426,
+      "step": 2410
+    },
+    {
+      "epoch": 1.8876755070202809,
+      "grad_norm": 3.7439956665039062,
+      "learning_rate": 0.0007306663695119234,
+      "loss": 0.2901,
+      "step": 2420
+    },
+    {
+      "epoch": 1.8954758190327614,
+      "grad_norm": 2.3777198791503906,
+      "learning_rate": 0.0007295520392244262,
+      "loss": 0.2837,
+      "step": 2430
+    },
+    {
+      "epoch": 1.9032761310452417,
+      "grad_norm": 2.7654988765716553,
+      "learning_rate": 0.000728437708936929,
+      "loss": 0.277,
+      "step": 2440
+    },
+    {
+      "epoch": 1.9110764430577223,
+      "grad_norm": 1.8758680820465088,
+      "learning_rate": 0.0007273233786494317,
+      "loss": 0.2706,
+      "step": 2450
+    },
+    {
+      "epoch": 1.9188767550702028,
+      "grad_norm": 2.8725340366363525,
+      "learning_rate": 0.0007262090483619345,
+      "loss": 0.2566,
+      "step": 2460
+    },
+    {
+      "epoch": 1.9266770670826832,
+      "grad_norm": 2.4021458625793457,
+      "learning_rate": 0.0007250947180744373,
+      "loss": 0.2645,
+      "step": 2470
+    },
+    {
+      "epoch": 1.934477379095164,
+      "grad_norm": 2.8407838344573975,
+      "learning_rate": 0.0007239803877869401,
+      "loss": 0.3152,
+      "step": 2480
+    },
+    {
+      "epoch": 1.9422776911076443,
+      "grad_norm": 3.606403112411499,
+      "learning_rate": 0.0007228660574994429,
+      "loss": 0.373,
+      "step": 2490
+    },
+    {
+      "epoch": 1.9500780031201248,
+      "grad_norm": 2.362473487854004,
+      "learning_rate": 0.0007217517272119456,
+      "loss": 0.4332,
+      "step": 2500
+    },
+    {
+      "epoch": 1.9578783151326054,
+      "grad_norm": 1.9711815118789673,
+      "learning_rate": 0.0007206373969244484,
+      "loss": 0.2606,
+      "step": 2510
+    },
+    {
+      "epoch": 1.9656786271450857,
+      "grad_norm": 2.683908224105835,
+      "learning_rate": 0.0007195230666369512,
+      "loss": 0.3582,
+      "step": 2520
+    },
+    {
+      "epoch": 1.9734789391575664,
+      "grad_norm": 2.5902493000030518,
+      "learning_rate": 0.000718408736349454,
+      "loss": 0.2694,
+      "step": 2530
+    },
+    {
+      "epoch": 1.9812792511700468,
+      "grad_norm": 3.92708420753479,
+      "learning_rate": 0.0007172944060619567,
+      "loss": 0.3083,
+      "step": 2540
+    },
+    {
+      "epoch": 1.9890795631825273,
+      "grad_norm": 2.6788370609283447,
+      "learning_rate": 0.0007161800757744596,
+      "loss": 0.2587,
+      "step": 2550
+    },
+    {
+      "epoch": 1.9968798751950079,
+      "grad_norm": 2.413313627243042,
+      "learning_rate": 0.0007150657454869624,
+      "loss": 0.2732,
+      "step": 2560
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 0.3617618680000305,
+      "eval_runtime": 36952.264,
+      "eval_samples_per_second": 0.015,
+      "eval_steps_per_second": 0.015,
+      "eval_wer": 22.793975670978952,
+      "step": 2564
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 8974,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 7,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 2.25899794169856e+18,
+  "train_batch_size": 3,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-2564/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:87bdde76802c6f6975c7a62016872baaa0d2d042f00a6968d4a9fadade014aa7
+size 5432

checkpoint-3846/README.md

@@ -0,0 +1,202 @@
+---
+base_model: openai/whisper-small
+library_name: peft
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.15.1

checkpoint-3846/adapter_config.json

@@ -0,0 +1,37 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": {
+    "base_model_class": "WhisperForConditionalGeneration",
+    "parent_library": "transformers.models.whisper.modeling_whisper"
+  },
+  "base_model_name_or_path": "openai/whisper-small",
+  "bias": "none",
+  "corda_config": null,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "q_proj",
+    "v_proj"
+  ],
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_rslora": false
+}

checkpoint-3846/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e76bd5cfed787c516ed6fce8d2d8c0ee2787834afc04871ba3e410ef9d5b0900
+size 14176064

checkpoint-3846/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:32f14dd0adfb3e40faa6cc63c3eaa286860e61936a189971626c3692302a21f1
+size 28432570

checkpoint-3846/preprocessor_config.json

@@ -0,0 +1,14 @@
+{
+  "chunk_length": 30,
+  "feature_extractor_type": "WhisperFeatureExtractor",
+  "feature_size": 80,
+  "hop_length": 160,
+  "n_fft": 400,
+  "n_samples": 480000,
+  "nb_max_frames": 3000,
+  "padding_side": "right",
+  "padding_value": 0.0,
+  "processor_class": "WhisperProcessor",
+  "return_attention_mask": false,
+  "sampling_rate": 16000
+}

checkpoint-3846/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6162a8a7b581c951d3b42caa27aa201d009d04a6cefc177ec4e39f366827eb25
+size 14244

checkpoint-3846/scaler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:46036d1e75960fd9e119e16473789f6ee322f2785c0f0da785f220b1a9e20f62
+size 988

checkpoint-3846/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:24acd7ee7f5c127e7cc14245ddd02a2a7e3440506b3b7a0e7e9f976f92870102
+size 1064

checkpoint-3846/trainer_state.json

@@ -0,0 +1,2748 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 3.0,
+  "eval_steps": 500,
+  "global_step": 3846,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.0078003120124804995,
+      "grad_norm": 2.144841432571411,
+      "learning_rate": 0.0009989971027412525,
+      "loss": 1.1174,
+      "step": 10
+    },
+    {
+      "epoch": 0.015600624024960999,
+      "grad_norm": 1.0625462532043457,
+      "learning_rate": 0.0009978827724537553,
+      "loss": 0.5035,
+      "step": 20
+    },
+    {
+      "epoch": 0.0234009360374415,
+      "grad_norm": 1.619573950767517,
+      "learning_rate": 0.0009967684421662581,
+      "loss": 0.3841,
+      "step": 30
+    },
+    {
+      "epoch": 0.031201248049921998,
+      "grad_norm": 3.2681403160095215,
+      "learning_rate": 0.000995654111878761,
+      "loss": 0.4974,
+      "step": 40
+    },
+    {
+      "epoch": 0.0390015600624025,
+      "grad_norm": 1.5944786071777344,
+      "learning_rate": 0.0009945397815912637,
+      "loss": 0.4797,
+      "step": 50
+    },
+    {
+      "epoch": 0.046801872074883,
+      "grad_norm": 1.8234444856643677,
+      "learning_rate": 0.0009934254513037665,
+      "loss": 0.637,
+      "step": 60
+    },
+    {
+      "epoch": 0.054602184087363496,
+      "grad_norm": 3.790844440460205,
+      "learning_rate": 0.0009923111210162691,
+      "loss": 0.4479,
+      "step": 70
+    },
+    {
+      "epoch": 0.062402496099843996,
+      "grad_norm": 2.9351046085357666,
+      "learning_rate": 0.000991196790728772,
+      "loss": 0.4332,
+      "step": 80
+    },
+    {
+      "epoch": 0.07020280811232449,
+      "grad_norm": 2.2770025730133057,
+      "learning_rate": 0.0009900824604412747,
+      "loss": 0.4399,
+      "step": 90
+    },
+    {
+      "epoch": 0.078003120124805,
+      "grad_norm": 1.86591637134552,
+      "learning_rate": 0.0009889681301537775,
+      "loss": 0.407,
+      "step": 100
+    },
+    {
+      "epoch": 0.08580343213728549,
+      "grad_norm": 1.6851640939712524,
+      "learning_rate": 0.0009878537998662805,
+      "loss": 0.3862,
+      "step": 110
+    },
+    {
+      "epoch": 0.093603744149766,
+      "grad_norm": 2.5469818115234375,
+      "learning_rate": 0.0009867394695787831,
+      "loss": 0.4166,
+      "step": 120
+    },
+    {
+      "epoch": 0.10140405616224649,
+      "grad_norm": 1.8210259675979614,
+      "learning_rate": 0.000985625139291286,
+      "loss": 0.3785,
+      "step": 130
+    },
+    {
+      "epoch": 0.10920436817472699,
+      "grad_norm": 2.031057119369507,
+      "learning_rate": 0.0009845108090037887,
+      "loss": 0.4177,
+      "step": 140
+    },
+    {
+      "epoch": 0.11700468018720749,
+      "grad_norm": 1.6646612882614136,
+      "learning_rate": 0.0009833964787162915,
+      "loss": 0.402,
+      "step": 150
+    },
+    {
+      "epoch": 0.12480499219968799,
+      "grad_norm": 1.8680285215377808,
+      "learning_rate": 0.0009822821484287943,
+      "loss": 0.3282,
+      "step": 160
+    },
+    {
+      "epoch": 0.13260530421216848,
+      "grad_norm": 1.8039604425430298,
+      "learning_rate": 0.0009811678181412971,
+      "loss": 0.3536,
+      "step": 170
+    },
+    {
+      "epoch": 0.14040561622464898,
+      "grad_norm": 3.3018901348114014,
+      "learning_rate": 0.0009800534878538,
+      "loss": 0.4595,
+      "step": 180
+    },
+    {
+      "epoch": 0.1482059282371295,
+      "grad_norm": 3.684013843536377,
+      "learning_rate": 0.0009789391575663027,
+      "loss": 0.4288,
+      "step": 190
+    },
+    {
+      "epoch": 0.15600624024961,
+      "grad_norm": 1.4512592554092407,
+      "learning_rate": 0.0009778248272788055,
+      "loss": 0.5086,
+      "step": 200
+    },
+    {
+      "epoch": 0.16380655226209048,
+      "grad_norm": 2.3981761932373047,
+      "learning_rate": 0.0009767104969913081,
+      "loss": 0.4084,
+      "step": 210
+    },
+    {
+      "epoch": 0.17160686427457097,
+      "grad_norm": 3.7943010330200195,
+      "learning_rate": 0.000975596166703811,
+      "loss": 0.4524,
+      "step": 220
+    },
+    {
+      "epoch": 0.1794071762870515,
+      "grad_norm": 2.657606840133667,
+      "learning_rate": 0.0009744818364163138,
+      "loss": 0.3592,
+      "step": 230
+    },
+    {
+      "epoch": 0.187207488299532,
+      "grad_norm": 2.7629363536834717,
+      "learning_rate": 0.0009733675061288166,
+      "loss": 0.4263,
+      "step": 240
+    },
+    {
+      "epoch": 0.19500780031201248,
+      "grad_norm": 1.3749983310699463,
+      "learning_rate": 0.0009722531758413193,
+      "loss": 0.48,
+      "step": 250
+    },
+    {
+      "epoch": 0.20280811232449297,
+      "grad_norm": 2.648716449737549,
+      "learning_rate": 0.0009711388455538221,
+      "loss": 0.416,
+      "step": 260
+    },
+    {
+      "epoch": 0.21060842433697347,
+      "grad_norm": 1.5672308206558228,
+      "learning_rate": 0.0009700245152663249,
+      "loss": 0.4223,
+      "step": 270
+    },
+    {
+      "epoch": 0.21840873634945399,
+      "grad_norm": 2.618163585662842,
+      "learning_rate": 0.0009689101849788277,
+      "loss": 0.4172,
+      "step": 280
+    },
+    {
+      "epoch": 0.22620904836193448,
+      "grad_norm": 3.6365268230438232,
+      "learning_rate": 0.0009677958546913305,
+      "loss": 0.5501,
+      "step": 290
+    },
+    {
+      "epoch": 0.23400936037441497,
+      "grad_norm": 2.740039825439453,
+      "learning_rate": 0.0009666815244038332,
+      "loss": 0.3553,
+      "step": 300
+    },
+    {
+      "epoch": 0.24180967238689546,
+      "grad_norm": 3.406210422515869,
+      "learning_rate": 0.000965567194116336,
+      "loss": 0.3518,
+      "step": 310
+    },
+    {
+      "epoch": 0.24960998439937598,
+      "grad_norm": 1.4707075357437134,
+      "learning_rate": 0.000964452863828839,
+      "loss": 0.3452,
+      "step": 320
+    },
+    {
+      "epoch": 0.2574102964118565,
+      "grad_norm": 1.608324408531189,
+      "learning_rate": 0.0009633385335413417,
+      "loss": 0.4908,
+      "step": 330
+    },
+    {
+      "epoch": 0.26521060842433697,
+      "grad_norm": 4.090480327606201,
+      "learning_rate": 0.0009622242032538444,
+      "loss": 0.4597,
+      "step": 340
+    },
+    {
+      "epoch": 0.27301092043681746,
+      "grad_norm": 2.2214395999908447,
+      "learning_rate": 0.0009611098729663472,
+      "loss": 0.4552,
+      "step": 350
+    },
+    {
+      "epoch": 0.28081123244929795,
+      "grad_norm": 1.9134166240692139,
+      "learning_rate": 0.00095999554267885,
+      "loss": 0.3571,
+      "step": 360
+    },
+    {
+      "epoch": 0.28861154446177845,
+      "grad_norm": 1.8127851486206055,
+      "learning_rate": 0.0009588812123913528,
+      "loss": 0.3808,
+      "step": 370
+    },
+    {
+      "epoch": 0.296411856474259,
+      "grad_norm": 2.2262885570526123,
+      "learning_rate": 0.0009577668821038556,
+      "loss": 0.4099,
+      "step": 380
+    },
+    {
+      "epoch": 0.3042121684867395,
+      "grad_norm": 2.8041303157806396,
+      "learning_rate": 0.0009566525518163583,
+      "loss": 0.3988,
+      "step": 390
+    },
+    {
+      "epoch": 0.31201248049922,
+      "grad_norm": 6.797432899475098,
+      "learning_rate": 0.0009555382215288611,
+      "loss": 0.4728,
+      "step": 400
+    },
+    {
+      "epoch": 0.31981279251170047,
+      "grad_norm": 3.1861369609832764,
+      "learning_rate": 0.000954423891241364,
+      "loss": 0.3502,
+      "step": 410
+    },
+    {
+      "epoch": 0.32761310452418096,
+      "grad_norm": 2.9223642349243164,
+      "learning_rate": 0.0009533095609538667,
+      "loss": 0.4215,
+      "step": 420
+    },
+    {
+      "epoch": 0.33541341653666146,
+      "grad_norm": 6.848895072937012,
+      "learning_rate": 0.0009521952306663694,
+      "loss": 0.415,
+      "step": 430
+    },
+    {
+      "epoch": 0.34321372854914195,
+      "grad_norm": 9.054282188415527,
+      "learning_rate": 0.0009510809003788722,
+      "loss": 0.4667,
+      "step": 440
+    },
+    {
+      "epoch": 0.35101404056162244,
+      "grad_norm": 2.3005900382995605,
+      "learning_rate": 0.0009499665700913752,
+      "loss": 0.6565,
+      "step": 450
+    },
+    {
+      "epoch": 0.358814352574103,
+      "grad_norm": 2.9467573165893555,
+      "learning_rate": 0.000948852239803878,
+      "loss": 0.3837,
+      "step": 460
+    },
+    {
+      "epoch": 0.3666146645865835,
+      "grad_norm": 1.5977652072906494,
+      "learning_rate": 0.0009477379095163808,
+      "loss": 0.407,
+      "step": 470
+    },
+    {
+      "epoch": 0.374414976599064,
+      "grad_norm": 2.8274600505828857,
+      "learning_rate": 0.0009466235792288835,
+      "loss": 0.4501,
+      "step": 480
+    },
+    {
+      "epoch": 0.38221528861154447,
+      "grad_norm": 1.8566502332687378,
+      "learning_rate": 0.0009455092489413863,
+      "loss": 0.3146,
+      "step": 490
+    },
+    {
+      "epoch": 0.39001560062402496,
+      "grad_norm": 2.5871951580047607,
+      "learning_rate": 0.0009443949186538891,
+      "loss": 0.3335,
+      "step": 500
+    },
+    {
+      "epoch": 0.39781591263650545,
+      "grad_norm": 2.0552711486816406,
+      "learning_rate": 0.0009432805883663919,
+      "loss": 0.3698,
+      "step": 510
+    },
+    {
+      "epoch": 0.40561622464898595,
+      "grad_norm": 1.5244548320770264,
+      "learning_rate": 0.0009421662580788946,
+      "loss": 0.581,
+      "step": 520
+    },
+    {
+      "epoch": 0.41341653666146644,
+      "grad_norm": 1.5146633386611938,
+      "learning_rate": 0.0009410519277913974,
+      "loss": 0.3634,
+      "step": 530
+    },
+    {
+      "epoch": 0.42121684867394693,
+      "grad_norm": 2.394819736480713,
+      "learning_rate": 0.0009399375975039002,
+      "loss": 0.338,
+      "step": 540
+    },
+    {
+      "epoch": 0.4290171606864275,
+      "grad_norm": 2.3500325679779053,
+      "learning_rate": 0.000938823267216403,
+      "loss": 0.4025,
+      "step": 550
+    },
+    {
+      "epoch": 0.43681747269890797,
+      "grad_norm": 2.4186370372772217,
+      "learning_rate": 0.0009377089369289058,
+      "loss": 0.4095,
+      "step": 560
+    },
+    {
+      "epoch": 0.44461778471138846,
+      "grad_norm": 1.9770065546035767,
+      "learning_rate": 0.0009365946066414085,
+      "loss": 0.4275,
+      "step": 570
+    },
+    {
+      "epoch": 0.45241809672386896,
+      "grad_norm": 1.7679632902145386,
+      "learning_rate": 0.0009354802763539114,
+      "loss": 0.3949,
+      "step": 580
+    },
+    {
+      "epoch": 0.46021840873634945,
+      "grad_norm": 2.0794620513916016,
+      "learning_rate": 0.0009343659460664142,
+      "loss": 0.4208,
+      "step": 590
+    },
+    {
+      "epoch": 0.46801872074882994,
+      "grad_norm": 2.677424192428589,
+      "learning_rate": 0.000933251615778917,
+      "loss": 0.4089,
+      "step": 600
+    },
+    {
+      "epoch": 0.47581903276131043,
+      "grad_norm": 1.526112675666809,
+      "learning_rate": 0.0009321372854914197,
+      "loss": 0.4196,
+      "step": 610
+    },
+    {
+      "epoch": 0.4836193447737909,
+      "grad_norm": 1.8656370639801025,
+      "learning_rate": 0.0009310229552039225,
+      "loss": 0.3732,
+      "step": 620
+    },
+    {
+      "epoch": 0.4914196567862715,
+      "grad_norm": 3.3338847160339355,
+      "learning_rate": 0.0009299086249164253,
+      "loss": 0.4231,
+      "step": 630
+    },
+    {
+      "epoch": 0.49921996879875197,
+      "grad_norm": 2.1057350635528564,
+      "learning_rate": 0.0009287942946289281,
+      "loss": 0.3921,
+      "step": 640
+    },
+    {
+      "epoch": 0.5070202808112324,
+      "grad_norm": 1.544977068901062,
+      "learning_rate": 0.0009276799643414309,
+      "loss": 0.3748,
+      "step": 650
+    },
+    {
+      "epoch": 0.514820592823713,
+      "grad_norm": 3.4070258140563965,
+      "learning_rate": 0.0009265656340539336,
+      "loss": 0.4027,
+      "step": 660
+    },
+    {
+      "epoch": 0.5226209048361935,
+      "grad_norm": 5.5486931800842285,
+      "learning_rate": 0.0009254513037664364,
+      "loss": 0.4326,
+      "step": 670
+    },
+    {
+      "epoch": 0.5304212168486739,
+      "grad_norm": 3.6824769973754883,
+      "learning_rate": 0.0009243369734789392,
+      "loss": 0.4809,
+      "step": 680
+    },
+    {
+      "epoch": 0.5382215288611545,
+      "grad_norm": 6.3154778480529785,
+      "learning_rate": 0.000923222643191442,
+      "loss": 0.3588,
+      "step": 690
+    },
+    {
+      "epoch": 0.5460218408736349,
+      "grad_norm": 3.133465528488159,
+      "learning_rate": 0.0009221083129039447,
+      "loss": 0.4712,
+      "step": 700
+    },
+    {
+      "epoch": 0.5538221528861155,
+      "grad_norm": 4.222598552703857,
+      "learning_rate": 0.0009209939826164475,
+      "loss": 0.4135,
+      "step": 710
+    },
+    {
+      "epoch": 0.5616224648985959,
+      "grad_norm": 4.6125078201293945,
+      "learning_rate": 0.0009198796523289504,
+      "loss": 0.393,
+      "step": 720
+    },
+    {
+      "epoch": 0.5694227769110765,
+      "grad_norm": 6.543318748474121,
+      "learning_rate": 0.0009187653220414532,
+      "loss": 0.6138,
+      "step": 730
+    },
+    {
+      "epoch": 0.5772230889235569,
+      "grad_norm": 2.596463680267334,
+      "learning_rate": 0.000917650991753956,
+      "loss": 0.3694,
+      "step": 740
+    },
+    {
+      "epoch": 0.5850234009360374,
+      "grad_norm": 2.428490161895752,
+      "learning_rate": 0.0009165366614664587,
+      "loss": 0.4535,
+      "step": 750
+    },
+    {
+      "epoch": 0.592823712948518,
+      "grad_norm": 1.8790688514709473,
+      "learning_rate": 0.0009154223311789615,
+      "loss": 0.3986,
+      "step": 760
+    },
+    {
+      "epoch": 0.6006240249609984,
+      "grad_norm": 3.141587734222412,
+      "learning_rate": 0.0009143080008914643,
+      "loss": 0.3881,
+      "step": 770
+    },
+    {
+      "epoch": 0.608424336973479,
+      "grad_norm": 2.125810146331787,
+      "learning_rate": 0.0009131936706039671,
+      "loss": 0.4225,
+      "step": 780
+    },
+    {
+      "epoch": 0.6162246489859594,
+      "grad_norm": 2.532404661178589,
+      "learning_rate": 0.0009120793403164698,
+      "loss": 0.3423,
+      "step": 790
+    },
+    {
+      "epoch": 0.62402496099844,
+      "grad_norm": 3.6324350833892822,
+      "learning_rate": 0.0009109650100289726,
+      "loss": 0.4205,
+      "step": 800
+    },
+    {
+      "epoch": 0.6318252730109204,
+      "grad_norm": 1.4804415702819824,
+      "learning_rate": 0.0009098506797414754,
+      "loss": 0.4042,
+      "step": 810
+    },
+    {
+      "epoch": 0.6396255850234009,
+      "grad_norm": 1.5140562057495117,
+      "learning_rate": 0.0009087363494539782,
+      "loss": 0.3056,
+      "step": 820
+    },
+    {
+      "epoch": 0.6474258970358814,
+      "grad_norm": 2.470576047897339,
+      "learning_rate": 0.000907622019166481,
+      "loss": 0.391,
+      "step": 830
+    },
+    {
+      "epoch": 0.6552262090483619,
+      "grad_norm": 3.4496209621429443,
+      "learning_rate": 0.0009065076888789837,
+      "loss": 0.4163,
+      "step": 840
+    },
+    {
+      "epoch": 0.6630265210608425,
+      "grad_norm": 1.8823250532150269,
+      "learning_rate": 0.0009053933585914866,
+      "loss": 0.3877,
+      "step": 850
+    },
+    {
+      "epoch": 0.6708268330733229,
+      "grad_norm": 2.492297410964966,
+      "learning_rate": 0.0009042790283039894,
+      "loss": 0.3542,
+      "step": 860
+    },
+    {
+      "epoch": 0.6786271450858035,
+      "grad_norm": 3.977569341659546,
+      "learning_rate": 0.0009031646980164922,
+      "loss": 0.4168,
+      "step": 870
+    },
+    {
+      "epoch": 0.6864274570982839,
+      "grad_norm": 3.938462495803833,
+      "learning_rate": 0.0009020503677289949,
+      "loss": 0.4894,
+      "step": 880
+    },
+    {
+      "epoch": 0.6942277691107644,
+      "grad_norm": 1.3457701206207275,
+      "learning_rate": 0.0009009360374414977,
+      "loss": 0.4903,
+      "step": 890
+    },
+    {
+      "epoch": 0.7020280811232449,
+      "grad_norm": 6.3473124504089355,
+      "learning_rate": 0.0008998217071540005,
+      "loss": 0.4766,
+      "step": 900
+    },
+    {
+      "epoch": 0.7098283931357254,
+      "grad_norm": 3.145792245864868,
+      "learning_rate": 0.0008987073768665033,
+      "loss": 0.4119,
+      "step": 910
+    },
+    {
+      "epoch": 0.717628705148206,
+      "grad_norm": 1.809446930885315,
+      "learning_rate": 0.0008975930465790061,
+      "loss": 0.3991,
+      "step": 920
+    },
+    {
+      "epoch": 0.7254290171606864,
+      "grad_norm": 1.7960044145584106,
+      "learning_rate": 0.0008964787162915088,
+      "loss": 0.3095,
+      "step": 930
+    },
+    {
+      "epoch": 0.733229329173167,
+      "grad_norm": 2.9710285663604736,
+      "learning_rate": 0.0008953643860040116,
+      "loss": 0.5104,
+      "step": 940
+    },
+    {
+      "epoch": 0.7410296411856474,
+      "grad_norm": 2.460524797439575,
+      "learning_rate": 0.0008942500557165144,
+      "loss": 0.4332,
+      "step": 950
+    },
+    {
+      "epoch": 0.748829953198128,
+      "grad_norm": 1.6166704893112183,
+      "learning_rate": 0.0008931357254290172,
+      "loss": 0.3856,
+      "step": 960
+    },
+    {
+      "epoch": 0.7566302652106084,
+      "grad_norm": 1.747750163078308,
+      "learning_rate": 0.0008920213951415199,
+      "loss": 0.3973,
+      "step": 970
+    },
+    {
+      "epoch": 0.7644305772230889,
+      "grad_norm": 1.4469414949417114,
+      "learning_rate": 0.0008909070648540227,
+      "loss": 0.3344,
+      "step": 980
+    },
+    {
+      "epoch": 0.7722308892355694,
+      "grad_norm": 4.661273956298828,
+      "learning_rate": 0.0008897927345665256,
+      "loss": 0.6003,
+      "step": 990
+    },
+    {
+      "epoch": 0.7800312012480499,
+      "grad_norm": 3.6588950157165527,
+      "learning_rate": 0.0008886784042790284,
+      "loss": 0.473,
+      "step": 1000
+    },
+    {
+      "epoch": 0.7878315132605305,
+      "grad_norm": 2.208383798599243,
+      "learning_rate": 0.0008876755070202809,
+      "loss": 0.3899,
+      "step": 1010
+    },
+    {
+      "epoch": 0.7956318252730109,
+      "grad_norm": 2.3569576740264893,
+      "learning_rate": 0.0008865611767327836,
+      "loss": 0.3871,
+      "step": 1020
+    },
+    {
+      "epoch": 0.8034321372854915,
+      "grad_norm": 2.7071454524993896,
+      "learning_rate": 0.0008854468464452864,
+      "loss": 0.3149,
+      "step": 1030
+    },
+    {
+      "epoch": 0.8112324492979719,
+      "grad_norm": 2.8024532794952393,
+      "learning_rate": 0.0008843325161577892,
+      "loss": 0.3328,
+      "step": 1040
+    },
+    {
+      "epoch": 0.8190327613104524,
+      "grad_norm": 3.0969290733337402,
+      "learning_rate": 0.000883218185870292,
+      "loss": 0.3948,
+      "step": 1050
+    },
+    {
+      "epoch": 0.8268330733229329,
+      "grad_norm": 2.982484817504883,
+      "learning_rate": 0.0008821038555827947,
+      "loss": 0.4323,
+      "step": 1060
+    },
+    {
+      "epoch": 0.8346333853354134,
+      "grad_norm": 3.133814573287964,
+      "learning_rate": 0.0008809895252952975,
+      "loss": 0.4393,
+      "step": 1070
+    },
+    {
+      "epoch": 0.8424336973478939,
+      "grad_norm": 3.3123364448547363,
+      "learning_rate": 0.0008798751950078003,
+      "loss": 0.3244,
+      "step": 1080
+    },
+    {
+      "epoch": 0.8502340093603744,
+      "grad_norm": 2.308555841445923,
+      "learning_rate": 0.0008787608647203032,
+      "loss": 0.424,
+      "step": 1090
+    },
+    {
+      "epoch": 0.858034321372855,
+      "grad_norm": 3.654137134552002,
+      "learning_rate": 0.000877646534432806,
+      "loss": 0.3445,
+      "step": 1100
+    },
+    {
+      "epoch": 0.8658346333853354,
+      "grad_norm": 2.149843692779541,
+      "learning_rate": 0.0008765322041453087,
+      "loss": 0.3398,
+      "step": 1110
+    },
+    {
+      "epoch": 0.8736349453978159,
+      "grad_norm": 3.1334431171417236,
+      "learning_rate": 0.0008754178738578115,
+      "loss": 0.3333,
+      "step": 1120
+    },
+    {
+      "epoch": 0.8814352574102964,
+      "grad_norm": 2.2942090034484863,
+      "learning_rate": 0.0008743035435703143,
+      "loss": 0.4188,
+      "step": 1130
+    },
+    {
+      "epoch": 0.8892355694227769,
+      "grad_norm": 2.0195343494415283,
+      "learning_rate": 0.0008731892132828171,
+      "loss": 0.4047,
+      "step": 1140
+    },
+    {
+      "epoch": 0.8970358814352574,
+      "grad_norm": 2.3850839138031006,
+      "learning_rate": 0.0008720748829953198,
+      "loss": 0.3931,
+      "step": 1150
+    },
+    {
+      "epoch": 0.9048361934477379,
+      "grad_norm": 1.6200228929519653,
+      "learning_rate": 0.0008709605527078226,
+      "loss": 0.409,
+      "step": 1160
+    },
+    {
+      "epoch": 0.9126365054602185,
+      "grad_norm": 2.9001989364624023,
+      "learning_rate": 0.0008698462224203254,
+      "loss": 0.4289,
+      "step": 1170
+    },
+    {
+      "epoch": 0.9204368174726989,
+      "grad_norm": 1.52889883518219,
+      "learning_rate": 0.0008687318921328282,
+      "loss": 0.3575,
+      "step": 1180
+    },
+    {
+      "epoch": 0.9282371294851794,
+      "grad_norm": 2.359733819961548,
+      "learning_rate": 0.000867617561845331,
+      "loss": 0.3837,
+      "step": 1190
+    },
+    {
+      "epoch": 0.9360374414976599,
+      "grad_norm": 2.3807597160339355,
+      "learning_rate": 0.0008665032315578337,
+      "loss": 0.4206,
+      "step": 1200
+    },
+    {
+      "epoch": 0.9438377535101404,
+      "grad_norm": 1.8366179466247559,
+      "learning_rate": 0.0008653889012703365,
+      "loss": 0.3101,
+      "step": 1210
+    },
+    {
+      "epoch": 0.9516380655226209,
+      "grad_norm": 3.1048014163970947,
+      "learning_rate": 0.0008642745709828393,
+      "loss": 0.4104,
+      "step": 1220
+    },
+    {
+      "epoch": 0.9594383775351014,
+      "grad_norm": 1.5314342975616455,
+      "learning_rate": 0.0008631602406953422,
+      "loss": 0.3539,
+      "step": 1230
+    },
+    {
+      "epoch": 0.9672386895475819,
+      "grad_norm": 2.8501791954040527,
+      "learning_rate": 0.0008620459104078449,
+      "loss": 0.4104,
+      "step": 1240
+    },
+    {
+      "epoch": 0.9750390015600624,
+      "grad_norm": 5.708191394805908,
+      "learning_rate": 0.0008609315801203477,
+      "loss": 0.4038,
+      "step": 1250
+    },
+    {
+      "epoch": 0.982839313572543,
+      "grad_norm": 2.173867702484131,
+      "learning_rate": 0.0008598172498328505,
+      "loss": 0.338,
+      "step": 1260
+    },
+    {
+      "epoch": 0.9906396255850234,
+      "grad_norm": 2.7057418823242188,
+      "learning_rate": 0.0008587029195453533,
+      "loss": 0.3941,
+      "step": 1270
+    },
+    {
+      "epoch": 0.9984399375975039,
+      "grad_norm": 1.3492989540100098,
+      "learning_rate": 0.0008575885892578561,
+      "loss": 0.3696,
+      "step": 1280
+    },
+    {
+      "epoch": 1.0,
+      "eval_loss": 0.4044143855571747,
+      "eval_runtime": 635.8286,
+      "eval_samples_per_second": 0.865,
+      "eval_steps_per_second": 0.865,
+      "eval_wer": 26.665379416875844,
+      "step": 1282
+    },
+    {
+      "epoch": 1.0062402496099845,
+      "grad_norm": 2.5727291107177734,
+      "learning_rate": 0.0008564742589703588,
+      "loss": 0.3369,
+      "step": 1290
+    },
+    {
+      "epoch": 1.0140405616224648,
+      "grad_norm": 2.681490182876587,
+      "learning_rate": 0.0008553599286828616,
+      "loss": 0.2434,
+      "step": 1300
+    },
+    {
+      "epoch": 1.0218408736349454,
+      "grad_norm": 2.0296504497528076,
+      "learning_rate": 0.0008542455983953644,
+      "loss": 0.2122,
+      "step": 1310
+    },
+    {
+      "epoch": 1.029641185647426,
+      "grad_norm": 2.277512311935425,
+      "learning_rate": 0.0008531312681078672,
+      "loss": 0.3932,
+      "step": 1320
+    },
+    {
+      "epoch": 1.0374414976599065,
+      "grad_norm": 4.77215576171875,
+      "learning_rate": 0.0008520169378203699,
+      "loss": 0.2658,
+      "step": 1330
+    },
+    {
+      "epoch": 1.045241809672387,
+      "grad_norm": 1.4027091264724731,
+      "learning_rate": 0.0008509026075328727,
+      "loss": 0.3524,
+      "step": 1340
+    },
+    {
+      "epoch": 1.0530421216848673,
+      "grad_norm": 2.2849514484405518,
+      "learning_rate": 0.0008497882772453755,
+      "loss": 0.2705,
+      "step": 1350
+    },
+    {
+      "epoch": 1.0608424336973479,
+      "grad_norm": 1.2896777391433716,
+      "learning_rate": 0.0008486739469578784,
+      "loss": 0.264,
+      "step": 1360
+    },
+    {
+      "epoch": 1.0686427457098284,
+      "grad_norm": 2.4552128314971924,
+      "learning_rate": 0.0008475596166703812,
+      "loss": 0.2833,
+      "step": 1370
+    },
+    {
+      "epoch": 1.076443057722309,
+      "grad_norm": 2.044693946838379,
+      "learning_rate": 0.0008464452863828839,
+      "loss": 0.2116,
+      "step": 1380
+    },
+    {
+      "epoch": 1.0842433697347893,
+      "grad_norm": 1.3727463483810425,
+      "learning_rate": 0.0008453309560953867,
+      "loss": 0.2533,
+      "step": 1390
+    },
+    {
+      "epoch": 1.0920436817472698,
+      "grad_norm": 1.6917822360992432,
+      "learning_rate": 0.0008442166258078895,
+      "loss": 0.4259,
+      "step": 1400
+    },
+    {
+      "epoch": 1.0998439937597504,
+      "grad_norm": 2.198549747467041,
+      "learning_rate": 0.0008431022955203923,
+      "loss": 0.3161,
+      "step": 1410
+    },
+    {
+      "epoch": 1.107644305772231,
+      "grad_norm": 1.7467869520187378,
+      "learning_rate": 0.000841987965232895,
+      "loss": 0.2377,
+      "step": 1420
+    },
+    {
+      "epoch": 1.1154446177847115,
+      "grad_norm": 2.5347695350646973,
+      "learning_rate": 0.0008408736349453978,
+      "loss": 0.3258,
+      "step": 1430
+    },
+    {
+      "epoch": 1.1232449297971918,
+      "grad_norm": 1.9081774950027466,
+      "learning_rate": 0.0008397593046579006,
+      "loss": 0.2462,
+      "step": 1440
+    },
+    {
+      "epoch": 1.1310452418096724,
+      "grad_norm": 1.5889848470687866,
+      "learning_rate": 0.0008386449743704034,
+      "loss": 0.2404,
+      "step": 1450
+    },
+    {
+      "epoch": 1.138845553822153,
+      "grad_norm": 1.8944768905639648,
+      "learning_rate": 0.0008375306440829062,
+      "loss": 0.2707,
+      "step": 1460
+    },
+    {
+      "epoch": 1.1466458658346335,
+      "grad_norm": 2.5448453426361084,
+      "learning_rate": 0.0008364163137954089,
+      "loss": 0.3342,
+      "step": 1470
+    },
+    {
+      "epoch": 1.154446177847114,
+      "grad_norm": 2.0936005115509033,
+      "learning_rate": 0.0008353019835079117,
+      "loss": 0.385,
+      "step": 1480
+    },
+    {
+      "epoch": 1.1622464898595943,
+      "grad_norm": 2.614129066467285,
+      "learning_rate": 0.0008341876532204145,
+      "loss": 0.2817,
+      "step": 1490
+    },
+    {
+      "epoch": 1.1700468018720749,
+      "grad_norm": 1.6156001091003418,
+      "learning_rate": 0.0008330733229329174,
+      "loss": 0.3527,
+      "step": 1500
+    },
+    {
+      "epoch": 1.1778471138845554,
+      "grad_norm": 1.4294220209121704,
+      "learning_rate": 0.0008319589926454201,
+      "loss": 0.2469,
+      "step": 1510
+    },
+    {
+      "epoch": 1.185647425897036,
+      "grad_norm": 3.197176456451416,
+      "learning_rate": 0.0008308446623579229,
+      "loss": 0.2753,
+      "step": 1520
+    },
+    {
+      "epoch": 1.1934477379095163,
+      "grad_norm": 2.1629223823547363,
+      "learning_rate": 0.0008297303320704257,
+      "loss": 0.3951,
+      "step": 1530
+    },
+    {
+      "epoch": 1.2012480499219969,
+      "grad_norm": 2.9824419021606445,
+      "learning_rate": 0.0008286160017829285,
+      "loss": 0.3278,
+      "step": 1540
+    },
+    {
+      "epoch": 1.2090483619344774,
+      "grad_norm": 2.866138219833374,
+      "learning_rate": 0.0008275016714954313,
+      "loss": 0.4267,
+      "step": 1550
+    },
+    {
+      "epoch": 1.216848673946958,
+      "grad_norm": 2.36781644821167,
+      "learning_rate": 0.000826387341207934,
+      "loss": 0.3414,
+      "step": 1560
+    },
+    {
+      "epoch": 1.2246489859594383,
+      "grad_norm": 1.8305447101593018,
+      "learning_rate": 0.0008252730109204368,
+      "loss": 0.2925,
+      "step": 1570
+    },
+    {
+      "epoch": 1.2324492979719188,
+      "grad_norm": 1.9879776239395142,
+      "learning_rate": 0.0008241586806329396,
+      "loss": 0.357,
+      "step": 1580
+    },
+    {
+      "epoch": 1.2402496099843994,
+      "grad_norm": 2.183350086212158,
+      "learning_rate": 0.0008230443503454424,
+      "loss": 0.3409,
+      "step": 1590
+    },
+    {
+      "epoch": 1.24804992199688,
+      "grad_norm": 2.197072744369507,
+      "learning_rate": 0.0008219300200579451,
+      "loss": 0.33,
+      "step": 1600
+    },
+    {
+      "epoch": 1.2558502340093605,
+      "grad_norm": 3.2065696716308594,
+      "learning_rate": 0.0008208156897704479,
+      "loss": 0.2853,
+      "step": 1610
+    },
+    {
+      "epoch": 1.2636505460218408,
+      "grad_norm": 2.0581350326538086,
+      "learning_rate": 0.0008197013594829507,
+      "loss": 0.3647,
+      "step": 1620
+    },
+    {
+      "epoch": 1.2714508580343213,
+      "grad_norm": 3.149153232574463,
+      "learning_rate": 0.0008185870291954536,
+      "loss": 0.3921,
+      "step": 1630
+    },
+    {
+      "epoch": 1.2792511700468019,
+      "grad_norm": 2.5097105503082275,
+      "learning_rate": 0.0008174726989079563,
+      "loss": 0.3302,
+      "step": 1640
+    },
+    {
+      "epoch": 1.2870514820592824,
+      "grad_norm": 2.7537474632263184,
+      "learning_rate": 0.0008163583686204591,
+      "loss": 0.3286,
+      "step": 1650
+    },
+    {
+      "epoch": 1.294851794071763,
+      "grad_norm": 1.966965675354004,
+      "learning_rate": 0.0008152440383329619,
+      "loss": 0.2785,
+      "step": 1660
+    },
+    {
+      "epoch": 1.3026521060842433,
+      "grad_norm": 1.9159988164901733,
+      "learning_rate": 0.0008141297080454647,
+      "loss": 0.3299,
+      "step": 1670
+    },
+    {
+      "epoch": 1.3104524180967239,
+      "grad_norm": 2.2212252616882324,
+      "learning_rate": 0.0008130153777579675,
+      "loss": 0.2807,
+      "step": 1680
+    },
+    {
+      "epoch": 1.3182527301092044,
+      "grad_norm": 4.194318771362305,
+      "learning_rate": 0.0008119010474704702,
+      "loss": 0.3161,
+      "step": 1690
+    },
+    {
+      "epoch": 1.3260530421216847,
+      "grad_norm": 1.7189604043960571,
+      "learning_rate": 0.000810786717182973,
+      "loss": 0.3356,
+      "step": 1700
+    },
+    {
+      "epoch": 1.3338533541341655,
+      "grad_norm": 1.5196418762207031,
+      "learning_rate": 0.0008096723868954758,
+      "loss": 0.2616,
+      "step": 1710
+    },
+    {
+      "epoch": 1.3416536661466458,
+      "grad_norm": 1.497450351715088,
+      "learning_rate": 0.0008085580566079786,
+      "loss": 0.2946,
+      "step": 1720
+    },
+    {
+      "epoch": 1.3494539781591264,
+      "grad_norm": 1.74885892868042,
+      "learning_rate": 0.0008074437263204813,
+      "loss": 0.2801,
+      "step": 1730
+    },
+    {
+      "epoch": 1.357254290171607,
+      "grad_norm": 2.040701389312744,
+      "learning_rate": 0.0008063293960329841,
+      "loss": 0.3203,
+      "step": 1740
+    },
+    {
+      "epoch": 1.3650546021840873,
+      "grad_norm": 3.760457754135132,
+      "learning_rate": 0.0008052150657454869,
+      "loss": 0.4569,
+      "step": 1750
+    },
+    {
+      "epoch": 1.3728549141965678,
+      "grad_norm": 2.92971134185791,
+      "learning_rate": 0.0008041007354579897,
+      "loss": 0.3321,
+      "step": 1760
+    },
+    {
+      "epoch": 1.3806552262090483,
+      "grad_norm": 1.9461047649383545,
+      "learning_rate": 0.0008029864051704926,
+      "loss": 0.2696,
+      "step": 1770
+    },
+    {
+      "epoch": 1.388455538221529,
+      "grad_norm": 3.2626147270202637,
+      "learning_rate": 0.0008018720748829953,
+      "loss": 0.3322,
+      "step": 1780
+    },
+    {
+      "epoch": 1.3962558502340094,
+      "grad_norm": 2.1270642280578613,
+      "learning_rate": 0.0008007577445954981,
+      "loss": 0.2965,
+      "step": 1790
+    },
+    {
+      "epoch": 1.4040561622464898,
+      "grad_norm": 2.3174221515655518,
+      "learning_rate": 0.0007996434143080009,
+      "loss": 0.3425,
+      "step": 1800
+    },
+    {
+      "epoch": 1.4118564742589703,
+      "grad_norm": 2.5749576091766357,
+      "learning_rate": 0.0007985290840205037,
+      "loss": 0.3622,
+      "step": 1810
+    },
+    {
+      "epoch": 1.4196567862714509,
+      "grad_norm": 1.873813509941101,
+      "learning_rate": 0.0007974147537330064,
+      "loss": 0.2497,
+      "step": 1820
+    },
+    {
+      "epoch": 1.4274570982839314,
+      "grad_norm": 3.633928060531616,
+      "learning_rate": 0.0007963004234455092,
+      "loss": 0.4058,
+      "step": 1830
+    },
+    {
+      "epoch": 1.435257410296412,
+      "grad_norm": 2.356269598007202,
+      "learning_rate": 0.000795186093158012,
+      "loss": 0.2635,
+      "step": 1840
+    },
+    {
+      "epoch": 1.4430577223088923,
+      "grad_norm": 1.9108752012252808,
+      "learning_rate": 0.0007940717628705148,
+      "loss": 0.336,
+      "step": 1850
+    },
+    {
+      "epoch": 1.4508580343213728,
+      "grad_norm": 1.5505330562591553,
+      "learning_rate": 0.0007929574325830176,
+      "loss": 0.3493,
+      "step": 1860
+    },
+    {
+      "epoch": 1.4586583463338534,
+      "grad_norm": 1.9970422983169556,
+      "learning_rate": 0.0007918431022955203,
+      "loss": 0.273,
+      "step": 1870
+    },
+    {
+      "epoch": 1.466458658346334,
+      "grad_norm": 2.753758192062378,
+      "learning_rate": 0.0007907287720080231,
+      "loss": 0.2845,
+      "step": 1880
+    },
+    {
+      "epoch": 1.4742589703588145,
+      "grad_norm": NaN,
+      "learning_rate": 0.0007897258747492757,
+      "loss": 0.2544,
+      "step": 1890
+    },
+    {
+      "epoch": 1.4820592823712948,
+      "grad_norm": 3.0995099544525146,
+      "learning_rate": 0.0007886115444617785,
+      "loss": 0.2884,
+      "step": 1900
+    },
+    {
+      "epoch": 1.4898595943837754,
+      "grad_norm": 5.728559970855713,
+      "learning_rate": 0.0007874972141742812,
+      "loss": 0.2681,
+      "step": 1910
+    },
+    {
+      "epoch": 1.497659906396256,
+      "grad_norm": 1.492622971534729,
+      "learning_rate": 0.000786382883886784,
+      "loss": 0.2891,
+      "step": 1920
+    },
+    {
+      "epoch": 1.5054602184087362,
+      "grad_norm": 1.7419252395629883,
+      "learning_rate": 0.0007852685535992868,
+      "loss": 0.4089,
+      "step": 1930
+    },
+    {
+      "epoch": 1.513260530421217,
+      "grad_norm": 6.814690589904785,
+      "learning_rate": 0.0007841542233117896,
+      "loss": 0.3372,
+      "step": 1940
+    },
+    {
+      "epoch": 1.5210608424336973,
+      "grad_norm": 2.380725860595703,
+      "learning_rate": 0.0007830398930242924,
+      "loss": 0.3189,
+      "step": 1950
+    },
+    {
+      "epoch": 1.5288611544461779,
+      "grad_norm": 5.004116058349609,
+      "learning_rate": 0.0007819255627367951,
+      "loss": 0.3018,
+      "step": 1960
+    },
+    {
+      "epoch": 1.5366614664586584,
+      "grad_norm": 2.604365825653076,
+      "learning_rate": 0.0007808112324492979,
+      "loss": 0.3054,
+      "step": 1970
+    },
+    {
+      "epoch": 1.5444617784711387,
+      "grad_norm": 1.585584044456482,
+      "learning_rate": 0.0007796969021618007,
+      "loss": 0.3477,
+      "step": 1980
+    },
+    {
+      "epoch": 1.5522620904836193,
+      "grad_norm": 1.8678693771362305,
+      "learning_rate": 0.0007785825718743035,
+      "loss": 0.3577,
+      "step": 1990
+    },
+    {
+      "epoch": 1.5600624024960998,
+      "grad_norm": 1.654689073562622,
+      "learning_rate": 0.0007774682415868062,
+      "loss": 0.2625,
+      "step": 2000
+    },
+    {
+      "epoch": 1.5678627145085804,
+      "grad_norm": 2.108919858932495,
+      "learning_rate": 0.0007763539112993092,
+      "loss": 0.2497,
+      "step": 2010
+    },
+    {
+      "epoch": 1.575663026521061,
+      "grad_norm": 7.198604106903076,
+      "learning_rate": 0.000775239581011812,
+      "loss": 0.3382,
+      "step": 2020
+    },
+    {
+      "epoch": 1.5834633385335413,
+      "grad_norm": 2.2285892963409424,
+      "learning_rate": 0.0007741252507243148,
+      "loss": 0.2598,
+      "step": 2030
+    },
+    {
+      "epoch": 1.5912636505460218,
+      "grad_norm": 1.7743014097213745,
+      "learning_rate": 0.0007730109204368176,
+      "loss": 0.2757,
+      "step": 2040
+    },
+    {
+      "epoch": 1.5990639625585024,
+      "grad_norm": 1.7763789892196655,
+      "learning_rate": 0.0007718965901493203,
+      "loss": 0.2703,
+      "step": 2050
+    },
+    {
+      "epoch": 1.6068642745709827,
+      "grad_norm": 2.159956693649292,
+      "learning_rate": 0.000770782259861823,
+      "loss": 0.2824,
+      "step": 2060
+    },
+    {
+      "epoch": 1.6146645865834635,
+      "grad_norm": 1.4845560789108276,
+      "learning_rate": 0.0007696679295743259,
+      "loss": 0.2528,
+      "step": 2070
+    },
+    {
+      "epoch": 1.6224648985959438,
+      "grad_norm": 3.627887010574341,
+      "learning_rate": 0.0007685535992868287,
+      "loss": 0.3197,
+      "step": 2080
+    },
+    {
+      "epoch": 1.6302652106084243,
+      "grad_norm": 2.2174973487854004,
+      "learning_rate": 0.0007674392689993314,
+      "loss": 0.2994,
+      "step": 2090
+    },
+    {
+      "epoch": 1.6380655226209049,
+      "grad_norm": 2.5977325439453125,
+      "learning_rate": 0.0007663249387118342,
+      "loss": 0.2991,
+      "step": 2100
+    },
+    {
+      "epoch": 1.6458658346333852,
+      "grad_norm": 1.9066824913024902,
+      "learning_rate": 0.000765210608424337,
+      "loss": 0.2166,
+      "step": 2110
+    },
+    {
+      "epoch": 1.653666146645866,
+      "grad_norm": 1.7197297811508179,
+      "learning_rate": 0.0007640962781368398,
+      "loss": 0.2948,
+      "step": 2120
+    },
+    {
+      "epoch": 1.6614664586583463,
+      "grad_norm": 2.054304361343384,
+      "learning_rate": 0.0007629819478493426,
+      "loss": 0.3073,
+      "step": 2130
+    },
+    {
+      "epoch": 1.6692667706708268,
+      "grad_norm": 1.7934963703155518,
+      "learning_rate": 0.0007618676175618453,
+      "loss": 0.2667,
+      "step": 2140
+    },
+    {
+      "epoch": 1.6770670826833074,
+      "grad_norm": 2.5259838104248047,
+      "learning_rate": 0.0007607532872743482,
+      "loss": 0.3322,
+      "step": 2150
+    },
+    {
+      "epoch": 1.6848673946957877,
+      "grad_norm": 3.6354122161865234,
+      "learning_rate": 0.000759638956986851,
+      "loss": 0.3909,
+      "step": 2160
+    },
+    {
+      "epoch": 1.6926677067082685,
+      "grad_norm": 1.6722809076309204,
+      "learning_rate": 0.0007585246266993538,
+      "loss": 0.2987,
+      "step": 2170
+    },
+    {
+      "epoch": 1.7004680187207488,
+      "grad_norm": 4.3235015869140625,
+      "learning_rate": 0.0007574102964118565,
+      "loss": 0.3112,
+      "step": 2180
+    },
+    {
+      "epoch": 1.7082683307332294,
+      "grad_norm": 2.236316442489624,
+      "learning_rate": 0.0007562959661243593,
+      "loss": 0.3174,
+      "step": 2190
+    },
+    {
+      "epoch": 1.71606864274571,
+      "grad_norm": 21.32891273498535,
+      "learning_rate": 0.0007551816358368621,
+      "loss": 0.3696,
+      "step": 2200
+    },
+    {
+      "epoch": 1.7238689547581902,
+      "grad_norm": 2.4251410961151123,
+      "learning_rate": 0.0007540673055493649,
+      "loss": 0.3171,
+      "step": 2210
+    },
+    {
+      "epoch": 1.7316692667706708,
+      "grad_norm": 2.4152424335479736,
+      "learning_rate": 0.0007529529752618677,
+      "loss": 0.328,
+      "step": 2220
+    },
+    {
+      "epoch": 1.7394695787831513,
+      "grad_norm": 2.0988574028015137,
+      "learning_rate": 0.0007518386449743704,
+      "loss": 0.3061,
+      "step": 2230
+    },
+    {
+      "epoch": 1.7472698907956317,
+      "grad_norm": 4.469291687011719,
+      "learning_rate": 0.0007507243146868732,
+      "loss": 0.3315,
+      "step": 2240
+    },
+    {
+      "epoch": 1.7550702028081124,
+      "grad_norm": 2.4917778968811035,
+      "learning_rate": 0.000749609984399376,
+      "loss": 0.4296,
+      "step": 2250
+    },
+    {
+      "epoch": 1.7628705148205928,
+      "grad_norm": 3.073840379714966,
+      "learning_rate": 0.0007484956541118788,
+      "loss": 0.3685,
+      "step": 2260
+    },
+    {
+      "epoch": 1.7706708268330733,
+      "grad_norm": 2.205733299255371,
+      "learning_rate": 0.0007473813238243815,
+      "loss": 0.2691,
+      "step": 2270
+    },
+    {
+      "epoch": 1.7784711388455539,
+      "grad_norm": 2.3948941230773926,
+      "learning_rate": 0.0007462669935368844,
+      "loss": 0.3065,
+      "step": 2280
+    },
+    {
+      "epoch": 1.7862714508580342,
+      "grad_norm": 2.6060824394226074,
+      "learning_rate": 0.0007451526632493872,
+      "loss": 0.3354,
+      "step": 2290
+    },
+    {
+      "epoch": 1.794071762870515,
+      "grad_norm": 3.2586774826049805,
+      "learning_rate": 0.00074403833296189,
+      "loss": 0.261,
+      "step": 2300
+    },
+    {
+      "epoch": 1.8018720748829953,
+      "grad_norm": 1.6417285203933716,
+      "learning_rate": 0.0007429240026743928,
+      "loss": 0.3828,
+      "step": 2310
+    },
+    {
+      "epoch": 1.8096723868954758,
+      "grad_norm": 4.006927967071533,
+      "learning_rate": 0.0007418096723868955,
+      "loss": 0.4228,
+      "step": 2320
+    },
+    {
+      "epoch": 1.8174726989079564,
+      "grad_norm": 2.5880374908447266,
+      "learning_rate": 0.0007406953420993983,
+      "loss": 0.2766,
+      "step": 2330
+    },
+    {
+      "epoch": 1.8252730109204367,
+      "grad_norm": 1.602337121963501,
+      "learning_rate": 0.0007395810118119011,
+      "loss": 0.2794,
+      "step": 2340
+    },
+    {
+      "epoch": 1.8330733229329175,
+      "grad_norm": 5.932153224945068,
+      "learning_rate": 0.0007384666815244039,
+      "loss": 0.3471,
+      "step": 2350
+    },
+    {
+      "epoch": 1.8408736349453978,
+      "grad_norm": 4.076808452606201,
+      "learning_rate": 0.0007373523512369066,
+      "loss": 0.2649,
+      "step": 2360
+    },
+    {
+      "epoch": 1.8486739469578783,
+      "grad_norm": 4.666397571563721,
+      "learning_rate": 0.0007362380209494094,
+      "loss": 0.2751,
+      "step": 2370
+    },
+    {
+      "epoch": 1.856474258970359,
+      "grad_norm": 3.792745590209961,
+      "learning_rate": 0.0007351236906619122,
+      "loss": 0.2606,
+      "step": 2380
+    },
+    {
+      "epoch": 1.8642745709828392,
+      "grad_norm": 2.5275423526763916,
+      "learning_rate": 0.000734009360374415,
+      "loss": 0.3469,
+      "step": 2390
+    },
+    {
+      "epoch": 1.8720748829953198,
+      "grad_norm": 1.59649658203125,
+      "learning_rate": 0.0007328950300869178,
+      "loss": 0.2724,
+      "step": 2400
+    },
+    {
+      "epoch": 1.8798751950078003,
+      "grad_norm": 3.7428267002105713,
+      "learning_rate": 0.0007317806997994206,
+      "loss": 0.4426,
+      "step": 2410
+    },
+    {
+      "epoch": 1.8876755070202809,
+      "grad_norm": 3.7439956665039062,
+      "learning_rate": 0.0007306663695119234,
+      "loss": 0.2901,
+      "step": 2420
+    },
+    {
+      "epoch": 1.8954758190327614,
+      "grad_norm": 2.3777198791503906,
+      "learning_rate": 0.0007295520392244262,
+      "loss": 0.2837,
+      "step": 2430
+    },
+    {
+      "epoch": 1.9032761310452417,
+      "grad_norm": 2.7654988765716553,
+      "learning_rate": 0.000728437708936929,
+      "loss": 0.277,
+      "step": 2440
+    },
+    {
+      "epoch": 1.9110764430577223,
+      "grad_norm": 1.8758680820465088,
+      "learning_rate": 0.0007273233786494317,
+      "loss": 0.2706,
+      "step": 2450
+    },
+    {
+      "epoch": 1.9188767550702028,
+      "grad_norm": 2.8725340366363525,
+      "learning_rate": 0.0007262090483619345,
+      "loss": 0.2566,
+      "step": 2460
+    },
+    {
+      "epoch": 1.9266770670826832,
+      "grad_norm": 2.4021458625793457,
+      "learning_rate": 0.0007250947180744373,
+      "loss": 0.2645,
+      "step": 2470
+    },
+    {
+      "epoch": 1.934477379095164,
+      "grad_norm": 2.8407838344573975,
+      "learning_rate": 0.0007239803877869401,
+      "loss": 0.3152,
+      "step": 2480
+    },
+    {
+      "epoch": 1.9422776911076443,
+      "grad_norm": 3.606403112411499,
+      "learning_rate": 0.0007228660574994429,
+      "loss": 0.373,
+      "step": 2490
+    },
+    {
+      "epoch": 1.9500780031201248,
+      "grad_norm": 2.362473487854004,
+      "learning_rate": 0.0007217517272119456,
+      "loss": 0.4332,
+      "step": 2500
+    },
+    {
+      "epoch": 1.9578783151326054,
+      "grad_norm": 1.9711815118789673,
+      "learning_rate": 0.0007206373969244484,
+      "loss": 0.2606,
+      "step": 2510
+    },
+    {
+      "epoch": 1.9656786271450857,
+      "grad_norm": 2.683908224105835,
+      "learning_rate": 0.0007195230666369512,
+      "loss": 0.3582,
+      "step": 2520
+    },
+    {
+      "epoch": 1.9734789391575664,
+      "grad_norm": 2.5902493000030518,
+      "learning_rate": 0.000718408736349454,
+      "loss": 0.2694,
+      "step": 2530
+    },
+    {
+      "epoch": 1.9812792511700468,
+      "grad_norm": 3.92708420753479,
+      "learning_rate": 0.0007172944060619567,
+      "loss": 0.3083,
+      "step": 2540
+    },
+    {
+      "epoch": 1.9890795631825273,
+      "grad_norm": 2.6788370609283447,
+      "learning_rate": 0.0007161800757744596,
+      "loss": 0.2587,
+      "step": 2550
+    },
+    {
+      "epoch": 1.9968798751950079,
+      "grad_norm": 2.413313627243042,
+      "learning_rate": 0.0007150657454869624,
+      "loss": 0.2732,
+      "step": 2560
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 0.3617618680000305,
+      "eval_runtime": 36952.264,
+      "eval_samples_per_second": 0.015,
+      "eval_steps_per_second": 0.015,
+      "eval_wer": 22.793975670978952,
+      "step": 2564
+    },
+    {
+      "epoch": 2.004680187207488,
+      "grad_norm": 1.572451114654541,
+      "learning_rate": 0.0007139514151994652,
+      "loss": 0.2147,
+      "step": 2570
+    },
+    {
+      "epoch": 2.012480499219969,
+      "grad_norm": 2.4759023189544678,
+      "learning_rate": 0.000712837084911968,
+      "loss": 0.2015,
+      "step": 2580
+    },
+    {
+      "epoch": 2.0202808112324493,
+      "grad_norm": 1.6903563737869263,
+      "learning_rate": 0.0007117227546244707,
+      "loss": 0.1703,
+      "step": 2590
+    },
+    {
+      "epoch": 2.0280811232449296,
+      "grad_norm": 3.494985580444336,
+      "learning_rate": 0.0007106084243369735,
+      "loss": 0.2063,
+      "step": 2600
+    },
+    {
+      "epoch": 2.0358814352574104,
+      "grad_norm": 1.5024439096450806,
+      "learning_rate": 0.0007094940940494763,
+      "loss": 0.1807,
+      "step": 2610
+    },
+    {
+      "epoch": 2.0436817472698907,
+      "grad_norm": 1.9423105716705322,
+      "learning_rate": 0.0007083797637619791,
+      "loss": 0.1953,
+      "step": 2620
+    },
+    {
+      "epoch": 2.0514820592823715,
+      "grad_norm": 0.8572360277175903,
+      "learning_rate": 0.0007072654334744818,
+      "loss": 0.3423,
+      "step": 2630
+    },
+    {
+      "epoch": 2.059282371294852,
+      "grad_norm": 2.574855327606201,
+      "learning_rate": 0.0007061511031869846,
+      "loss": 0.1916,
+      "step": 2640
+    },
+    {
+      "epoch": 2.067082683307332,
+      "grad_norm": 2.2941219806671143,
+      "learning_rate": 0.0007050367728994874,
+      "loss": 0.2168,
+      "step": 2650
+    },
+    {
+      "epoch": 2.074882995319813,
+      "grad_norm": 2.0448150634765625,
+      "learning_rate": 0.0007039224426119902,
+      "loss": 0.2412,
+      "step": 2660
+    },
+    {
+      "epoch": 2.0826833073322932,
+      "grad_norm": 1.8044381141662598,
+      "learning_rate": 0.000702808112324493,
+      "loss": 0.3261,
+      "step": 2670
+    },
+    {
+      "epoch": 2.090483619344774,
+      "grad_norm": 2.752220630645752,
+      "learning_rate": 0.0007016937820369958,
+      "loss": 0.2007,
+      "step": 2680
+    },
+    {
+      "epoch": 2.0982839313572543,
+      "grad_norm": 1.4598050117492676,
+      "learning_rate": 0.0007005794517494986,
+      "loss": 0.1856,
+      "step": 2690
+    },
+    {
+      "epoch": 2.1060842433697347,
+      "grad_norm": 3.578192710876465,
+      "learning_rate": 0.0006994651214620014,
+      "loss": 0.1991,
+      "step": 2700
+    },
+    {
+      "epoch": 2.1138845553822154,
+      "grad_norm": 2.6971054077148438,
+      "learning_rate": 0.0006983507911745042,
+      "loss": 0.2134,
+      "step": 2710
+    },
+    {
+      "epoch": 2.1216848673946958,
+      "grad_norm": 2.437596559524536,
+      "learning_rate": 0.0006972364608870069,
+      "loss": 0.228,
+      "step": 2720
+    },
+    {
+      "epoch": 2.129485179407176,
+      "grad_norm": 2.6254658699035645,
+      "learning_rate": 0.0006961221305995097,
+      "loss": 0.1671,
+      "step": 2730
+    },
+    {
+      "epoch": 2.137285491419657,
+      "grad_norm": 1.3765720129013062,
+      "learning_rate": 0.0006950078003120125,
+      "loss": 0.2391,
+      "step": 2740
+    },
+    {
+      "epoch": 2.145085803432137,
+      "grad_norm": 2.192396879196167,
+      "learning_rate": 0.0006938934700245153,
+      "loss": 0.2331,
+      "step": 2750
+    },
+    {
+      "epoch": 2.152886115444618,
+      "grad_norm": 1.4418809413909912,
+      "learning_rate": 0.0006927791397370181,
+      "loss": 0.2041,
+      "step": 2760
+    },
+    {
+      "epoch": 2.1606864274570983,
+      "grad_norm": 2.553459882736206,
+      "learning_rate": 0.0006916648094495208,
+      "loss": 0.2342,
+      "step": 2770
+    },
+    {
+      "epoch": 2.1684867394695786,
+      "grad_norm": 1.7199114561080933,
+      "learning_rate": 0.0006905504791620236,
+      "loss": 0.1841,
+      "step": 2780
+    },
+    {
+      "epoch": 2.1762870514820594,
+      "grad_norm": 2.2145979404449463,
+      "learning_rate": 0.0006894361488745264,
+      "loss": 0.1821,
+      "step": 2790
+    },
+    {
+      "epoch": 2.1840873634945397,
+      "grad_norm": 2.434779405593872,
+      "learning_rate": 0.0006883218185870292,
+      "loss": 0.2115,
+      "step": 2800
+    },
+    {
+      "epoch": 2.1918876755070205,
+      "grad_norm": 1.0518474578857422,
+      "learning_rate": 0.0006872074882995319,
+      "loss": 0.2186,
+      "step": 2810
+    },
+    {
+      "epoch": 2.199687987519501,
+      "grad_norm": 1.9856785535812378,
+      "learning_rate": 0.0006860931580120348,
+      "loss": 0.2448,
+      "step": 2820
+    },
+    {
+      "epoch": 2.207488299531981,
+      "grad_norm": 17.3148136138916,
+      "learning_rate": 0.0006849788277245376,
+      "loss": 0.2372,
+      "step": 2830
+    },
+    {
+      "epoch": 2.215288611544462,
+      "grad_norm": 3.4590959548950195,
+      "learning_rate": 0.0006838644974370404,
+      "loss": 0.3179,
+      "step": 2840
+    },
+    {
+      "epoch": 2.223088923556942,
+      "grad_norm": 2.127650260925293,
+      "learning_rate": 0.0006827501671495432,
+      "loss": 0.2445,
+      "step": 2850
+    },
+    {
+      "epoch": 2.230889235569423,
+      "grad_norm": 1.895729660987854,
+      "learning_rate": 0.0006816358368620459,
+      "loss": 0.1966,
+      "step": 2860
+    },
+    {
+      "epoch": 2.2386895475819033,
+      "grad_norm": 1.4693467617034912,
+      "learning_rate": 0.0006805215065745487,
+      "loss": 0.2351,
+      "step": 2870
+    },
+    {
+      "epoch": 2.2464898595943836,
+      "grad_norm": 1.500453233718872,
+      "learning_rate": 0.0006794071762870515,
+      "loss": 0.1946,
+      "step": 2880
+    },
+    {
+      "epoch": 2.2542901716068644,
+      "grad_norm": 2.8631374835968018,
+      "learning_rate": 0.0006782928459995543,
+      "loss": 0.2367,
+      "step": 2890
+    },
+    {
+      "epoch": 2.2620904836193447,
+      "grad_norm": 1.5281766653060913,
+      "learning_rate": 0.000677178515712057,
+      "loss": 0.1855,
+      "step": 2900
+    },
+    {
+      "epoch": 2.2698907956318255,
+      "grad_norm": 2.3027195930480957,
+      "learning_rate": 0.0006760641854245598,
+      "loss": 0.2117,
+      "step": 2910
+    },
+    {
+      "epoch": 2.277691107644306,
+      "grad_norm": 2.9962668418884277,
+      "learning_rate": 0.0006749498551370626,
+      "loss": 0.2441,
+      "step": 2920
+    },
+    {
+      "epoch": 2.285491419656786,
+      "grad_norm": 2.489192008972168,
+      "learning_rate": 0.0006738355248495654,
+      "loss": 0.2212,
+      "step": 2930
+    },
+    {
+      "epoch": 2.293291731669267,
+      "grad_norm": 1.366945743560791,
+      "learning_rate": 0.0006727211945620682,
+      "loss": 0.2003,
+      "step": 2940
+    },
+    {
+      "epoch": 2.3010920436817472,
+      "grad_norm": 1.4768056869506836,
+      "learning_rate": 0.000671606864274571,
+      "loss": 0.3107,
+      "step": 2950
+    },
+    {
+      "epoch": 2.308892355694228,
+      "grad_norm": 1.797979474067688,
+      "learning_rate": 0.0006704925339870738,
+      "loss": 0.1978,
+      "step": 2960
+    },
+    {
+      "epoch": 2.3166926677067083,
+      "grad_norm": 1.9439573287963867,
+      "learning_rate": 0.0006693782036995766,
+      "loss": 0.2306,
+      "step": 2970
+    },
+    {
+      "epoch": 2.3244929797191887,
+      "grad_norm": 1.582478642463684,
+      "learning_rate": 0.0006682638734120794,
+      "loss": 0.2079,
+      "step": 2980
+    },
+    {
+      "epoch": 2.3322932917316694,
+      "grad_norm": 2.1245157718658447,
+      "learning_rate": 0.0006671495431245821,
+      "loss": 0.2258,
+      "step": 2990
+    },
+    {
+      "epoch": 2.3400936037441498,
+      "grad_norm": 2.427675724029541,
+      "learning_rate": 0.0006660352128370849,
+      "loss": 0.2846,
+      "step": 3000
+    },
+    {
+      "epoch": 2.34789391575663,
+      "grad_norm": 0.9095527529716492,
+      "learning_rate": 0.0006649208825495877,
+      "loss": 0.1746,
+      "step": 3010
+    },
+    {
+      "epoch": 2.355694227769111,
+      "grad_norm": 2.0468294620513916,
+      "learning_rate": 0.0006638065522620905,
+      "loss": 0.2645,
+      "step": 3020
+    },
+    {
+      "epoch": 2.363494539781591,
+      "grad_norm": 1.5703836679458618,
+      "learning_rate": 0.0006626922219745933,
+      "loss": 0.1896,
+      "step": 3030
+    },
+    {
+      "epoch": 2.371294851794072,
+      "grad_norm": 2.37263822555542,
+      "learning_rate": 0.000661577891687096,
+      "loss": 0.232,
+      "step": 3040
+    },
+    {
+      "epoch": 2.3790951638065523,
+      "grad_norm": 1.6041433811187744,
+      "learning_rate": 0.0006604635613995988,
+      "loss": 0.2135,
+      "step": 3050
+    },
+    {
+      "epoch": 2.3868954758190326,
+      "grad_norm": 1.87883722782135,
+      "learning_rate": 0.0006593492311121016,
+      "loss": 0.1993,
+      "step": 3060
+    },
+    {
+      "epoch": 2.3946957878315134,
+      "grad_norm": 2.5502099990844727,
+      "learning_rate": 0.0006582349008246044,
+      "loss": 0.2468,
+      "step": 3070
+    },
+    {
+      "epoch": 2.4024960998439937,
+      "grad_norm": 2.681384801864624,
+      "learning_rate": 0.0006571205705371071,
+      "loss": 0.2364,
+      "step": 3080
+    },
+    {
+      "epoch": 2.410296411856474,
+      "grad_norm": 1.2032707929611206,
+      "learning_rate": 0.00065600624024961,
+      "loss": 0.2732,
+      "step": 3090
+    },
+    {
+      "epoch": 2.418096723868955,
+      "grad_norm": 1.553661584854126,
+      "learning_rate": 0.0006548919099621128,
+      "loss": 0.2214,
+      "step": 3100
+    },
+    {
+      "epoch": 2.425897035881435,
+      "grad_norm": 0.9736389517784119,
+      "learning_rate": 0.0006537775796746156,
+      "loss": 0.1778,
+      "step": 3110
+    },
+    {
+      "epoch": 2.433697347893916,
+      "grad_norm": 1.911352276802063,
+      "learning_rate": 0.0006526632493871184,
+      "loss": 0.2062,
+      "step": 3120
+    },
+    {
+      "epoch": 2.4414976599063962,
+      "grad_norm": 1.4338595867156982,
+      "learning_rate": 0.0006515489190996211,
+      "loss": 0.193,
+      "step": 3130
+    },
+    {
+      "epoch": 2.4492979719188765,
+      "grad_norm": 1.3027153015136719,
+      "learning_rate": 0.0006504345888121239,
+      "loss": 0.3175,
+      "step": 3140
+    },
+    {
+      "epoch": 2.4570982839313573,
+      "grad_norm": 2.262709140777588,
+      "learning_rate": 0.0006493202585246267,
+      "loss": 0.2874,
+      "step": 3150
+    },
+    {
+      "epoch": 2.4648985959438376,
+      "grad_norm": 1.36016845703125,
+      "learning_rate": 0.0006482059282371295,
+      "loss": 0.1603,
+      "step": 3160
+    },
+    {
+      "epoch": 2.4726989079563184,
+      "grad_norm": 3.441779613494873,
+      "learning_rate": 0.0006470915979496322,
+      "loss": 0.4564,
+      "step": 3170
+    },
+    {
+      "epoch": 2.4804992199687987,
+      "grad_norm": 1.8022549152374268,
+      "learning_rate": 0.000645977267662135,
+      "loss": 0.1928,
+      "step": 3180
+    },
+    {
+      "epoch": 2.488299531981279,
+      "grad_norm": 2.104497194290161,
+      "learning_rate": 0.0006448629373746378,
+      "loss": 0.2047,
+      "step": 3190
+    },
+    {
+      "epoch": 2.49609984399376,
+      "grad_norm": 2.5888454914093018,
+      "learning_rate": 0.0006437486070871406,
+      "loss": 0.2591,
+      "step": 3200
+    },
+    {
+      "epoch": 2.50390015600624,
+      "grad_norm": 1.3476183414459229,
+      "learning_rate": 0.0006426342767996434,
+      "loss": 0.2636,
+      "step": 3210
+    },
+    {
+      "epoch": 2.511700468018721,
+      "grad_norm": 2.888965606689453,
+      "learning_rate": 0.0006415199465121462,
+      "loss": 0.216,
+      "step": 3220
+    },
+    {
+      "epoch": 2.5195007800312013,
+      "grad_norm": 1.8190357685089111,
+      "learning_rate": 0.000640405616224649,
+      "loss": 0.3498,
+      "step": 3230
+    },
+    {
+      "epoch": 2.5273010920436816,
+      "grad_norm": 1.5132287740707397,
+      "learning_rate": 0.0006392912859371518,
+      "loss": 0.227,
+      "step": 3240
+    },
+    {
+      "epoch": 2.5351014040561624,
+      "grad_norm": 2.8061325550079346,
+      "learning_rate": 0.0006381769556496546,
+      "loss": 0.2089,
+      "step": 3250
+    },
+    {
+      "epoch": 2.5429017160686427,
+      "grad_norm": 2.2521767616271973,
+      "learning_rate": 0.0006370626253621573,
+      "loss": 0.2362,
+      "step": 3260
+    },
+    {
+      "epoch": 2.5507020280811235,
+      "grad_norm": 1.4635623693466187,
+      "learning_rate": 0.0006359482950746601,
+      "loss": 0.1901,
+      "step": 3270
+    },
+    {
+      "epoch": 2.5585023400936038,
+      "grad_norm": 1.9882755279541016,
+      "learning_rate": 0.0006348339647871629,
+      "loss": 0.2387,
+      "step": 3280
+    },
+    {
+      "epoch": 2.566302652106084,
+      "grad_norm": 6.737502098083496,
+      "learning_rate": 0.0006337196344996657,
+      "loss": 0.2278,
+      "step": 3290
+    },
+    {
+      "epoch": 2.574102964118565,
+      "grad_norm": 1.686926245689392,
+      "learning_rate": 0.0006326053042121685,
+      "loss": 0.3118,
+      "step": 3300
+    },
+    {
+      "epoch": 2.581903276131045,
+      "grad_norm": 1.436584234237671,
+      "learning_rate": 0.0006314909739246712,
+      "loss": 0.1787,
+      "step": 3310
+    },
+    {
+      "epoch": 2.589703588143526,
+      "grad_norm": 3.646476984024048,
+      "learning_rate": 0.000630376643637174,
+      "loss": 0.2165,
+      "step": 3320
+    },
+    {
+      "epoch": 2.5975039001560063,
+      "grad_norm": 1.8787113428115845,
+      "learning_rate": 0.0006292623133496768,
+      "loss": 0.1993,
+      "step": 3330
+    },
+    {
+      "epoch": 2.6053042121684866,
+      "grad_norm": 2.4485440254211426,
+      "learning_rate": 0.0006281479830621796,
+      "loss": 0.24,
+      "step": 3340
+    },
+    {
+      "epoch": 2.6131045241809674,
+      "grad_norm": 2.73197865486145,
+      "learning_rate": 0.0006270336527746824,
+      "loss": 0.228,
+      "step": 3350
+    },
+    {
+      "epoch": 2.6209048361934477,
+      "grad_norm": 3.2943315505981445,
+      "learning_rate": 0.0006259193224871852,
+      "loss": 0.2289,
+      "step": 3360
+    },
+    {
+      "epoch": 2.6287051482059285,
+      "grad_norm": 2.165308952331543,
+      "learning_rate": 0.000624804992199688,
+      "loss": 0.2435,
+      "step": 3370
+    },
+    {
+      "epoch": 2.636505460218409,
+      "grad_norm": 2.3766629695892334,
+      "learning_rate": 0.0006236906619121908,
+      "loss": 0.26,
+      "step": 3380
+    },
+    {
+      "epoch": 2.644305772230889,
+      "grad_norm": 1.1527057886123657,
+      "learning_rate": 0.0006225763316246936,
+      "loss": 0.2486,
+      "step": 3390
+    },
+    {
+      "epoch": 2.6521060842433695,
+      "grad_norm": 2.6304874420166016,
+      "learning_rate": 0.0006214620013371963,
+      "loss": 0.3827,
+      "step": 3400
+    },
+    {
+      "epoch": 2.6599063962558502,
+      "grad_norm": 1.5219537019729614,
+      "learning_rate": 0.0006203476710496991,
+      "loss": 0.2231,
+      "step": 3410
+    },
+    {
+      "epoch": 2.667706708268331,
+      "grad_norm": 1.9267528057098389,
+      "learning_rate": 0.0006192333407622019,
+      "loss": 0.2468,
+      "step": 3420
+    },
+    {
+      "epoch": 2.6755070202808113,
+      "grad_norm": 2.247861385345459,
+      "learning_rate": 0.0006181190104747047,
+      "loss": 0.2463,
+      "step": 3430
+    },
+    {
+      "epoch": 2.6833073322932917,
+      "grad_norm": 2.6133053302764893,
+      "learning_rate": 0.0006170046801872074,
+      "loss": 0.2636,
+      "step": 3440
+    },
+    {
+      "epoch": 2.691107644305772,
+      "grad_norm": 1.4675954580307007,
+      "learning_rate": 0.0006158903498997102,
+      "loss": 0.2141,
+      "step": 3450
+    },
+    {
+      "epoch": 2.6989079563182528,
+      "grad_norm": 3.3972527980804443,
+      "learning_rate": 0.000614776019612213,
+      "loss": 0.2102,
+      "step": 3460
+    },
+    {
+      "epoch": 2.706708268330733,
+      "grad_norm": 2.0020527839660645,
+      "learning_rate": 0.0006136616893247158,
+      "loss": 0.2618,
+      "step": 3470
+    },
+    {
+      "epoch": 2.714508580343214,
+      "grad_norm": 1.4799424409866333,
+      "learning_rate": 0.0006125473590372186,
+      "loss": 0.2644,
+      "step": 3480
+    },
+    {
+      "epoch": 2.722308892355694,
+      "grad_norm": 2.207921028137207,
+      "learning_rate": 0.0006114330287497214,
+      "loss": 0.2496,
+      "step": 3490
+    },
+    {
+      "epoch": 2.7301092043681745,
+      "grad_norm": 2.7361607551574707,
+      "learning_rate": 0.0006103186984622242,
+      "loss": 0.1917,
+      "step": 3500
+    },
+    {
+      "epoch": 2.7379095163806553,
+      "grad_norm": 2.6031532287597656,
+      "learning_rate": 0.000609204368174727,
+      "loss": 0.2358,
+      "step": 3510
+    },
+    {
+      "epoch": 2.7457098283931356,
+      "grad_norm": 1.442651629447937,
+      "learning_rate": 0.0006080900378872299,
+      "loss": 0.2929,
+      "step": 3520
+    },
+    {
+      "epoch": 2.7535101404056164,
+      "grad_norm": 2.853076457977295,
+      "learning_rate": 0.0006069757075997325,
+      "loss": 0.2287,
+      "step": 3530
+    },
+    {
+      "epoch": 2.7613104524180967,
+      "grad_norm": 1.827863335609436,
+      "learning_rate": 0.0006058613773122353,
+      "loss": 0.284,
+      "step": 3540
+    },
+    {
+      "epoch": 2.769110764430577,
+      "grad_norm": 1.29642915725708,
+      "learning_rate": 0.0006047470470247382,
+      "loss": 0.2156,
+      "step": 3550
+    },
+    {
+      "epoch": 2.776911076443058,
+      "grad_norm": 2.0758543014526367,
+      "learning_rate": 0.000603632716737241,
+      "loss": 0.4312,
+      "step": 3560
+    },
+    {
+      "epoch": 2.784711388455538,
+      "grad_norm": 2.1832942962646484,
+      "learning_rate": 0.0006025183864497438,
+      "loss": 0.22,
+      "step": 3570
+    },
+    {
+      "epoch": 2.792511700468019,
+      "grad_norm": 1.541040301322937,
+      "learning_rate": 0.0006014040561622464,
+      "loss": 0.2452,
+      "step": 3580
+    },
+    {
+      "epoch": 2.800312012480499,
+      "grad_norm": 1.5947539806365967,
+      "learning_rate": 0.0006002897258747492,
+      "loss": 0.2708,
+      "step": 3590
+    },
+    {
+      "epoch": 2.8081123244929795,
+      "grad_norm": 2.130390167236328,
+      "learning_rate": 0.000599175395587252,
+      "loss": 0.2194,
+      "step": 3600
+    },
+    {
+      "epoch": 2.8159126365054603,
+      "grad_norm": 2.382166862487793,
+      "learning_rate": 0.0005980610652997549,
+      "loss": 0.1931,
+      "step": 3610
+    },
+    {
+      "epoch": 2.8237129485179406,
+      "grad_norm": 4.421852111816406,
+      "learning_rate": 0.0005969467350122577,
+      "loss": 0.3081,
+      "step": 3620
+    },
+    {
+      "epoch": 2.8315132605304214,
+      "grad_norm": 1.1889768838882446,
+      "learning_rate": 0.0005958324047247605,
+      "loss": 0.2134,
+      "step": 3630
+    },
+    {
+      "epoch": 2.8393135725429017,
+      "grad_norm": 2.6204874515533447,
+      "learning_rate": 0.0005947180744372633,
+      "loss": 0.2128,
+      "step": 3640
+    },
+    {
+      "epoch": 2.847113884555382,
+      "grad_norm": 1.4705913066864014,
+      "learning_rate": 0.0005936037441497661,
+      "loss": 0.1894,
+      "step": 3650
+    },
+    {
+      "epoch": 2.854914196567863,
+      "grad_norm": 3.110135555267334,
+      "learning_rate": 0.0005924894138622688,
+      "loss": 0.2072,
+      "step": 3660
+    },
+    {
+      "epoch": 2.862714508580343,
+      "grad_norm": 1.3255491256713867,
+      "learning_rate": 0.0005913750835747716,
+      "loss": 0.2325,
+      "step": 3670
+    },
+    {
+      "epoch": 2.870514820592824,
+      "grad_norm": 2.2520713806152344,
+      "learning_rate": 0.0005902607532872744,
+      "loss": 0.2121,
+      "step": 3680
+    },
+    {
+      "epoch": 2.8783151326053042,
+      "grad_norm": 1.4630913734436035,
+      "learning_rate": 0.0005891464229997772,
+      "loss": 0.2281,
+      "step": 3690
+    },
+    {
+      "epoch": 2.8861154446177846,
+      "grad_norm": 2.0491292476654053,
+      "learning_rate": 0.00058803209271228,
+      "loss": 0.2227,
+      "step": 3700
+    },
+    {
+      "epoch": 2.8939157566302653,
+      "grad_norm": 6.133053302764893,
+      "learning_rate": 0.0005869177624247827,
+      "loss": 0.3106,
+      "step": 3710
+    },
+    {
+      "epoch": 2.9017160686427457,
+      "grad_norm": 2.3226382732391357,
+      "learning_rate": 0.0005858034321372855,
+      "loss": 0.2193,
+      "step": 3720
+    },
+    {
+      "epoch": 2.9095163806552264,
+      "grad_norm": 1.351330041885376,
+      "learning_rate": 0.0005846891018497883,
+      "loss": 0.1928,
+      "step": 3730
+    },
+    {
+      "epoch": 2.9173166926677068,
+      "grad_norm": 0.9889002442359924,
+      "learning_rate": 0.0005835747715622911,
+      "loss": 0.2395,
+      "step": 3740
+    },
+    {
+      "epoch": 2.925117004680187,
+      "grad_norm": 3.9231808185577393,
+      "learning_rate": 0.0005824604412747938,
+      "loss": 0.2288,
+      "step": 3750
+    },
+    {
+      "epoch": 2.932917316692668,
+      "grad_norm": 1.1299773454666138,
+      "learning_rate": 0.0005813461109872967,
+      "loss": 0.1996,
+      "step": 3760
+    },
+    {
+      "epoch": 2.940717628705148,
+      "grad_norm": 1.894411563873291,
+      "learning_rate": 0.0005802317806997995,
+      "loss": 0.2444,
+      "step": 3770
+    },
+    {
+      "epoch": 2.948517940717629,
+      "grad_norm": 3.100918769836426,
+      "learning_rate": 0.0005791174504123023,
+      "loss": 0.2723,
+      "step": 3780
+    },
+    {
+      "epoch": 2.9563182527301093,
+      "grad_norm": 2.921398639678955,
+      "learning_rate": 0.0005780031201248051,
+      "loss": 0.1963,
+      "step": 3790
+    },
+    {
+      "epoch": 2.9641185647425896,
+      "grad_norm": 2.898193120956421,
+      "learning_rate": 0.0005768887898373078,
+      "loss": 0.2262,
+      "step": 3800
+    },
+    {
+      "epoch": 2.97191887675507,
+      "grad_norm": 1.8510135412216187,
+      "learning_rate": 0.0005757744595498106,
+      "loss": 0.2253,
+      "step": 3810
+    },
+    {
+      "epoch": 2.9797191887675507,
+      "grad_norm": 2.716972827911377,
+      "learning_rate": 0.0005746601292623134,
+      "loss": 0.222,
+      "step": 3820
+    },
+    {
+      "epoch": 2.9875195007800315,
+      "grad_norm": 2.466111660003662,
+      "learning_rate": 0.0005735457989748162,
+      "loss": 0.2903,
+      "step": 3830
+    },
+    {
+      "epoch": 2.995319812792512,
+      "grad_norm": 2.409747838973999,
+      "learning_rate": 0.0005724314686873189,
+      "loss": 0.3088,
+      "step": 3840
+    },
+    {
+      "epoch": 3.0,
+      "eval_loss": 0.34428906440734863,
+      "eval_runtime": 557.5413,
+      "eval_samples_per_second": 0.986,
+      "eval_steps_per_second": 0.986,
+      "eval_wer": 24.454527901139215,
+      "step": 3846
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 8974,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 7,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 3.38849691254784e+18,
+  "train_batch_size": 3,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-3846/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:87bdde76802c6f6975c7a62016872baaa0d2d042f00a6968d4a9fadade014aa7
+size 5432

checkpoint-5128/README.md

@@ -0,0 +1,202 @@
+---
+base_model: openai/whisper-small
+library_name: peft
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.15.1

checkpoint-5128/adapter_config.json

@@ -0,0 +1,37 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": {
+    "base_model_class": "WhisperForConditionalGeneration",
+    "parent_library": "transformers.models.whisper.modeling_whisper"
+  },
+  "base_model_name_or_path": "openai/whisper-small",
+  "bias": "none",
+  "corda_config": null,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "q_proj",
+    "v_proj"
+  ],
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_rslora": false
+}

checkpoint-5128/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:db6500851e4c23f2fdbc19fbaf720c94b1d2fb427b034746b2ee225fb915cd26
+size 14176064

checkpoint-5128/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:85ed114d260da569a3506e6d06e0c4bbe851146f2f984d9a7a2d6b38049f52ce
+size 28432570

checkpoint-5128/preprocessor_config.json

@@ -0,0 +1,14 @@
+{
+  "chunk_length": 30,
+  "feature_extractor_type": "WhisperFeatureExtractor",
+  "feature_size": 80,
+  "hop_length": 160,
+  "n_fft": 400,
+  "n_samples": 480000,
+  "nb_max_frames": 3000,
+  "padding_side": "right",
+  "padding_value": 0.0,
+  "processor_class": "WhisperProcessor",
+  "return_attention_mask": false,
+  "sampling_rate": 16000
+}

checkpoint-5128/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:20f106a2e11d68effc698b58fefbecdb20e9ae022c8ce82d791541a4f1277fa6
+size 14244

checkpoint-5128/scaler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a4ea290651c7978e28e955dab01609852087e0a268908587c5b313eac7ba2bd5
+size 988

checkpoint-5128/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2376d1d1866375dad0a10468cbd5d9af6fe0bf3766014dca978a7a379cc04208
+size 1064

checkpoint-5128/trainer_state.json

@@ -0,0 +1,3653 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 4.0,
+  "eval_steps": 500,
+  "global_step": 5128,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.0078003120124804995,
+      "grad_norm": 2.144841432571411,
+      "learning_rate": 0.0009989971027412525,
+      "loss": 1.1174,
+      "step": 10
+    },
+    {
+      "epoch": 0.015600624024960999,
+      "grad_norm": 1.0625462532043457,
+      "learning_rate": 0.0009978827724537553,
+      "loss": 0.5035,
+      "step": 20
+    },
+    {
+      "epoch": 0.0234009360374415,
+      "grad_norm": 1.619573950767517,
+      "learning_rate": 0.0009967684421662581,
+      "loss": 0.3841,
+      "step": 30
+    },
+    {
+      "epoch": 0.031201248049921998,
+      "grad_norm": 3.2681403160095215,
+      "learning_rate": 0.000995654111878761,
+      "loss": 0.4974,
+      "step": 40
+    },
+    {
+      "epoch": 0.0390015600624025,
+      "grad_norm": 1.5944786071777344,
+      "learning_rate": 0.0009945397815912637,
+      "loss": 0.4797,
+      "step": 50
+    },
+    {
+      "epoch": 0.046801872074883,
+      "grad_norm": 1.8234444856643677,
+      "learning_rate": 0.0009934254513037665,
+      "loss": 0.637,
+      "step": 60
+    },
+    {
+      "epoch": 0.054602184087363496,
+      "grad_norm": 3.790844440460205,
+      "learning_rate": 0.0009923111210162691,
+      "loss": 0.4479,
+      "step": 70
+    },
+    {
+      "epoch": 0.062402496099843996,
+      "grad_norm": 2.9351046085357666,
+      "learning_rate": 0.000991196790728772,
+      "loss": 0.4332,
+      "step": 80
+    },
+    {
+      "epoch": 0.07020280811232449,
+      "grad_norm": 2.2770025730133057,
+      "learning_rate": 0.0009900824604412747,
+      "loss": 0.4399,
+      "step": 90
+    },
+    {
+      "epoch": 0.078003120124805,
+      "grad_norm": 1.86591637134552,
+      "learning_rate": 0.0009889681301537775,
+      "loss": 0.407,
+      "step": 100
+    },
+    {
+      "epoch": 0.08580343213728549,
+      "grad_norm": 1.6851640939712524,
+      "learning_rate": 0.0009878537998662805,
+      "loss": 0.3862,
+      "step": 110
+    },
+    {
+      "epoch": 0.093603744149766,
+      "grad_norm": 2.5469818115234375,
+      "learning_rate": 0.0009867394695787831,
+      "loss": 0.4166,
+      "step": 120
+    },
+    {
+      "epoch": 0.10140405616224649,
+      "grad_norm": 1.8210259675979614,
+      "learning_rate": 0.000985625139291286,
+      "loss": 0.3785,
+      "step": 130
+    },
+    {
+      "epoch": 0.10920436817472699,
+      "grad_norm": 2.031057119369507,
+      "learning_rate": 0.0009845108090037887,
+      "loss": 0.4177,
+      "step": 140
+    },
+    {
+      "epoch": 0.11700468018720749,
+      "grad_norm": 1.6646612882614136,
+      "learning_rate": 0.0009833964787162915,
+      "loss": 0.402,
+      "step": 150
+    },
+    {
+      "epoch": 0.12480499219968799,
+      "grad_norm": 1.8680285215377808,
+      "learning_rate": 0.0009822821484287943,
+      "loss": 0.3282,
+      "step": 160
+    },
+    {
+      "epoch": 0.13260530421216848,
+      "grad_norm": 1.8039604425430298,
+      "learning_rate": 0.0009811678181412971,
+      "loss": 0.3536,
+      "step": 170
+    },
+    {
+      "epoch": 0.14040561622464898,
+      "grad_norm": 3.3018901348114014,
+      "learning_rate": 0.0009800534878538,
+      "loss": 0.4595,
+      "step": 180
+    },
+    {
+      "epoch": 0.1482059282371295,
+      "grad_norm": 3.684013843536377,
+      "learning_rate": 0.0009789391575663027,
+      "loss": 0.4288,
+      "step": 190
+    },
+    {
+      "epoch": 0.15600624024961,
+      "grad_norm": 1.4512592554092407,
+      "learning_rate": 0.0009778248272788055,
+      "loss": 0.5086,
+      "step": 200
+    },
+    {
+      "epoch": 0.16380655226209048,
+      "grad_norm": 2.3981761932373047,
+      "learning_rate": 0.0009767104969913081,
+      "loss": 0.4084,
+      "step": 210
+    },
+    {
+      "epoch": 0.17160686427457097,
+      "grad_norm": 3.7943010330200195,
+      "learning_rate": 0.000975596166703811,
+      "loss": 0.4524,
+      "step": 220
+    },
+    {
+      "epoch": 0.1794071762870515,
+      "grad_norm": 2.657606840133667,
+      "learning_rate": 0.0009744818364163138,
+      "loss": 0.3592,
+      "step": 230
+    },
+    {
+      "epoch": 0.187207488299532,
+      "grad_norm": 2.7629363536834717,
+      "learning_rate": 0.0009733675061288166,
+      "loss": 0.4263,
+      "step": 240
+    },
+    {
+      "epoch": 0.19500780031201248,
+      "grad_norm": 1.3749983310699463,
+      "learning_rate": 0.0009722531758413193,
+      "loss": 0.48,
+      "step": 250
+    },
+    {
+      "epoch": 0.20280811232449297,
+      "grad_norm": 2.648716449737549,
+      "learning_rate": 0.0009711388455538221,
+      "loss": 0.416,
+      "step": 260
+    },
+    {
+      "epoch": 0.21060842433697347,
+      "grad_norm": 1.5672308206558228,
+      "learning_rate": 0.0009700245152663249,
+      "loss": 0.4223,
+      "step": 270
+    },
+    {
+      "epoch": 0.21840873634945399,
+      "grad_norm": 2.618163585662842,
+      "learning_rate": 0.0009689101849788277,
+      "loss": 0.4172,
+      "step": 280
+    },
+    {
+      "epoch": 0.22620904836193448,
+      "grad_norm": 3.6365268230438232,
+      "learning_rate": 0.0009677958546913305,
+      "loss": 0.5501,
+      "step": 290
+    },
+    {
+      "epoch": 0.23400936037441497,
+      "grad_norm": 2.740039825439453,
+      "learning_rate": 0.0009666815244038332,
+      "loss": 0.3553,
+      "step": 300
+    },
+    {
+      "epoch": 0.24180967238689546,
+      "grad_norm": 3.406210422515869,
+      "learning_rate": 0.000965567194116336,
+      "loss": 0.3518,
+      "step": 310
+    },
+    {
+      "epoch": 0.24960998439937598,
+      "grad_norm": 1.4707075357437134,
+      "learning_rate": 0.000964452863828839,
+      "loss": 0.3452,
+      "step": 320
+    },
+    {
+      "epoch": 0.2574102964118565,
+      "grad_norm": 1.608324408531189,
+      "learning_rate": 0.0009633385335413417,
+      "loss": 0.4908,
+      "step": 330
+    },
+    {
+      "epoch": 0.26521060842433697,
+      "grad_norm": 4.090480327606201,
+      "learning_rate": 0.0009622242032538444,
+      "loss": 0.4597,
+      "step": 340
+    },
+    {
+      "epoch": 0.27301092043681746,
+      "grad_norm": 2.2214395999908447,
+      "learning_rate": 0.0009611098729663472,
+      "loss": 0.4552,
+      "step": 350
+    },
+    {
+      "epoch": 0.28081123244929795,
+      "grad_norm": 1.9134166240692139,
+      "learning_rate": 0.00095999554267885,
+      "loss": 0.3571,
+      "step": 360
+    },
+    {
+      "epoch": 0.28861154446177845,
+      "grad_norm": 1.8127851486206055,
+      "learning_rate": 0.0009588812123913528,
+      "loss": 0.3808,
+      "step": 370
+    },
+    {
+      "epoch": 0.296411856474259,
+      "grad_norm": 2.2262885570526123,
+      "learning_rate": 0.0009577668821038556,
+      "loss": 0.4099,
+      "step": 380
+    },
+    {
+      "epoch": 0.3042121684867395,
+      "grad_norm": 2.8041303157806396,
+      "learning_rate": 0.0009566525518163583,
+      "loss": 0.3988,
+      "step": 390
+    },
+    {
+      "epoch": 0.31201248049922,
+      "grad_norm": 6.797432899475098,
+      "learning_rate": 0.0009555382215288611,
+      "loss": 0.4728,
+      "step": 400
+    },
+    {
+      "epoch": 0.31981279251170047,
+      "grad_norm": 3.1861369609832764,
+      "learning_rate": 0.000954423891241364,
+      "loss": 0.3502,
+      "step": 410
+    },
+    {
+      "epoch": 0.32761310452418096,
+      "grad_norm": 2.9223642349243164,
+      "learning_rate": 0.0009533095609538667,
+      "loss": 0.4215,
+      "step": 420
+    },
+    {
+      "epoch": 0.33541341653666146,
+      "grad_norm": 6.848895072937012,
+      "learning_rate": 0.0009521952306663694,
+      "loss": 0.415,
+      "step": 430
+    },
+    {
+      "epoch": 0.34321372854914195,
+      "grad_norm": 9.054282188415527,
+      "learning_rate": 0.0009510809003788722,
+      "loss": 0.4667,
+      "step": 440
+    },
+    {
+      "epoch": 0.35101404056162244,
+      "grad_norm": 2.3005900382995605,
+      "learning_rate": 0.0009499665700913752,
+      "loss": 0.6565,
+      "step": 450
+    },
+    {
+      "epoch": 0.358814352574103,
+      "grad_norm": 2.9467573165893555,
+      "learning_rate": 0.000948852239803878,
+      "loss": 0.3837,
+      "step": 460
+    },
+    {
+      "epoch": 0.3666146645865835,
+      "grad_norm": 1.5977652072906494,
+      "learning_rate": 0.0009477379095163808,
+      "loss": 0.407,
+      "step": 470
+    },
+    {
+      "epoch": 0.374414976599064,
+      "grad_norm": 2.8274600505828857,
+      "learning_rate": 0.0009466235792288835,
+      "loss": 0.4501,
+      "step": 480
+    },
+    {
+      "epoch": 0.38221528861154447,
+      "grad_norm": 1.8566502332687378,
+      "learning_rate": 0.0009455092489413863,
+      "loss": 0.3146,
+      "step": 490
+    },
+    {
+      "epoch": 0.39001560062402496,
+      "grad_norm": 2.5871951580047607,
+      "learning_rate": 0.0009443949186538891,
+      "loss": 0.3335,
+      "step": 500
+    },
+    {
+      "epoch": 0.39781591263650545,
+      "grad_norm": 2.0552711486816406,
+      "learning_rate": 0.0009432805883663919,
+      "loss": 0.3698,
+      "step": 510
+    },
+    {
+      "epoch": 0.40561622464898595,
+      "grad_norm": 1.5244548320770264,
+      "learning_rate": 0.0009421662580788946,
+      "loss": 0.581,
+      "step": 520
+    },
+    {
+      "epoch": 0.41341653666146644,
+      "grad_norm": 1.5146633386611938,
+      "learning_rate": 0.0009410519277913974,
+      "loss": 0.3634,
+      "step": 530
+    },
+    {
+      "epoch": 0.42121684867394693,
+      "grad_norm": 2.394819736480713,
+      "learning_rate": 0.0009399375975039002,
+      "loss": 0.338,
+      "step": 540
+    },
+    {
+      "epoch": 0.4290171606864275,
+      "grad_norm": 2.3500325679779053,
+      "learning_rate": 0.000938823267216403,
+      "loss": 0.4025,
+      "step": 550
+    },
+    {
+      "epoch": 0.43681747269890797,
+      "grad_norm": 2.4186370372772217,
+      "learning_rate": 0.0009377089369289058,
+      "loss": 0.4095,
+      "step": 560
+    },
+    {
+      "epoch": 0.44461778471138846,
+      "grad_norm": 1.9770065546035767,
+      "learning_rate": 0.0009365946066414085,
+      "loss": 0.4275,
+      "step": 570
+    },
+    {
+      "epoch": 0.45241809672386896,
+      "grad_norm": 1.7679632902145386,
+      "learning_rate": 0.0009354802763539114,
+      "loss": 0.3949,
+      "step": 580
+    },
+    {
+      "epoch": 0.46021840873634945,
+      "grad_norm": 2.0794620513916016,
+      "learning_rate": 0.0009343659460664142,
+      "loss": 0.4208,
+      "step": 590
+    },
+    {
+      "epoch": 0.46801872074882994,
+      "grad_norm": 2.677424192428589,
+      "learning_rate": 0.000933251615778917,
+      "loss": 0.4089,
+      "step": 600
+    },
+    {
+      "epoch": 0.47581903276131043,
+      "grad_norm": 1.526112675666809,
+      "learning_rate": 0.0009321372854914197,
+      "loss": 0.4196,
+      "step": 610
+    },
+    {
+      "epoch": 0.4836193447737909,
+      "grad_norm": 1.8656370639801025,
+      "learning_rate": 0.0009310229552039225,
+      "loss": 0.3732,
+      "step": 620
+    },
+    {
+      "epoch": 0.4914196567862715,
+      "grad_norm": 3.3338847160339355,
+      "learning_rate": 0.0009299086249164253,
+      "loss": 0.4231,
+      "step": 630
+    },
+    {
+      "epoch": 0.49921996879875197,
+      "grad_norm": 2.1057350635528564,
+      "learning_rate": 0.0009287942946289281,
+      "loss": 0.3921,
+      "step": 640
+    },
+    {
+      "epoch": 0.5070202808112324,
+      "grad_norm": 1.544977068901062,
+      "learning_rate": 0.0009276799643414309,
+      "loss": 0.3748,
+      "step": 650
+    },
+    {
+      "epoch": 0.514820592823713,
+      "grad_norm": 3.4070258140563965,
+      "learning_rate": 0.0009265656340539336,
+      "loss": 0.4027,
+      "step": 660
+    },
+    {
+      "epoch": 0.5226209048361935,
+      "grad_norm": 5.5486931800842285,
+      "learning_rate": 0.0009254513037664364,
+      "loss": 0.4326,
+      "step": 670
+    },
+    {
+      "epoch": 0.5304212168486739,
+      "grad_norm": 3.6824769973754883,
+      "learning_rate": 0.0009243369734789392,
+      "loss": 0.4809,
+      "step": 680
+    },
+    {
+      "epoch": 0.5382215288611545,
+      "grad_norm": 6.3154778480529785,
+      "learning_rate": 0.000923222643191442,
+      "loss": 0.3588,
+      "step": 690
+    },
+    {
+      "epoch": 0.5460218408736349,
+      "grad_norm": 3.133465528488159,
+      "learning_rate": 0.0009221083129039447,
+      "loss": 0.4712,
+      "step": 700
+    },
+    {
+      "epoch": 0.5538221528861155,
+      "grad_norm": 4.222598552703857,
+      "learning_rate": 0.0009209939826164475,
+      "loss": 0.4135,
+      "step": 710
+    },
+    {
+      "epoch": 0.5616224648985959,
+      "grad_norm": 4.6125078201293945,
+      "learning_rate": 0.0009198796523289504,
+      "loss": 0.393,
+      "step": 720
+    },
+    {
+      "epoch": 0.5694227769110765,
+      "grad_norm": 6.543318748474121,
+      "learning_rate": 0.0009187653220414532,
+      "loss": 0.6138,
+      "step": 730
+    },
+    {
+      "epoch": 0.5772230889235569,
+      "grad_norm": 2.596463680267334,
+      "learning_rate": 0.000917650991753956,
+      "loss": 0.3694,
+      "step": 740
+    },
+    {
+      "epoch": 0.5850234009360374,
+      "grad_norm": 2.428490161895752,
+      "learning_rate": 0.0009165366614664587,
+      "loss": 0.4535,
+      "step": 750
+    },
+    {
+      "epoch": 0.592823712948518,
+      "grad_norm": 1.8790688514709473,
+      "learning_rate": 0.0009154223311789615,
+      "loss": 0.3986,
+      "step": 760
+    },
+    {
+      "epoch": 0.6006240249609984,
+      "grad_norm": 3.141587734222412,
+      "learning_rate": 0.0009143080008914643,
+      "loss": 0.3881,
+      "step": 770
+    },
+    {
+      "epoch": 0.608424336973479,
+      "grad_norm": 2.125810146331787,
+      "learning_rate": 0.0009131936706039671,
+      "loss": 0.4225,
+      "step": 780
+    },
+    {
+      "epoch": 0.6162246489859594,
+      "grad_norm": 2.532404661178589,
+      "learning_rate": 0.0009120793403164698,
+      "loss": 0.3423,
+      "step": 790
+    },
+    {
+      "epoch": 0.62402496099844,
+      "grad_norm": 3.6324350833892822,
+      "learning_rate": 0.0009109650100289726,
+      "loss": 0.4205,
+      "step": 800
+    },
+    {
+      "epoch": 0.6318252730109204,
+      "grad_norm": 1.4804415702819824,
+      "learning_rate": 0.0009098506797414754,
+      "loss": 0.4042,
+      "step": 810
+    },
+    {
+      "epoch": 0.6396255850234009,
+      "grad_norm": 1.5140562057495117,
+      "learning_rate": 0.0009087363494539782,
+      "loss": 0.3056,
+      "step": 820
+    },
+    {
+      "epoch": 0.6474258970358814,
+      "grad_norm": 2.470576047897339,
+      "learning_rate": 0.000907622019166481,
+      "loss": 0.391,
+      "step": 830
+    },
+    {
+      "epoch": 0.6552262090483619,
+      "grad_norm": 3.4496209621429443,
+      "learning_rate": 0.0009065076888789837,
+      "loss": 0.4163,
+      "step": 840
+    },
+    {
+      "epoch": 0.6630265210608425,
+      "grad_norm": 1.8823250532150269,
+      "learning_rate": 0.0009053933585914866,
+      "loss": 0.3877,
+      "step": 850
+    },
+    {
+      "epoch": 0.6708268330733229,
+      "grad_norm": 2.492297410964966,
+      "learning_rate": 0.0009042790283039894,
+      "loss": 0.3542,
+      "step": 860
+    },
+    {
+      "epoch": 0.6786271450858035,
+      "grad_norm": 3.977569341659546,
+      "learning_rate": 0.0009031646980164922,
+      "loss": 0.4168,
+      "step": 870
+    },
+    {
+      "epoch": 0.6864274570982839,
+      "grad_norm": 3.938462495803833,
+      "learning_rate": 0.0009020503677289949,
+      "loss": 0.4894,
+      "step": 880
+    },
+    {
+      "epoch": 0.6942277691107644,
+      "grad_norm": 1.3457701206207275,
+      "learning_rate": 0.0009009360374414977,
+      "loss": 0.4903,
+      "step": 890
+    },
+    {
+      "epoch": 0.7020280811232449,
+      "grad_norm": 6.3473124504089355,
+      "learning_rate": 0.0008998217071540005,
+      "loss": 0.4766,
+      "step": 900
+    },
+    {
+      "epoch": 0.7098283931357254,
+      "grad_norm": 3.145792245864868,
+      "learning_rate": 0.0008987073768665033,
+      "loss": 0.4119,
+      "step": 910
+    },
+    {
+      "epoch": 0.717628705148206,
+      "grad_norm": 1.809446930885315,
+      "learning_rate": 0.0008975930465790061,
+      "loss": 0.3991,
+      "step": 920
+    },
+    {
+      "epoch": 0.7254290171606864,
+      "grad_norm": 1.7960044145584106,
+      "learning_rate": 0.0008964787162915088,
+      "loss": 0.3095,
+      "step": 930
+    },
+    {
+      "epoch": 0.733229329173167,
+      "grad_norm": 2.9710285663604736,
+      "learning_rate": 0.0008953643860040116,
+      "loss": 0.5104,
+      "step": 940
+    },
+    {
+      "epoch": 0.7410296411856474,
+      "grad_norm": 2.460524797439575,
+      "learning_rate": 0.0008942500557165144,
+      "loss": 0.4332,
+      "step": 950
+    },
+    {
+      "epoch": 0.748829953198128,
+      "grad_norm": 1.6166704893112183,
+      "learning_rate": 0.0008931357254290172,
+      "loss": 0.3856,
+      "step": 960
+    },
+    {
+      "epoch": 0.7566302652106084,
+      "grad_norm": 1.747750163078308,
+      "learning_rate": 0.0008920213951415199,
+      "loss": 0.3973,
+      "step": 970
+    },
+    {
+      "epoch": 0.7644305772230889,
+      "grad_norm": 1.4469414949417114,
+      "learning_rate": 0.0008909070648540227,
+      "loss": 0.3344,
+      "step": 980
+    },
+    {
+      "epoch": 0.7722308892355694,
+      "grad_norm": 4.661273956298828,
+      "learning_rate": 0.0008897927345665256,
+      "loss": 0.6003,
+      "step": 990
+    },
+    {
+      "epoch": 0.7800312012480499,
+      "grad_norm": 3.6588950157165527,
+      "learning_rate": 0.0008886784042790284,
+      "loss": 0.473,
+      "step": 1000
+    },
+    {
+      "epoch": 0.7878315132605305,
+      "grad_norm": 2.208383798599243,
+      "learning_rate": 0.0008876755070202809,
+      "loss": 0.3899,
+      "step": 1010
+    },
+    {
+      "epoch": 0.7956318252730109,
+      "grad_norm": 2.3569576740264893,
+      "learning_rate": 0.0008865611767327836,
+      "loss": 0.3871,
+      "step": 1020
+    },
+    {
+      "epoch": 0.8034321372854915,
+      "grad_norm": 2.7071454524993896,
+      "learning_rate": 0.0008854468464452864,
+      "loss": 0.3149,
+      "step": 1030
+    },
+    {
+      "epoch": 0.8112324492979719,
+      "grad_norm": 2.8024532794952393,
+      "learning_rate": 0.0008843325161577892,
+      "loss": 0.3328,
+      "step": 1040
+    },
+    {
+      "epoch": 0.8190327613104524,
+      "grad_norm": 3.0969290733337402,
+      "learning_rate": 0.000883218185870292,
+      "loss": 0.3948,
+      "step": 1050
+    },
+    {
+      "epoch": 0.8268330733229329,
+      "grad_norm": 2.982484817504883,
+      "learning_rate": 0.0008821038555827947,
+      "loss": 0.4323,
+      "step": 1060
+    },
+    {
+      "epoch": 0.8346333853354134,
+      "grad_norm": 3.133814573287964,
+      "learning_rate": 0.0008809895252952975,
+      "loss": 0.4393,
+      "step": 1070
+    },
+    {
+      "epoch": 0.8424336973478939,
+      "grad_norm": 3.3123364448547363,
+      "learning_rate": 0.0008798751950078003,
+      "loss": 0.3244,
+      "step": 1080
+    },
+    {
+      "epoch": 0.8502340093603744,
+      "grad_norm": 2.308555841445923,
+      "learning_rate": 0.0008787608647203032,
+      "loss": 0.424,
+      "step": 1090
+    },
+    {
+      "epoch": 0.858034321372855,
+      "grad_norm": 3.654137134552002,
+      "learning_rate": 0.000877646534432806,
+      "loss": 0.3445,
+      "step": 1100
+    },
+    {
+      "epoch": 0.8658346333853354,
+      "grad_norm": 2.149843692779541,
+      "learning_rate": 0.0008765322041453087,
+      "loss": 0.3398,
+      "step": 1110
+    },
+    {
+      "epoch": 0.8736349453978159,
+      "grad_norm": 3.1334431171417236,
+      "learning_rate": 0.0008754178738578115,
+      "loss": 0.3333,
+      "step": 1120
+    },
+    {
+      "epoch": 0.8814352574102964,
+      "grad_norm": 2.2942090034484863,
+      "learning_rate": 0.0008743035435703143,
+      "loss": 0.4188,
+      "step": 1130
+    },
+    {
+      "epoch": 0.8892355694227769,
+      "grad_norm": 2.0195343494415283,
+      "learning_rate": 0.0008731892132828171,
+      "loss": 0.4047,
+      "step": 1140
+    },
+    {
+      "epoch": 0.8970358814352574,
+      "grad_norm": 2.3850839138031006,
+      "learning_rate": 0.0008720748829953198,
+      "loss": 0.3931,
+      "step": 1150
+    },
+    {
+      "epoch": 0.9048361934477379,
+      "grad_norm": 1.6200228929519653,
+      "learning_rate": 0.0008709605527078226,
+      "loss": 0.409,
+      "step": 1160
+    },
+    {
+      "epoch": 0.9126365054602185,
+      "grad_norm": 2.9001989364624023,
+      "learning_rate": 0.0008698462224203254,
+      "loss": 0.4289,
+      "step": 1170
+    },
+    {
+      "epoch": 0.9204368174726989,
+      "grad_norm": 1.52889883518219,
+      "learning_rate": 0.0008687318921328282,
+      "loss": 0.3575,
+      "step": 1180
+    },
+    {
+      "epoch": 0.9282371294851794,
+      "grad_norm": 2.359733819961548,
+      "learning_rate": 0.000867617561845331,
+      "loss": 0.3837,
+      "step": 1190
+    },
+    {
+      "epoch": 0.9360374414976599,
+      "grad_norm": 2.3807597160339355,
+      "learning_rate": 0.0008665032315578337,
+      "loss": 0.4206,
+      "step": 1200
+    },
+    {
+      "epoch": 0.9438377535101404,
+      "grad_norm": 1.8366179466247559,
+      "learning_rate": 0.0008653889012703365,
+      "loss": 0.3101,
+      "step": 1210
+    },
+    {
+      "epoch": 0.9516380655226209,
+      "grad_norm": 3.1048014163970947,
+      "learning_rate": 0.0008642745709828393,
+      "loss": 0.4104,
+      "step": 1220
+    },
+    {
+      "epoch": 0.9594383775351014,
+      "grad_norm": 1.5314342975616455,
+      "learning_rate": 0.0008631602406953422,
+      "loss": 0.3539,
+      "step": 1230
+    },
+    {
+      "epoch": 0.9672386895475819,
+      "grad_norm": 2.8501791954040527,
+      "learning_rate": 0.0008620459104078449,
+      "loss": 0.4104,
+      "step": 1240
+    },
+    {
+      "epoch": 0.9750390015600624,
+      "grad_norm": 5.708191394805908,
+      "learning_rate": 0.0008609315801203477,
+      "loss": 0.4038,
+      "step": 1250
+    },
+    {
+      "epoch": 0.982839313572543,
+      "grad_norm": 2.173867702484131,
+      "learning_rate": 0.0008598172498328505,
+      "loss": 0.338,
+      "step": 1260
+    },
+    {
+      "epoch": 0.9906396255850234,
+      "grad_norm": 2.7057418823242188,
+      "learning_rate": 0.0008587029195453533,
+      "loss": 0.3941,
+      "step": 1270
+    },
+    {
+      "epoch": 0.9984399375975039,
+      "grad_norm": 1.3492989540100098,
+      "learning_rate": 0.0008575885892578561,
+      "loss": 0.3696,
+      "step": 1280
+    },
+    {
+      "epoch": 1.0,
+      "eval_loss": 0.4044143855571747,
+      "eval_runtime": 635.8286,
+      "eval_samples_per_second": 0.865,
+      "eval_steps_per_second": 0.865,
+      "eval_wer": 26.665379416875844,
+      "step": 1282
+    },
+    {
+      "epoch": 1.0062402496099845,
+      "grad_norm": 2.5727291107177734,
+      "learning_rate": 0.0008564742589703588,
+      "loss": 0.3369,
+      "step": 1290
+    },
+    {
+      "epoch": 1.0140405616224648,
+      "grad_norm": 2.681490182876587,
+      "learning_rate": 0.0008553599286828616,
+      "loss": 0.2434,
+      "step": 1300
+    },
+    {
+      "epoch": 1.0218408736349454,
+      "grad_norm": 2.0296504497528076,
+      "learning_rate": 0.0008542455983953644,
+      "loss": 0.2122,
+      "step": 1310
+    },
+    {
+      "epoch": 1.029641185647426,
+      "grad_norm": 2.277512311935425,
+      "learning_rate": 0.0008531312681078672,
+      "loss": 0.3932,
+      "step": 1320
+    },
+    {
+      "epoch": 1.0374414976599065,
+      "grad_norm": 4.77215576171875,
+      "learning_rate": 0.0008520169378203699,
+      "loss": 0.2658,
+      "step": 1330
+    },
+    {
+      "epoch": 1.045241809672387,
+      "grad_norm": 1.4027091264724731,
+      "learning_rate": 0.0008509026075328727,
+      "loss": 0.3524,
+      "step": 1340
+    },
+    {
+      "epoch": 1.0530421216848673,
+      "grad_norm": 2.2849514484405518,
+      "learning_rate": 0.0008497882772453755,
+      "loss": 0.2705,
+      "step": 1350
+    },
+    {
+      "epoch": 1.0608424336973479,
+      "grad_norm": 1.2896777391433716,
+      "learning_rate": 0.0008486739469578784,
+      "loss": 0.264,
+      "step": 1360
+    },
+    {
+      "epoch": 1.0686427457098284,
+      "grad_norm": 2.4552128314971924,
+      "learning_rate": 0.0008475596166703812,
+      "loss": 0.2833,
+      "step": 1370
+    },
+    {
+      "epoch": 1.076443057722309,
+      "grad_norm": 2.044693946838379,
+      "learning_rate": 0.0008464452863828839,
+      "loss": 0.2116,
+      "step": 1380
+    },
+    {
+      "epoch": 1.0842433697347893,
+      "grad_norm": 1.3727463483810425,
+      "learning_rate": 0.0008453309560953867,
+      "loss": 0.2533,
+      "step": 1390
+    },
+    {
+      "epoch": 1.0920436817472698,
+      "grad_norm": 1.6917822360992432,
+      "learning_rate": 0.0008442166258078895,
+      "loss": 0.4259,
+      "step": 1400
+    },
+    {
+      "epoch": 1.0998439937597504,
+      "grad_norm": 2.198549747467041,
+      "learning_rate": 0.0008431022955203923,
+      "loss": 0.3161,
+      "step": 1410
+    },
+    {
+      "epoch": 1.107644305772231,
+      "grad_norm": 1.7467869520187378,
+      "learning_rate": 0.000841987965232895,
+      "loss": 0.2377,
+      "step": 1420
+    },
+    {
+      "epoch": 1.1154446177847115,
+      "grad_norm": 2.5347695350646973,
+      "learning_rate": 0.0008408736349453978,
+      "loss": 0.3258,
+      "step": 1430
+    },
+    {
+      "epoch": 1.1232449297971918,
+      "grad_norm": 1.9081774950027466,
+      "learning_rate": 0.0008397593046579006,
+      "loss": 0.2462,
+      "step": 1440
+    },
+    {
+      "epoch": 1.1310452418096724,
+      "grad_norm": 1.5889848470687866,
+      "learning_rate": 0.0008386449743704034,
+      "loss": 0.2404,
+      "step": 1450
+    },
+    {
+      "epoch": 1.138845553822153,
+      "grad_norm": 1.8944768905639648,
+      "learning_rate": 0.0008375306440829062,
+      "loss": 0.2707,
+      "step": 1460
+    },
+    {
+      "epoch": 1.1466458658346335,
+      "grad_norm": 2.5448453426361084,
+      "learning_rate": 0.0008364163137954089,
+      "loss": 0.3342,
+      "step": 1470
+    },
+    {
+      "epoch": 1.154446177847114,
+      "grad_norm": 2.0936005115509033,
+      "learning_rate": 0.0008353019835079117,
+      "loss": 0.385,
+      "step": 1480
+    },
+    {
+      "epoch": 1.1622464898595943,
+      "grad_norm": 2.614129066467285,
+      "learning_rate": 0.0008341876532204145,
+      "loss": 0.2817,
+      "step": 1490
+    },
+    {
+      "epoch": 1.1700468018720749,
+      "grad_norm": 1.6156001091003418,
+      "learning_rate": 0.0008330733229329174,
+      "loss": 0.3527,
+      "step": 1500
+    },
+    {
+      "epoch": 1.1778471138845554,
+      "grad_norm": 1.4294220209121704,
+      "learning_rate": 0.0008319589926454201,
+      "loss": 0.2469,
+      "step": 1510
+    },
+    {
+      "epoch": 1.185647425897036,
+      "grad_norm": 3.197176456451416,
+      "learning_rate": 0.0008308446623579229,
+      "loss": 0.2753,
+      "step": 1520
+    },
+    {
+      "epoch": 1.1934477379095163,
+      "grad_norm": 2.1629223823547363,
+      "learning_rate": 0.0008297303320704257,
+      "loss": 0.3951,
+      "step": 1530
+    },
+    {
+      "epoch": 1.2012480499219969,
+      "grad_norm": 2.9824419021606445,
+      "learning_rate": 0.0008286160017829285,
+      "loss": 0.3278,
+      "step": 1540
+    },
+    {
+      "epoch": 1.2090483619344774,
+      "grad_norm": 2.866138219833374,
+      "learning_rate": 0.0008275016714954313,
+      "loss": 0.4267,
+      "step": 1550
+    },
+    {
+      "epoch": 1.216848673946958,
+      "grad_norm": 2.36781644821167,
+      "learning_rate": 0.000826387341207934,
+      "loss": 0.3414,
+      "step": 1560
+    },
+    {
+      "epoch": 1.2246489859594383,
+      "grad_norm": 1.8305447101593018,
+      "learning_rate": 0.0008252730109204368,
+      "loss": 0.2925,
+      "step": 1570
+    },
+    {
+      "epoch": 1.2324492979719188,
+      "grad_norm": 1.9879776239395142,
+      "learning_rate": 0.0008241586806329396,
+      "loss": 0.357,
+      "step": 1580
+    },
+    {
+      "epoch": 1.2402496099843994,
+      "grad_norm": 2.183350086212158,
+      "learning_rate": 0.0008230443503454424,
+      "loss": 0.3409,
+      "step": 1590
+    },
+    {
+      "epoch": 1.24804992199688,
+      "grad_norm": 2.197072744369507,
+      "learning_rate": 0.0008219300200579451,
+      "loss": 0.33,
+      "step": 1600
+    },
+    {
+      "epoch": 1.2558502340093605,
+      "grad_norm": 3.2065696716308594,
+      "learning_rate": 0.0008208156897704479,
+      "loss": 0.2853,
+      "step": 1610
+    },
+    {
+      "epoch": 1.2636505460218408,
+      "grad_norm": 2.0581350326538086,
+      "learning_rate": 0.0008197013594829507,
+      "loss": 0.3647,
+      "step": 1620
+    },
+    {
+      "epoch": 1.2714508580343213,
+      "grad_norm": 3.149153232574463,
+      "learning_rate": 0.0008185870291954536,
+      "loss": 0.3921,
+      "step": 1630
+    },
+    {
+      "epoch": 1.2792511700468019,
+      "grad_norm": 2.5097105503082275,
+      "learning_rate": 0.0008174726989079563,
+      "loss": 0.3302,
+      "step": 1640
+    },
+    {
+      "epoch": 1.2870514820592824,
+      "grad_norm": 2.7537474632263184,
+      "learning_rate": 0.0008163583686204591,
+      "loss": 0.3286,
+      "step": 1650
+    },
+    {
+      "epoch": 1.294851794071763,
+      "grad_norm": 1.966965675354004,
+      "learning_rate": 0.0008152440383329619,
+      "loss": 0.2785,
+      "step": 1660
+    },
+    {
+      "epoch": 1.3026521060842433,
+      "grad_norm": 1.9159988164901733,
+      "learning_rate": 0.0008141297080454647,
+      "loss": 0.3299,
+      "step": 1670
+    },
+    {
+      "epoch": 1.3104524180967239,
+      "grad_norm": 2.2212252616882324,
+      "learning_rate": 0.0008130153777579675,
+      "loss": 0.2807,
+      "step": 1680
+    },
+    {
+      "epoch": 1.3182527301092044,
+      "grad_norm": 4.194318771362305,
+      "learning_rate": 0.0008119010474704702,
+      "loss": 0.3161,
+      "step": 1690
+    },
+    {
+      "epoch": 1.3260530421216847,
+      "grad_norm": 1.7189604043960571,
+      "learning_rate": 0.000810786717182973,
+      "loss": 0.3356,
+      "step": 1700
+    },
+    {
+      "epoch": 1.3338533541341655,
+      "grad_norm": 1.5196418762207031,
+      "learning_rate": 0.0008096723868954758,
+      "loss": 0.2616,
+      "step": 1710
+    },
+    {
+      "epoch": 1.3416536661466458,
+      "grad_norm": 1.497450351715088,
+      "learning_rate": 0.0008085580566079786,
+      "loss": 0.2946,
+      "step": 1720
+    },
+    {
+      "epoch": 1.3494539781591264,
+      "grad_norm": 1.74885892868042,
+      "learning_rate": 0.0008074437263204813,
+      "loss": 0.2801,
+      "step": 1730
+    },
+    {
+      "epoch": 1.357254290171607,
+      "grad_norm": 2.040701389312744,
+      "learning_rate": 0.0008063293960329841,
+      "loss": 0.3203,
+      "step": 1740
+    },
+    {
+      "epoch": 1.3650546021840873,
+      "grad_norm": 3.760457754135132,
+      "learning_rate": 0.0008052150657454869,
+      "loss": 0.4569,
+      "step": 1750
+    },
+    {
+      "epoch": 1.3728549141965678,
+      "grad_norm": 2.92971134185791,
+      "learning_rate": 0.0008041007354579897,
+      "loss": 0.3321,
+      "step": 1760
+    },
+    {
+      "epoch": 1.3806552262090483,
+      "grad_norm": 1.9461047649383545,
+      "learning_rate": 0.0008029864051704926,
+      "loss": 0.2696,
+      "step": 1770
+    },
+    {
+      "epoch": 1.388455538221529,
+      "grad_norm": 3.2626147270202637,
+      "learning_rate": 0.0008018720748829953,
+      "loss": 0.3322,
+      "step": 1780
+    },
+    {
+      "epoch": 1.3962558502340094,
+      "grad_norm": 2.1270642280578613,
+      "learning_rate": 0.0008007577445954981,
+      "loss": 0.2965,
+      "step": 1790
+    },
+    {
+      "epoch": 1.4040561622464898,
+      "grad_norm": 2.3174221515655518,
+      "learning_rate": 0.0007996434143080009,
+      "loss": 0.3425,
+      "step": 1800
+    },
+    {
+      "epoch": 1.4118564742589703,
+      "grad_norm": 2.5749576091766357,
+      "learning_rate": 0.0007985290840205037,
+      "loss": 0.3622,
+      "step": 1810
+    },
+    {
+      "epoch": 1.4196567862714509,
+      "grad_norm": 1.873813509941101,
+      "learning_rate": 0.0007974147537330064,
+      "loss": 0.2497,
+      "step": 1820
+    },
+    {
+      "epoch": 1.4274570982839314,
+      "grad_norm": 3.633928060531616,
+      "learning_rate": 0.0007963004234455092,
+      "loss": 0.4058,
+      "step": 1830
+    },
+    {
+      "epoch": 1.435257410296412,
+      "grad_norm": 2.356269598007202,
+      "learning_rate": 0.000795186093158012,
+      "loss": 0.2635,
+      "step": 1840
+    },
+    {
+      "epoch": 1.4430577223088923,
+      "grad_norm": 1.9108752012252808,
+      "learning_rate": 0.0007940717628705148,
+      "loss": 0.336,
+      "step": 1850
+    },
+    {
+      "epoch": 1.4508580343213728,
+      "grad_norm": 1.5505330562591553,
+      "learning_rate": 0.0007929574325830176,
+      "loss": 0.3493,
+      "step": 1860
+    },
+    {
+      "epoch": 1.4586583463338534,
+      "grad_norm": 1.9970422983169556,
+      "learning_rate": 0.0007918431022955203,
+      "loss": 0.273,
+      "step": 1870
+    },
+    {
+      "epoch": 1.466458658346334,
+      "grad_norm": 2.753758192062378,
+      "learning_rate": 0.0007907287720080231,
+      "loss": 0.2845,
+      "step": 1880
+    },
+    {
+      "epoch": 1.4742589703588145,
+      "grad_norm": NaN,
+      "learning_rate": 0.0007897258747492757,
+      "loss": 0.2544,
+      "step": 1890
+    },
+    {
+      "epoch": 1.4820592823712948,
+      "grad_norm": 3.0995099544525146,
+      "learning_rate": 0.0007886115444617785,
+      "loss": 0.2884,
+      "step": 1900
+    },
+    {
+      "epoch": 1.4898595943837754,
+      "grad_norm": 5.728559970855713,
+      "learning_rate": 0.0007874972141742812,
+      "loss": 0.2681,
+      "step": 1910
+    },
+    {
+      "epoch": 1.497659906396256,
+      "grad_norm": 1.492622971534729,
+      "learning_rate": 0.000786382883886784,
+      "loss": 0.2891,
+      "step": 1920
+    },
+    {
+      "epoch": 1.5054602184087362,
+      "grad_norm": 1.7419252395629883,
+      "learning_rate": 0.0007852685535992868,
+      "loss": 0.4089,
+      "step": 1930
+    },
+    {
+      "epoch": 1.513260530421217,
+      "grad_norm": 6.814690589904785,
+      "learning_rate": 0.0007841542233117896,
+      "loss": 0.3372,
+      "step": 1940
+    },
+    {
+      "epoch": 1.5210608424336973,
+      "grad_norm": 2.380725860595703,
+      "learning_rate": 0.0007830398930242924,
+      "loss": 0.3189,
+      "step": 1950
+    },
+    {
+      "epoch": 1.5288611544461779,
+      "grad_norm": 5.004116058349609,
+      "learning_rate": 0.0007819255627367951,
+      "loss": 0.3018,
+      "step": 1960
+    },
+    {
+      "epoch": 1.5366614664586584,
+      "grad_norm": 2.604365825653076,
+      "learning_rate": 0.0007808112324492979,
+      "loss": 0.3054,
+      "step": 1970
+    },
+    {
+      "epoch": 1.5444617784711387,
+      "grad_norm": 1.585584044456482,
+      "learning_rate": 0.0007796969021618007,
+      "loss": 0.3477,
+      "step": 1980
+    },
+    {
+      "epoch": 1.5522620904836193,
+      "grad_norm": 1.8678693771362305,
+      "learning_rate": 0.0007785825718743035,
+      "loss": 0.3577,
+      "step": 1990
+    },
+    {
+      "epoch": 1.5600624024960998,
+      "grad_norm": 1.654689073562622,
+      "learning_rate": 0.0007774682415868062,
+      "loss": 0.2625,
+      "step": 2000
+    },
+    {
+      "epoch": 1.5678627145085804,
+      "grad_norm": 2.108919858932495,
+      "learning_rate": 0.0007763539112993092,
+      "loss": 0.2497,
+      "step": 2010
+    },
+    {
+      "epoch": 1.575663026521061,
+      "grad_norm": 7.198604106903076,
+      "learning_rate": 0.000775239581011812,
+      "loss": 0.3382,
+      "step": 2020
+    },
+    {
+      "epoch": 1.5834633385335413,
+      "grad_norm": 2.2285892963409424,
+      "learning_rate": 0.0007741252507243148,
+      "loss": 0.2598,
+      "step": 2030
+    },
+    {
+      "epoch": 1.5912636505460218,
+      "grad_norm": 1.7743014097213745,
+      "learning_rate": 0.0007730109204368176,
+      "loss": 0.2757,
+      "step": 2040
+    },
+    {
+      "epoch": 1.5990639625585024,
+      "grad_norm": 1.7763789892196655,
+      "learning_rate": 0.0007718965901493203,
+      "loss": 0.2703,
+      "step": 2050
+    },
+    {
+      "epoch": 1.6068642745709827,
+      "grad_norm": 2.159956693649292,
+      "learning_rate": 0.000770782259861823,
+      "loss": 0.2824,
+      "step": 2060
+    },
+    {
+      "epoch": 1.6146645865834635,
+      "grad_norm": 1.4845560789108276,
+      "learning_rate": 0.0007696679295743259,
+      "loss": 0.2528,
+      "step": 2070
+    },
+    {
+      "epoch": 1.6224648985959438,
+      "grad_norm": 3.627887010574341,
+      "learning_rate": 0.0007685535992868287,
+      "loss": 0.3197,
+      "step": 2080
+    },
+    {
+      "epoch": 1.6302652106084243,
+      "grad_norm": 2.2174973487854004,
+      "learning_rate": 0.0007674392689993314,
+      "loss": 0.2994,
+      "step": 2090
+    },
+    {
+      "epoch": 1.6380655226209049,
+      "grad_norm": 2.5977325439453125,
+      "learning_rate": 0.0007663249387118342,
+      "loss": 0.2991,
+      "step": 2100
+    },
+    {
+      "epoch": 1.6458658346333852,
+      "grad_norm": 1.9066824913024902,
+      "learning_rate": 0.000765210608424337,
+      "loss": 0.2166,
+      "step": 2110
+    },
+    {
+      "epoch": 1.653666146645866,
+      "grad_norm": 1.7197297811508179,
+      "learning_rate": 0.0007640962781368398,
+      "loss": 0.2948,
+      "step": 2120
+    },
+    {
+      "epoch": 1.6614664586583463,
+      "grad_norm": 2.054304361343384,
+      "learning_rate": 0.0007629819478493426,
+      "loss": 0.3073,
+      "step": 2130
+    },
+    {
+      "epoch": 1.6692667706708268,
+      "grad_norm": 1.7934963703155518,
+      "learning_rate": 0.0007618676175618453,
+      "loss": 0.2667,
+      "step": 2140
+    },
+    {
+      "epoch": 1.6770670826833074,
+      "grad_norm": 2.5259838104248047,
+      "learning_rate": 0.0007607532872743482,
+      "loss": 0.3322,
+      "step": 2150
+    },
+    {
+      "epoch": 1.6848673946957877,
+      "grad_norm": 3.6354122161865234,
+      "learning_rate": 0.000759638956986851,
+      "loss": 0.3909,
+      "step": 2160
+    },
+    {
+      "epoch": 1.6926677067082685,
+      "grad_norm": 1.6722809076309204,
+      "learning_rate": 0.0007585246266993538,
+      "loss": 0.2987,
+      "step": 2170
+    },
+    {
+      "epoch": 1.7004680187207488,
+      "grad_norm": 4.3235015869140625,
+      "learning_rate": 0.0007574102964118565,
+      "loss": 0.3112,
+      "step": 2180
+    },
+    {
+      "epoch": 1.7082683307332294,
+      "grad_norm": 2.236316442489624,
+      "learning_rate": 0.0007562959661243593,
+      "loss": 0.3174,
+      "step": 2190
+    },
+    {
+      "epoch": 1.71606864274571,
+      "grad_norm": 21.32891273498535,
+      "learning_rate": 0.0007551816358368621,
+      "loss": 0.3696,
+      "step": 2200
+    },
+    {
+      "epoch": 1.7238689547581902,
+      "grad_norm": 2.4251410961151123,
+      "learning_rate": 0.0007540673055493649,
+      "loss": 0.3171,
+      "step": 2210
+    },
+    {
+      "epoch": 1.7316692667706708,
+      "grad_norm": 2.4152424335479736,
+      "learning_rate": 0.0007529529752618677,
+      "loss": 0.328,
+      "step": 2220
+    },
+    {
+      "epoch": 1.7394695787831513,
+      "grad_norm": 2.0988574028015137,
+      "learning_rate": 0.0007518386449743704,
+      "loss": 0.3061,
+      "step": 2230
+    },
+    {
+      "epoch": 1.7472698907956317,
+      "grad_norm": 4.469291687011719,
+      "learning_rate": 0.0007507243146868732,
+      "loss": 0.3315,
+      "step": 2240
+    },
+    {
+      "epoch": 1.7550702028081124,
+      "grad_norm": 2.4917778968811035,
+      "learning_rate": 0.000749609984399376,
+      "loss": 0.4296,
+      "step": 2250
+    },
+    {
+      "epoch": 1.7628705148205928,
+      "grad_norm": 3.073840379714966,
+      "learning_rate": 0.0007484956541118788,
+      "loss": 0.3685,
+      "step": 2260
+    },
+    {
+      "epoch": 1.7706708268330733,
+      "grad_norm": 2.205733299255371,
+      "learning_rate": 0.0007473813238243815,
+      "loss": 0.2691,
+      "step": 2270
+    },
+    {
+      "epoch": 1.7784711388455539,
+      "grad_norm": 2.3948941230773926,
+      "learning_rate": 0.0007462669935368844,
+      "loss": 0.3065,
+      "step": 2280
+    },
+    {
+      "epoch": 1.7862714508580342,
+      "grad_norm": 2.6060824394226074,
+      "learning_rate": 0.0007451526632493872,
+      "loss": 0.3354,
+      "step": 2290
+    },
+    {
+      "epoch": 1.794071762870515,
+      "grad_norm": 3.2586774826049805,
+      "learning_rate": 0.00074403833296189,
+      "loss": 0.261,
+      "step": 2300
+    },
+    {
+      "epoch": 1.8018720748829953,
+      "grad_norm": 1.6417285203933716,
+      "learning_rate": 0.0007429240026743928,
+      "loss": 0.3828,
+      "step": 2310
+    },
+    {
+      "epoch": 1.8096723868954758,
+      "grad_norm": 4.006927967071533,
+      "learning_rate": 0.0007418096723868955,
+      "loss": 0.4228,
+      "step": 2320
+    },
+    {
+      "epoch": 1.8174726989079564,
+      "grad_norm": 2.5880374908447266,
+      "learning_rate": 0.0007406953420993983,
+      "loss": 0.2766,
+      "step": 2330
+    },
+    {
+      "epoch": 1.8252730109204367,
+      "grad_norm": 1.602337121963501,
+      "learning_rate": 0.0007395810118119011,
+      "loss": 0.2794,
+      "step": 2340
+    },
+    {
+      "epoch": 1.8330733229329175,
+      "grad_norm": 5.932153224945068,
+      "learning_rate": 0.0007384666815244039,
+      "loss": 0.3471,
+      "step": 2350
+    },
+    {
+      "epoch": 1.8408736349453978,
+      "grad_norm": 4.076808452606201,
+      "learning_rate": 0.0007373523512369066,
+      "loss": 0.2649,
+      "step": 2360
+    },
+    {
+      "epoch": 1.8486739469578783,
+      "grad_norm": 4.666397571563721,
+      "learning_rate": 0.0007362380209494094,
+      "loss": 0.2751,
+      "step": 2370
+    },
+    {
+      "epoch": 1.856474258970359,
+      "grad_norm": 3.792745590209961,
+      "learning_rate": 0.0007351236906619122,
+      "loss": 0.2606,
+      "step": 2380
+    },
+    {
+      "epoch": 1.8642745709828392,
+      "grad_norm": 2.5275423526763916,
+      "learning_rate": 0.000734009360374415,
+      "loss": 0.3469,
+      "step": 2390
+    },
+    {
+      "epoch": 1.8720748829953198,
+      "grad_norm": 1.59649658203125,
+      "learning_rate": 0.0007328950300869178,
+      "loss": 0.2724,
+      "step": 2400
+    },
+    {
+      "epoch": 1.8798751950078003,
+      "grad_norm": 3.7428267002105713,
+      "learning_rate": 0.0007317806997994206,
+      "loss": 0.4426,
+      "step": 2410
+    },
+    {
+      "epoch": 1.8876755070202809,
+      "grad_norm": 3.7439956665039062,
+      "learning_rate": 0.0007306663695119234,
+      "loss": 0.2901,
+      "step": 2420
+    },
+    {
+      "epoch": 1.8954758190327614,
+      "grad_norm": 2.3777198791503906,
+      "learning_rate": 0.0007295520392244262,
+      "loss": 0.2837,
+      "step": 2430
+    },
+    {
+      "epoch": 1.9032761310452417,
+      "grad_norm": 2.7654988765716553,
+      "learning_rate": 0.000728437708936929,
+      "loss": 0.277,
+      "step": 2440
+    },
+    {
+      "epoch": 1.9110764430577223,
+      "grad_norm": 1.8758680820465088,
+      "learning_rate": 0.0007273233786494317,
+      "loss": 0.2706,
+      "step": 2450
+    },
+    {
+      "epoch": 1.9188767550702028,
+      "grad_norm": 2.8725340366363525,
+      "learning_rate": 0.0007262090483619345,
+      "loss": 0.2566,
+      "step": 2460
+    },
+    {
+      "epoch": 1.9266770670826832,
+      "grad_norm": 2.4021458625793457,
+      "learning_rate": 0.0007250947180744373,
+      "loss": 0.2645,
+      "step": 2470
+    },
+    {
+      "epoch": 1.934477379095164,
+      "grad_norm": 2.8407838344573975,
+      "learning_rate": 0.0007239803877869401,
+      "loss": 0.3152,
+      "step": 2480
+    },
+    {
+      "epoch": 1.9422776911076443,
+      "grad_norm": 3.606403112411499,
+      "learning_rate": 0.0007228660574994429,
+      "loss": 0.373,
+      "step": 2490
+    },
+    {
+      "epoch": 1.9500780031201248,
+      "grad_norm": 2.362473487854004,
+      "learning_rate": 0.0007217517272119456,
+      "loss": 0.4332,
+      "step": 2500
+    },
+    {
+      "epoch": 1.9578783151326054,
+      "grad_norm": 1.9711815118789673,
+      "learning_rate": 0.0007206373969244484,
+      "loss": 0.2606,
+      "step": 2510
+    },
+    {
+      "epoch": 1.9656786271450857,
+      "grad_norm": 2.683908224105835,
+      "learning_rate": 0.0007195230666369512,
+      "loss": 0.3582,
+      "step": 2520
+    },
+    {
+      "epoch": 1.9734789391575664,
+      "grad_norm": 2.5902493000030518,
+      "learning_rate": 0.000718408736349454,
+      "loss": 0.2694,
+      "step": 2530
+    },
+    {
+      "epoch": 1.9812792511700468,
+      "grad_norm": 3.92708420753479,
+      "learning_rate": 0.0007172944060619567,
+      "loss": 0.3083,
+      "step": 2540
+    },
+    {
+      "epoch": 1.9890795631825273,
+      "grad_norm": 2.6788370609283447,
+      "learning_rate": 0.0007161800757744596,
+      "loss": 0.2587,
+      "step": 2550
+    },
+    {
+      "epoch": 1.9968798751950079,
+      "grad_norm": 2.413313627243042,
+      "learning_rate": 0.0007150657454869624,
+      "loss": 0.2732,
+      "step": 2560
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 0.3617618680000305,
+      "eval_runtime": 36952.264,
+      "eval_samples_per_second": 0.015,
+      "eval_steps_per_second": 0.015,
+      "eval_wer": 22.793975670978952,
+      "step": 2564
+    },
+    {
+      "epoch": 2.004680187207488,
+      "grad_norm": 1.572451114654541,
+      "learning_rate": 0.0007139514151994652,
+      "loss": 0.2147,
+      "step": 2570
+    },
+    {
+      "epoch": 2.012480499219969,
+      "grad_norm": 2.4759023189544678,
+      "learning_rate": 0.000712837084911968,
+      "loss": 0.2015,
+      "step": 2580
+    },
+    {
+      "epoch": 2.0202808112324493,
+      "grad_norm": 1.6903563737869263,
+      "learning_rate": 0.0007117227546244707,
+      "loss": 0.1703,
+      "step": 2590
+    },
+    {
+      "epoch": 2.0280811232449296,
+      "grad_norm": 3.494985580444336,
+      "learning_rate": 0.0007106084243369735,
+      "loss": 0.2063,
+      "step": 2600
+    },
+    {
+      "epoch": 2.0358814352574104,
+      "grad_norm": 1.5024439096450806,
+      "learning_rate": 0.0007094940940494763,
+      "loss": 0.1807,
+      "step": 2610
+    },
+    {
+      "epoch": 2.0436817472698907,
+      "grad_norm": 1.9423105716705322,
+      "learning_rate": 0.0007083797637619791,
+      "loss": 0.1953,
+      "step": 2620
+    },
+    {
+      "epoch": 2.0514820592823715,
+      "grad_norm": 0.8572360277175903,
+      "learning_rate": 0.0007072654334744818,
+      "loss": 0.3423,
+      "step": 2630
+    },
+    {
+      "epoch": 2.059282371294852,
+      "grad_norm": 2.574855327606201,
+      "learning_rate": 0.0007061511031869846,
+      "loss": 0.1916,
+      "step": 2640
+    },
+    {
+      "epoch": 2.067082683307332,
+      "grad_norm": 2.2941219806671143,
+      "learning_rate": 0.0007050367728994874,
+      "loss": 0.2168,
+      "step": 2650
+    },
+    {
+      "epoch": 2.074882995319813,
+      "grad_norm": 2.0448150634765625,
+      "learning_rate": 0.0007039224426119902,
+      "loss": 0.2412,
+      "step": 2660
+    },
+    {
+      "epoch": 2.0826833073322932,
+      "grad_norm": 1.8044381141662598,
+      "learning_rate": 0.000702808112324493,
+      "loss": 0.3261,
+      "step": 2670
+    },
+    {
+      "epoch": 2.090483619344774,
+      "grad_norm": 2.752220630645752,
+      "learning_rate": 0.0007016937820369958,
+      "loss": 0.2007,
+      "step": 2680
+    },
+    {
+      "epoch": 2.0982839313572543,
+      "grad_norm": 1.4598050117492676,
+      "learning_rate": 0.0007005794517494986,
+      "loss": 0.1856,
+      "step": 2690
+    },
+    {
+      "epoch": 2.1060842433697347,
+      "grad_norm": 3.578192710876465,
+      "learning_rate": 0.0006994651214620014,
+      "loss": 0.1991,
+      "step": 2700
+    },
+    {
+      "epoch": 2.1138845553822154,
+      "grad_norm": 2.6971054077148438,
+      "learning_rate": 0.0006983507911745042,
+      "loss": 0.2134,
+      "step": 2710
+    },
+    {
+      "epoch": 2.1216848673946958,
+      "grad_norm": 2.437596559524536,
+      "learning_rate": 0.0006972364608870069,
+      "loss": 0.228,
+      "step": 2720
+    },
+    {
+      "epoch": 2.129485179407176,
+      "grad_norm": 2.6254658699035645,
+      "learning_rate": 0.0006961221305995097,
+      "loss": 0.1671,
+      "step": 2730
+    },
+    {
+      "epoch": 2.137285491419657,
+      "grad_norm": 1.3765720129013062,
+      "learning_rate": 0.0006950078003120125,
+      "loss": 0.2391,
+      "step": 2740
+    },
+    {
+      "epoch": 2.145085803432137,
+      "grad_norm": 2.192396879196167,
+      "learning_rate": 0.0006938934700245153,
+      "loss": 0.2331,
+      "step": 2750
+    },
+    {
+      "epoch": 2.152886115444618,
+      "grad_norm": 1.4418809413909912,
+      "learning_rate": 0.0006927791397370181,
+      "loss": 0.2041,
+      "step": 2760
+    },
+    {
+      "epoch": 2.1606864274570983,
+      "grad_norm": 2.553459882736206,
+      "learning_rate": 0.0006916648094495208,
+      "loss": 0.2342,
+      "step": 2770
+    },
+    {
+      "epoch": 2.1684867394695786,
+      "grad_norm": 1.7199114561080933,
+      "learning_rate": 0.0006905504791620236,
+      "loss": 0.1841,
+      "step": 2780
+    },
+    {
+      "epoch": 2.1762870514820594,
+      "grad_norm": 2.2145979404449463,
+      "learning_rate": 0.0006894361488745264,
+      "loss": 0.1821,
+      "step": 2790
+    },
+    {
+      "epoch": 2.1840873634945397,
+      "grad_norm": 2.434779405593872,
+      "learning_rate": 0.0006883218185870292,
+      "loss": 0.2115,
+      "step": 2800
+    },
+    {
+      "epoch": 2.1918876755070205,
+      "grad_norm": 1.0518474578857422,
+      "learning_rate": 0.0006872074882995319,
+      "loss": 0.2186,
+      "step": 2810
+    },
+    {
+      "epoch": 2.199687987519501,
+      "grad_norm": 1.9856785535812378,
+      "learning_rate": 0.0006860931580120348,
+      "loss": 0.2448,
+      "step": 2820
+    },
+    {
+      "epoch": 2.207488299531981,
+      "grad_norm": 17.3148136138916,
+      "learning_rate": 0.0006849788277245376,
+      "loss": 0.2372,
+      "step": 2830
+    },
+    {
+      "epoch": 2.215288611544462,
+      "grad_norm": 3.4590959548950195,
+      "learning_rate": 0.0006838644974370404,
+      "loss": 0.3179,
+      "step": 2840
+    },
+    {
+      "epoch": 2.223088923556942,
+      "grad_norm": 2.127650260925293,
+      "learning_rate": 0.0006827501671495432,
+      "loss": 0.2445,
+      "step": 2850
+    },
+    {
+      "epoch": 2.230889235569423,
+      "grad_norm": 1.895729660987854,
+      "learning_rate": 0.0006816358368620459,
+      "loss": 0.1966,
+      "step": 2860
+    },
+    {
+      "epoch": 2.2386895475819033,
+      "grad_norm": 1.4693467617034912,
+      "learning_rate": 0.0006805215065745487,
+      "loss": 0.2351,
+      "step": 2870
+    },
+    {
+      "epoch": 2.2464898595943836,
+      "grad_norm": 1.500453233718872,
+      "learning_rate": 0.0006794071762870515,
+      "loss": 0.1946,
+      "step": 2880
+    },
+    {
+      "epoch": 2.2542901716068644,
+      "grad_norm": 2.8631374835968018,
+      "learning_rate": 0.0006782928459995543,
+      "loss": 0.2367,
+      "step": 2890
+    },
+    {
+      "epoch": 2.2620904836193447,
+      "grad_norm": 1.5281766653060913,
+      "learning_rate": 0.000677178515712057,
+      "loss": 0.1855,
+      "step": 2900
+    },
+    {
+      "epoch": 2.2698907956318255,
+      "grad_norm": 2.3027195930480957,
+      "learning_rate": 0.0006760641854245598,
+      "loss": 0.2117,
+      "step": 2910
+    },
+    {
+      "epoch": 2.277691107644306,
+      "grad_norm": 2.9962668418884277,
+      "learning_rate": 0.0006749498551370626,
+      "loss": 0.2441,
+      "step": 2920
+    },
+    {
+      "epoch": 2.285491419656786,
+      "grad_norm": 2.489192008972168,
+      "learning_rate": 0.0006738355248495654,
+      "loss": 0.2212,
+      "step": 2930
+    },
+    {
+      "epoch": 2.293291731669267,
+      "grad_norm": 1.366945743560791,
+      "learning_rate": 0.0006727211945620682,
+      "loss": 0.2003,
+      "step": 2940
+    },
+    {
+      "epoch": 2.3010920436817472,
+      "grad_norm": 1.4768056869506836,
+      "learning_rate": 0.000671606864274571,
+      "loss": 0.3107,
+      "step": 2950
+    },
+    {
+      "epoch": 2.308892355694228,
+      "grad_norm": 1.797979474067688,
+      "learning_rate": 0.0006704925339870738,
+      "loss": 0.1978,
+      "step": 2960
+    },
+    {
+      "epoch": 2.3166926677067083,
+      "grad_norm": 1.9439573287963867,
+      "learning_rate": 0.0006693782036995766,
+      "loss": 0.2306,
+      "step": 2970
+    },
+    {
+      "epoch": 2.3244929797191887,
+      "grad_norm": 1.582478642463684,
+      "learning_rate": 0.0006682638734120794,
+      "loss": 0.2079,
+      "step": 2980
+    },
+    {
+      "epoch": 2.3322932917316694,
+      "grad_norm": 2.1245157718658447,
+      "learning_rate": 0.0006671495431245821,
+      "loss": 0.2258,
+      "step": 2990
+    },
+    {
+      "epoch": 2.3400936037441498,
+      "grad_norm": 2.427675724029541,
+      "learning_rate": 0.0006660352128370849,
+      "loss": 0.2846,
+      "step": 3000
+    },
+    {
+      "epoch": 2.34789391575663,
+      "grad_norm": 0.9095527529716492,
+      "learning_rate": 0.0006649208825495877,
+      "loss": 0.1746,
+      "step": 3010
+    },
+    {
+      "epoch": 2.355694227769111,
+      "grad_norm": 2.0468294620513916,
+      "learning_rate": 0.0006638065522620905,
+      "loss": 0.2645,
+      "step": 3020
+    },
+    {
+      "epoch": 2.363494539781591,
+      "grad_norm": 1.5703836679458618,
+      "learning_rate": 0.0006626922219745933,
+      "loss": 0.1896,
+      "step": 3030
+    },
+    {
+      "epoch": 2.371294851794072,
+      "grad_norm": 2.37263822555542,
+      "learning_rate": 0.000661577891687096,
+      "loss": 0.232,
+      "step": 3040
+    },
+    {
+      "epoch": 2.3790951638065523,
+      "grad_norm": 1.6041433811187744,
+      "learning_rate": 0.0006604635613995988,
+      "loss": 0.2135,
+      "step": 3050
+    },
+    {
+      "epoch": 2.3868954758190326,
+      "grad_norm": 1.87883722782135,
+      "learning_rate": 0.0006593492311121016,
+      "loss": 0.1993,
+      "step": 3060
+    },
+    {
+      "epoch": 2.3946957878315134,
+      "grad_norm": 2.5502099990844727,
+      "learning_rate": 0.0006582349008246044,
+      "loss": 0.2468,
+      "step": 3070
+    },
+    {
+      "epoch": 2.4024960998439937,
+      "grad_norm": 2.681384801864624,
+      "learning_rate": 0.0006571205705371071,
+      "loss": 0.2364,
+      "step": 3080
+    },
+    {
+      "epoch": 2.410296411856474,
+      "grad_norm": 1.2032707929611206,
+      "learning_rate": 0.00065600624024961,
+      "loss": 0.2732,
+      "step": 3090
+    },
+    {
+      "epoch": 2.418096723868955,
+      "grad_norm": 1.553661584854126,
+      "learning_rate": 0.0006548919099621128,
+      "loss": 0.2214,
+      "step": 3100
+    },
+    {
+      "epoch": 2.425897035881435,
+      "grad_norm": 0.9736389517784119,
+      "learning_rate": 0.0006537775796746156,
+      "loss": 0.1778,
+      "step": 3110
+    },
+    {
+      "epoch": 2.433697347893916,
+      "grad_norm": 1.911352276802063,
+      "learning_rate": 0.0006526632493871184,
+      "loss": 0.2062,
+      "step": 3120
+    },
+    {
+      "epoch": 2.4414976599063962,
+      "grad_norm": 1.4338595867156982,
+      "learning_rate": 0.0006515489190996211,
+      "loss": 0.193,
+      "step": 3130
+    },
+    {
+      "epoch": 2.4492979719188765,
+      "grad_norm": 1.3027153015136719,
+      "learning_rate": 0.0006504345888121239,
+      "loss": 0.3175,
+      "step": 3140
+    },
+    {
+      "epoch": 2.4570982839313573,
+      "grad_norm": 2.262709140777588,
+      "learning_rate": 0.0006493202585246267,
+      "loss": 0.2874,
+      "step": 3150
+    },
+    {
+      "epoch": 2.4648985959438376,
+      "grad_norm": 1.36016845703125,
+      "learning_rate": 0.0006482059282371295,
+      "loss": 0.1603,
+      "step": 3160
+    },
+    {
+      "epoch": 2.4726989079563184,
+      "grad_norm": 3.441779613494873,
+      "learning_rate": 0.0006470915979496322,
+      "loss": 0.4564,
+      "step": 3170
+    },
+    {
+      "epoch": 2.4804992199687987,
+      "grad_norm": 1.8022549152374268,
+      "learning_rate": 0.000645977267662135,
+      "loss": 0.1928,
+      "step": 3180
+    },
+    {
+      "epoch": 2.488299531981279,
+      "grad_norm": 2.104497194290161,
+      "learning_rate": 0.0006448629373746378,
+      "loss": 0.2047,
+      "step": 3190
+    },
+    {
+      "epoch": 2.49609984399376,
+      "grad_norm": 2.5888454914093018,
+      "learning_rate": 0.0006437486070871406,
+      "loss": 0.2591,
+      "step": 3200
+    },
+    {
+      "epoch": 2.50390015600624,
+      "grad_norm": 1.3476183414459229,
+      "learning_rate": 0.0006426342767996434,
+      "loss": 0.2636,
+      "step": 3210
+    },
+    {
+      "epoch": 2.511700468018721,
+      "grad_norm": 2.888965606689453,
+      "learning_rate": 0.0006415199465121462,
+      "loss": 0.216,
+      "step": 3220
+    },
+    {
+      "epoch": 2.5195007800312013,
+      "grad_norm": 1.8190357685089111,
+      "learning_rate": 0.000640405616224649,
+      "loss": 0.3498,
+      "step": 3230
+    },
+    {
+      "epoch": 2.5273010920436816,
+      "grad_norm": 1.5132287740707397,
+      "learning_rate": 0.0006392912859371518,
+      "loss": 0.227,
+      "step": 3240
+    },
+    {
+      "epoch": 2.5351014040561624,
+      "grad_norm": 2.8061325550079346,
+      "learning_rate": 0.0006381769556496546,
+      "loss": 0.2089,
+      "step": 3250
+    },
+    {
+      "epoch": 2.5429017160686427,
+      "grad_norm": 2.2521767616271973,
+      "learning_rate": 0.0006370626253621573,
+      "loss": 0.2362,
+      "step": 3260
+    },
+    {
+      "epoch": 2.5507020280811235,
+      "grad_norm": 1.4635623693466187,
+      "learning_rate": 0.0006359482950746601,
+      "loss": 0.1901,
+      "step": 3270
+    },
+    {
+      "epoch": 2.5585023400936038,
+      "grad_norm": 1.9882755279541016,
+      "learning_rate": 0.0006348339647871629,
+      "loss": 0.2387,
+      "step": 3280
+    },
+    {
+      "epoch": 2.566302652106084,
+      "grad_norm": 6.737502098083496,
+      "learning_rate": 0.0006337196344996657,
+      "loss": 0.2278,
+      "step": 3290
+    },
+    {
+      "epoch": 2.574102964118565,
+      "grad_norm": 1.686926245689392,
+      "learning_rate": 0.0006326053042121685,
+      "loss": 0.3118,
+      "step": 3300
+    },
+    {
+      "epoch": 2.581903276131045,
+      "grad_norm": 1.436584234237671,
+      "learning_rate": 0.0006314909739246712,
+      "loss": 0.1787,
+      "step": 3310
+    },
+    {
+      "epoch": 2.589703588143526,
+      "grad_norm": 3.646476984024048,
+      "learning_rate": 0.000630376643637174,
+      "loss": 0.2165,
+      "step": 3320
+    },
+    {
+      "epoch": 2.5975039001560063,
+      "grad_norm": 1.8787113428115845,
+      "learning_rate": 0.0006292623133496768,
+      "loss": 0.1993,
+      "step": 3330
+    },
+    {
+      "epoch": 2.6053042121684866,
+      "grad_norm": 2.4485440254211426,
+      "learning_rate": 0.0006281479830621796,
+      "loss": 0.24,
+      "step": 3340
+    },
+    {
+      "epoch": 2.6131045241809674,
+      "grad_norm": 2.73197865486145,
+      "learning_rate": 0.0006270336527746824,
+      "loss": 0.228,
+      "step": 3350
+    },
+    {
+      "epoch": 2.6209048361934477,
+      "grad_norm": 3.2943315505981445,
+      "learning_rate": 0.0006259193224871852,
+      "loss": 0.2289,
+      "step": 3360
+    },
+    {
+      "epoch": 2.6287051482059285,
+      "grad_norm": 2.165308952331543,
+      "learning_rate": 0.000624804992199688,
+      "loss": 0.2435,
+      "step": 3370
+    },
+    {
+      "epoch": 2.636505460218409,
+      "grad_norm": 2.3766629695892334,
+      "learning_rate": 0.0006236906619121908,
+      "loss": 0.26,
+      "step": 3380
+    },
+    {
+      "epoch": 2.644305772230889,
+      "grad_norm": 1.1527057886123657,
+      "learning_rate": 0.0006225763316246936,
+      "loss": 0.2486,
+      "step": 3390
+    },
+    {
+      "epoch": 2.6521060842433695,
+      "grad_norm": 2.6304874420166016,
+      "learning_rate": 0.0006214620013371963,
+      "loss": 0.3827,
+      "step": 3400
+    },
+    {
+      "epoch": 2.6599063962558502,
+      "grad_norm": 1.5219537019729614,
+      "learning_rate": 0.0006203476710496991,
+      "loss": 0.2231,
+      "step": 3410
+    },
+    {
+      "epoch": 2.667706708268331,
+      "grad_norm": 1.9267528057098389,
+      "learning_rate": 0.0006192333407622019,
+      "loss": 0.2468,
+      "step": 3420
+    },
+    {
+      "epoch": 2.6755070202808113,
+      "grad_norm": 2.247861385345459,
+      "learning_rate": 0.0006181190104747047,
+      "loss": 0.2463,
+      "step": 3430
+    },
+    {
+      "epoch": 2.6833073322932917,
+      "grad_norm": 2.6133053302764893,
+      "learning_rate": 0.0006170046801872074,
+      "loss": 0.2636,
+      "step": 3440
+    },
+    {
+      "epoch": 2.691107644305772,
+      "grad_norm": 1.4675954580307007,
+      "learning_rate": 0.0006158903498997102,
+      "loss": 0.2141,
+      "step": 3450
+    },
+    {
+      "epoch": 2.6989079563182528,
+      "grad_norm": 3.3972527980804443,
+      "learning_rate": 0.000614776019612213,
+      "loss": 0.2102,
+      "step": 3460
+    },
+    {
+      "epoch": 2.706708268330733,
+      "grad_norm": 2.0020527839660645,
+      "learning_rate": 0.0006136616893247158,
+      "loss": 0.2618,
+      "step": 3470
+    },
+    {
+      "epoch": 2.714508580343214,
+      "grad_norm": 1.4799424409866333,
+      "learning_rate": 0.0006125473590372186,
+      "loss": 0.2644,
+      "step": 3480
+    },
+    {
+      "epoch": 2.722308892355694,
+      "grad_norm": 2.207921028137207,
+      "learning_rate": 0.0006114330287497214,
+      "loss": 0.2496,
+      "step": 3490
+    },
+    {
+      "epoch": 2.7301092043681745,
+      "grad_norm": 2.7361607551574707,
+      "learning_rate": 0.0006103186984622242,
+      "loss": 0.1917,
+      "step": 3500
+    },
+    {
+      "epoch": 2.7379095163806553,
+      "grad_norm": 2.6031532287597656,
+      "learning_rate": 0.000609204368174727,
+      "loss": 0.2358,
+      "step": 3510
+    },
+    {
+      "epoch": 2.7457098283931356,
+      "grad_norm": 1.442651629447937,
+      "learning_rate": 0.0006080900378872299,
+      "loss": 0.2929,
+      "step": 3520
+    },
+    {
+      "epoch": 2.7535101404056164,
+      "grad_norm": 2.853076457977295,
+      "learning_rate": 0.0006069757075997325,
+      "loss": 0.2287,
+      "step": 3530
+    },
+    {
+      "epoch": 2.7613104524180967,
+      "grad_norm": 1.827863335609436,
+      "learning_rate": 0.0006058613773122353,
+      "loss": 0.284,
+      "step": 3540
+    },
+    {
+      "epoch": 2.769110764430577,
+      "grad_norm": 1.29642915725708,
+      "learning_rate": 0.0006047470470247382,
+      "loss": 0.2156,
+      "step": 3550
+    },
+    {
+      "epoch": 2.776911076443058,
+      "grad_norm": 2.0758543014526367,
+      "learning_rate": 0.000603632716737241,
+      "loss": 0.4312,
+      "step": 3560
+    },
+    {
+      "epoch": 2.784711388455538,
+      "grad_norm": 2.1832942962646484,
+      "learning_rate": 0.0006025183864497438,
+      "loss": 0.22,
+      "step": 3570
+    },
+    {
+      "epoch": 2.792511700468019,
+      "grad_norm": 1.541040301322937,
+      "learning_rate": 0.0006014040561622464,
+      "loss": 0.2452,
+      "step": 3580
+    },
+    {
+      "epoch": 2.800312012480499,
+      "grad_norm": 1.5947539806365967,
+      "learning_rate": 0.0006002897258747492,
+      "loss": 0.2708,
+      "step": 3590
+    },
+    {
+      "epoch": 2.8081123244929795,
+      "grad_norm": 2.130390167236328,
+      "learning_rate": 0.000599175395587252,
+      "loss": 0.2194,
+      "step": 3600
+    },
+    {
+      "epoch": 2.8159126365054603,
+      "grad_norm": 2.382166862487793,
+      "learning_rate": 0.0005980610652997549,
+      "loss": 0.1931,
+      "step": 3610
+    },
+    {
+      "epoch": 2.8237129485179406,
+      "grad_norm": 4.421852111816406,
+      "learning_rate": 0.0005969467350122577,
+      "loss": 0.3081,
+      "step": 3620
+    },
+    {
+      "epoch": 2.8315132605304214,
+      "grad_norm": 1.1889768838882446,
+      "learning_rate": 0.0005958324047247605,
+      "loss": 0.2134,
+      "step": 3630
+    },
+    {
+      "epoch": 2.8393135725429017,
+      "grad_norm": 2.6204874515533447,
+      "learning_rate": 0.0005947180744372633,
+      "loss": 0.2128,
+      "step": 3640
+    },
+    {
+      "epoch": 2.847113884555382,
+      "grad_norm": 1.4705913066864014,
+      "learning_rate": 0.0005936037441497661,
+      "loss": 0.1894,
+      "step": 3650
+    },
+    {
+      "epoch": 2.854914196567863,
+      "grad_norm": 3.110135555267334,
+      "learning_rate": 0.0005924894138622688,
+      "loss": 0.2072,
+      "step": 3660
+    },
+    {
+      "epoch": 2.862714508580343,
+      "grad_norm": 1.3255491256713867,
+      "learning_rate": 0.0005913750835747716,
+      "loss": 0.2325,
+      "step": 3670
+    },
+    {
+      "epoch": 2.870514820592824,
+      "grad_norm": 2.2520713806152344,
+      "learning_rate": 0.0005902607532872744,
+      "loss": 0.2121,
+      "step": 3680
+    },
+    {
+      "epoch": 2.8783151326053042,
+      "grad_norm": 1.4630913734436035,
+      "learning_rate": 0.0005891464229997772,
+      "loss": 0.2281,
+      "step": 3690
+    },
+    {
+      "epoch": 2.8861154446177846,
+      "grad_norm": 2.0491292476654053,
+      "learning_rate": 0.00058803209271228,
+      "loss": 0.2227,
+      "step": 3700
+    },
+    {
+      "epoch": 2.8939157566302653,
+      "grad_norm": 6.133053302764893,
+      "learning_rate": 0.0005869177624247827,
+      "loss": 0.3106,
+      "step": 3710
+    },
+    {
+      "epoch": 2.9017160686427457,
+      "grad_norm": 2.3226382732391357,
+      "learning_rate": 0.0005858034321372855,
+      "loss": 0.2193,
+      "step": 3720
+    },
+    {
+      "epoch": 2.9095163806552264,
+      "grad_norm": 1.351330041885376,
+      "learning_rate": 0.0005846891018497883,
+      "loss": 0.1928,
+      "step": 3730
+    },
+    {
+      "epoch": 2.9173166926677068,
+      "grad_norm": 0.9889002442359924,
+      "learning_rate": 0.0005835747715622911,
+      "loss": 0.2395,
+      "step": 3740
+    },
+    {
+      "epoch": 2.925117004680187,
+      "grad_norm": 3.9231808185577393,
+      "learning_rate": 0.0005824604412747938,
+      "loss": 0.2288,
+      "step": 3750
+    },
+    {
+      "epoch": 2.932917316692668,
+      "grad_norm": 1.1299773454666138,
+      "learning_rate": 0.0005813461109872967,
+      "loss": 0.1996,
+      "step": 3760
+    },
+    {
+      "epoch": 2.940717628705148,
+      "grad_norm": 1.894411563873291,
+      "learning_rate": 0.0005802317806997995,
+      "loss": 0.2444,
+      "step": 3770
+    },
+    {
+      "epoch": 2.948517940717629,
+      "grad_norm": 3.100918769836426,
+      "learning_rate": 0.0005791174504123023,
+      "loss": 0.2723,
+      "step": 3780
+    },
+    {
+      "epoch": 2.9563182527301093,
+      "grad_norm": 2.921398639678955,
+      "learning_rate": 0.0005780031201248051,
+      "loss": 0.1963,
+      "step": 3790
+    },
+    {
+      "epoch": 2.9641185647425896,
+      "grad_norm": 2.898193120956421,
+      "learning_rate": 0.0005768887898373078,
+      "loss": 0.2262,
+      "step": 3800
+    },
+    {
+      "epoch": 2.97191887675507,
+      "grad_norm": 1.8510135412216187,
+      "learning_rate": 0.0005757744595498106,
+      "loss": 0.2253,
+      "step": 3810
+    },
+    {
+      "epoch": 2.9797191887675507,
+      "grad_norm": 2.716972827911377,
+      "learning_rate": 0.0005746601292623134,
+      "loss": 0.222,
+      "step": 3820
+    },
+    {
+      "epoch": 2.9875195007800315,
+      "grad_norm": 2.466111660003662,
+      "learning_rate": 0.0005735457989748162,
+      "loss": 0.2903,
+      "step": 3830
+    },
+    {
+      "epoch": 2.995319812792512,
+      "grad_norm": 2.409747838973999,
+      "learning_rate": 0.0005724314686873189,
+      "loss": 0.3088,
+      "step": 3840
+    },
+    {
+      "epoch": 3.0,
+      "eval_loss": 0.34428906440734863,
+      "eval_runtime": 557.5413,
+      "eval_samples_per_second": 0.986,
+      "eval_steps_per_second": 0.986,
+      "eval_wer": 24.454527901139215,
+      "step": 3846
+    },
+    {
+      "epoch": 3.003120124804992,
+      "grad_norm": 1.5137571096420288,
+      "learning_rate": 0.0005713171383998217,
+      "loss": 0.197,
+      "step": 3850
+    },
+    {
+      "epoch": 3.010920436817473,
+      "grad_norm": 0.7653738856315613,
+      "learning_rate": 0.0005702028081123245,
+      "loss": 0.1295,
+      "step": 3860
+    },
+    {
+      "epoch": 3.0187207488299532,
+      "grad_norm": 1.05947744846344,
+      "learning_rate": 0.0005690884778248273,
+      "loss": 0.1684,
+      "step": 3870
+    },
+    {
+      "epoch": 3.0265210608424336,
+      "grad_norm": 1.4757158756256104,
+      "learning_rate": 0.0005679741475373301,
+      "loss": 0.1379,
+      "step": 3880
+    },
+    {
+      "epoch": 3.0343213728549143,
+      "grad_norm": 1.789844274520874,
+      "learning_rate": 0.0005668598172498329,
+      "loss": 0.1711,
+      "step": 3890
+    },
+    {
+      "epoch": 3.0421216848673946,
+      "grad_norm": 1.6123414039611816,
+      "learning_rate": 0.0005657454869623357,
+      "loss": 0.1375,
+      "step": 3900
+    },
+    {
+      "epoch": 3.049921996879875,
+      "grad_norm": 1.1085964441299438,
+      "learning_rate": 0.0005646311566748385,
+      "loss": 0.1552,
+      "step": 3910
+    },
+    {
+      "epoch": 3.0577223088923557,
+      "grad_norm": 0.5497006773948669,
+      "learning_rate": 0.0005635168263873413,
+      "loss": 0.2185,
+      "step": 3920
+    },
+    {
+      "epoch": 3.065522620904836,
+      "grad_norm": 1.584539532661438,
+      "learning_rate": 0.000562402496099844,
+      "loss": 0.1398,
+      "step": 3930
+    },
+    {
+      "epoch": 3.073322932917317,
+      "grad_norm": 1.2666114568710327,
+      "learning_rate": 0.0005612881658123468,
+      "loss": 0.1372,
+      "step": 3940
+    },
+    {
+      "epoch": 3.081123244929797,
+      "grad_norm": 2.3011374473571777,
+      "learning_rate": 0.0005601738355248496,
+      "loss": 0.1354,
+      "step": 3950
+    },
+    {
+      "epoch": 3.0889235569422775,
+      "grad_norm": 1.05606210231781,
+      "learning_rate": 0.0005590595052373524,
+      "loss": 0.1158,
+      "step": 3960
+    },
+    {
+      "epoch": 3.0967238689547583,
+      "grad_norm": 1.0332001447677612,
+      "learning_rate": 0.0005579451749498552,
+      "loss": 0.1333,
+      "step": 3970
+    },
+    {
+      "epoch": 3.1045241809672386,
+      "grad_norm": 1.020912766456604,
+      "learning_rate": 0.0005568308446623579,
+      "loss": 0.1259,
+      "step": 3980
+    },
+    {
+      "epoch": 3.1123244929797194,
+      "grad_norm": 1.2610992193222046,
+      "learning_rate": 0.0005557165143748607,
+      "loss": 0.1391,
+      "step": 3990
+    },
+    {
+      "epoch": 3.1201248049921997,
+      "grad_norm": 1.1720095872879028,
+      "learning_rate": 0.0005546021840873635,
+      "loss": 0.188,
+      "step": 4000
+    },
+    {
+      "epoch": 3.12792511700468,
+      "grad_norm": 3.7483229637145996,
+      "learning_rate": 0.0005534878537998663,
+      "loss": 0.1198,
+      "step": 4010
+    },
+    {
+      "epoch": 3.135725429017161,
+      "grad_norm": 0.9843854308128357,
+      "learning_rate": 0.000552373523512369,
+      "loss": 0.1405,
+      "step": 4020
+    },
+    {
+      "epoch": 3.143525741029641,
+      "grad_norm": 2.162191152572632,
+      "learning_rate": 0.0005512591932248719,
+      "loss": 0.1268,
+      "step": 4030
+    },
+    {
+      "epoch": 3.151326053042122,
+      "grad_norm": 1.1201051473617554,
+      "learning_rate": 0.0005501448629373747,
+      "loss": 0.1354,
+      "step": 4040
+    },
+    {
+      "epoch": 3.159126365054602,
+      "grad_norm": 8.070413589477539,
+      "learning_rate": 0.0005490305326498775,
+      "loss": 0.1438,
+      "step": 4050
+    },
+    {
+      "epoch": 3.1669266770670825,
+      "grad_norm": 0.971108078956604,
+      "learning_rate": 0.0005479162023623803,
+      "loss": 0.1521,
+      "step": 4060
+    },
+    {
+      "epoch": 3.1747269890795633,
+      "grad_norm": 0.9375523924827576,
+      "learning_rate": 0.000546801872074883,
+      "loss": 0.1313,
+      "step": 4070
+    },
+    {
+      "epoch": 3.1825273010920436,
+      "grad_norm": 1.3351547718048096,
+      "learning_rate": 0.0005456875417873858,
+      "loss": 0.1626,
+      "step": 4080
+    },
+    {
+      "epoch": 3.1903276131045244,
+      "grad_norm": 2.073784589767456,
+      "learning_rate": 0.0005445732114998886,
+      "loss": 0.1435,
+      "step": 4090
+    },
+    {
+      "epoch": 3.1981279251170047,
+      "grad_norm": 1.691977620124817,
+      "learning_rate": 0.0005434588812123914,
+      "loss": 0.1742,
+      "step": 4100
+    },
+    {
+      "epoch": 3.205928237129485,
+      "grad_norm": 22.18892478942871,
+      "learning_rate": 0.0005423445509248941,
+      "loss": 0.1618,
+      "step": 4110
+    },
+    {
+      "epoch": 3.213728549141966,
+      "grad_norm": 1.7861751317977905,
+      "learning_rate": 0.0005412302206373969,
+      "loss": 0.1351,
+      "step": 4120
+    },
+    {
+      "epoch": 3.221528861154446,
+      "grad_norm": 1.1942836046218872,
+      "learning_rate": 0.0005401158903498997,
+      "loss": 0.1403,
+      "step": 4130
+    },
+    {
+      "epoch": 3.2293291731669265,
+      "grad_norm": 4.223365306854248,
+      "learning_rate": 0.0005390015600624025,
+      "loss": 0.1568,
+      "step": 4140
+    },
+    {
+      "epoch": 3.2371294851794072,
+      "grad_norm": 1.828805685043335,
+      "learning_rate": 0.0005378872297749053,
+      "loss": 0.1732,
+      "step": 4150
+    },
+    {
+      "epoch": 3.2449297971918876,
+      "grad_norm": 1.3445240259170532,
+      "learning_rate": 0.0005367728994874081,
+      "loss": 0.1609,
+      "step": 4160
+    },
+    {
+      "epoch": 3.2527301092043683,
+      "grad_norm": 1.8210989236831665,
+      "learning_rate": 0.0005356585691999109,
+      "loss": 0.1529,
+      "step": 4170
+    },
+    {
+      "epoch": 3.2605304212168487,
+      "grad_norm": 1.5427830219268799,
+      "learning_rate": 0.0005345442389124137,
+      "loss": 0.1245,
+      "step": 4180
+    },
+    {
+      "epoch": 3.268330733229329,
+      "grad_norm": 2.216970920562744,
+      "learning_rate": 0.0005334299086249165,
+      "loss": 0.1321,
+      "step": 4190
+    },
+    {
+      "epoch": 3.2761310452418098,
+      "grad_norm": 2.470158338546753,
+      "learning_rate": 0.0005323155783374192,
+      "loss": 0.1466,
+      "step": 4200
+    },
+    {
+      "epoch": 3.28393135725429,
+      "grad_norm": 1.8444827795028687,
+      "learning_rate": 0.000531201248049922,
+      "loss": 0.141,
+      "step": 4210
+    },
+    {
+      "epoch": 3.291731669266771,
+      "grad_norm": 1.771625280380249,
+      "learning_rate": 0.0005300869177624248,
+      "loss": 0.1443,
+      "step": 4220
+    },
+    {
+      "epoch": 3.299531981279251,
+      "grad_norm": 2.486222267150879,
+      "learning_rate": 0.0005289725874749276,
+      "loss": 0.1665,
+      "step": 4230
+    },
+    {
+      "epoch": 3.3073322932917315,
+      "grad_norm": 2.0185546875,
+      "learning_rate": 0.0005278582571874304,
+      "loss": 0.2224,
+      "step": 4240
+    },
+    {
+      "epoch": 3.3151326053042123,
+      "grad_norm": 1.810857892036438,
+      "learning_rate": 0.0005267439268999331,
+      "loss": 0.1384,
+      "step": 4250
+    },
+    {
+      "epoch": 3.3229329173166926,
+      "grad_norm": 0.9884628057479858,
+      "learning_rate": 0.0005256295966124359,
+      "loss": 0.1028,
+      "step": 4260
+    },
+    {
+      "epoch": 3.330733229329173,
+      "grad_norm": 0.9811989068984985,
+      "learning_rate": 0.0005245152663249387,
+      "loss": 0.1327,
+      "step": 4270
+    },
+    {
+      "epoch": 3.3385335413416537,
+      "grad_norm": 0.9268475770950317,
+      "learning_rate": 0.0005234009360374415,
+      "loss": 0.2011,
+      "step": 4280
+    },
+    {
+      "epoch": 3.346333853354134,
+      "grad_norm": 2.9813990592956543,
+      "learning_rate": 0.0005222866057499443,
+      "loss": 0.1729,
+      "step": 4290
+    },
+    {
+      "epoch": 3.354134165366615,
+      "grad_norm": 1.216805338859558,
+      "learning_rate": 0.0005211722754624471,
+      "loss": 0.148,
+      "step": 4300
+    },
+    {
+      "epoch": 3.361934477379095,
+      "grad_norm": 4.585384845733643,
+      "learning_rate": 0.0005200579451749499,
+      "loss": 0.28,
+      "step": 4310
+    },
+    {
+      "epoch": 3.3697347893915754,
+      "grad_norm": 1.331865906715393,
+      "learning_rate": 0.0005189436148874527,
+      "loss": 0.1757,
+      "step": 4320
+    },
+    {
+      "epoch": 3.377535101404056,
+      "grad_norm": 3.4767305850982666,
+      "learning_rate": 0.0005178292845999555,
+      "loss": 0.1955,
+      "step": 4330
+    },
+    {
+      "epoch": 3.3853354134165365,
+      "grad_norm": 1.2598906755447388,
+      "learning_rate": 0.0005167149543124582,
+      "loss": 0.1633,
+      "step": 4340
+    },
+    {
+      "epoch": 3.3931357254290173,
+      "grad_norm": 1.9768121242523193,
+      "learning_rate": 0.000515600624024961,
+      "loss": 0.1314,
+      "step": 4350
+    },
+    {
+      "epoch": 3.4009360374414976,
+      "grad_norm": 1.916244387626648,
+      "learning_rate": 0.0005144862937374638,
+      "loss": 0.1826,
+      "step": 4360
+    },
+    {
+      "epoch": 3.408736349453978,
+      "grad_norm": 1.159989595413208,
+      "learning_rate": 0.0005133719634499666,
+      "loss": 0.1655,
+      "step": 4370
+    },
+    {
+      "epoch": 3.4165366614664587,
+      "grad_norm": 2.1428520679473877,
+      "learning_rate": 0.0005122576331624693,
+      "loss": 0.1868,
+      "step": 4380
+    },
+    {
+      "epoch": 3.424336973478939,
+      "grad_norm": 1.3343071937561035,
+      "learning_rate": 0.0005111433028749721,
+      "loss": 0.1539,
+      "step": 4390
+    },
+    {
+      "epoch": 3.43213728549142,
+      "grad_norm": 2.33966326713562,
+      "learning_rate": 0.0005100289725874749,
+      "loss": 0.1613,
+      "step": 4400
+    },
+    {
+      "epoch": 3.4399375975039,
+      "grad_norm": 1.662204384803772,
+      "learning_rate": 0.0005089146422999777,
+      "loss": 0.1554,
+      "step": 4410
+    },
+    {
+      "epoch": 3.4477379095163805,
+      "grad_norm": 0.7760083675384521,
+      "learning_rate": 0.0005078003120124805,
+      "loss": 0.0962,
+      "step": 4420
+    },
+    {
+      "epoch": 3.4555382215288613,
+      "grad_norm": 1.860338807106018,
+      "learning_rate": 0.0005066859817249833,
+      "loss": 0.1757,
+      "step": 4430
+    },
+    {
+      "epoch": 3.4633385335413416,
+      "grad_norm": 0.6259675025939941,
+      "learning_rate": 0.0005055716514374861,
+      "loss": 0.176,
+      "step": 4440
+    },
+    {
+      "epoch": 3.4711388455538223,
+      "grad_norm": 2.819291591644287,
+      "learning_rate": 0.0005044573211499889,
+      "loss": 0.2447,
+      "step": 4450
+    },
+    {
+      "epoch": 3.4789391575663027,
+      "grad_norm": 1.125827431678772,
+      "learning_rate": 0.0005033429908624917,
+      "loss": 0.1265,
+      "step": 4460
+    },
+    {
+      "epoch": 3.486739469578783,
+      "grad_norm": 1.1929935216903687,
+      "learning_rate": 0.0005022286605749944,
+      "loss": 0.1428,
+      "step": 4470
+    },
+    {
+      "epoch": 3.4945397815912638,
+      "grad_norm": 1.2320295572280884,
+      "learning_rate": 0.0005011143302874972,
+      "loss": 0.1372,
+      "step": 4480
+    },
+    {
+      "epoch": 3.502340093603744,
+      "grad_norm": 1.1627320051193237,
+      "learning_rate": 0.0005,
+      "loss": 0.1143,
+      "step": 4490
+    },
+    {
+      "epoch": 3.510140405616225,
+      "grad_norm": 2.7850723266601562,
+      "learning_rate": 0.0004988856697125028,
+      "loss": 0.1431,
+      "step": 4500
+    },
+    {
+      "epoch": 3.517940717628705,
+      "grad_norm": 1.2920804023742676,
+      "learning_rate": 0.0004977713394250056,
+      "loss": 0.17,
+      "step": 4510
+    },
+    {
+      "epoch": 3.5257410296411855,
+      "grad_norm": 1.2807234525680542,
+      "learning_rate": 0.0004966570091375084,
+      "loss": 0.1379,
+      "step": 4520
+    },
+    {
+      "epoch": 3.5335413416536663,
+      "grad_norm": 1.2454224824905396,
+      "learning_rate": 0.0004955426788500111,
+      "loss": 0.1439,
+      "step": 4530
+    },
+    {
+      "epoch": 3.5413416536661466,
+      "grad_norm": 3.7264480590820312,
+      "learning_rate": 0.0004944283485625139,
+      "loss": 0.1691,
+      "step": 4540
+    },
+    {
+      "epoch": 3.5491419656786274,
+      "grad_norm": 1.653113842010498,
+      "learning_rate": 0.0004933140182750167,
+      "loss": 0.157,
+      "step": 4550
+    },
+    {
+      "epoch": 3.5569422776911077,
+      "grad_norm": 2.0576372146606445,
+      "learning_rate": 0.0004921996879875195,
+      "loss": 0.171,
+      "step": 4560
+    },
+    {
+      "epoch": 3.564742589703588,
+      "grad_norm": 1.499207615852356,
+      "learning_rate": 0.0004910853577000223,
+      "loss": 0.1297,
+      "step": 4570
+    },
+    {
+      "epoch": 3.572542901716069,
+      "grad_norm": 1.2240318059921265,
+      "learning_rate": 0.0004899710274125251,
+      "loss": 0.2274,
+      "step": 4580
+    },
+    {
+      "epoch": 3.580343213728549,
+      "grad_norm": 1.4562804698944092,
+      "learning_rate": 0.0004888566971250279,
+      "loss": 0.1884,
+      "step": 4590
+    },
+    {
+      "epoch": 3.58814352574103,
+      "grad_norm": 2.0862207412719727,
+      "learning_rate": 0.00048774236683753067,
+      "loss": 0.1323,
+      "step": 4600
+    },
+    {
+      "epoch": 3.5959438377535102,
+      "grad_norm": 1.8791660070419312,
+      "learning_rate": 0.00048662803655003347,
+      "loss": 0.1446,
+      "step": 4610
+    },
+    {
+      "epoch": 3.6037441497659906,
+      "grad_norm": 2.589106798171997,
+      "learning_rate": 0.0004855137062625362,
+      "loss": 0.1722,
+      "step": 4620
+    },
+    {
+      "epoch": 3.611544461778471,
+      "grad_norm": 1.3053058385849,
+      "learning_rate": 0.000484399375975039,
+      "loss": 0.1458,
+      "step": 4630
+    },
+    {
+      "epoch": 3.6193447737909517,
+      "grad_norm": 1.4898595809936523,
+      "learning_rate": 0.00048328504568754177,
+      "loss": 0.155,
+      "step": 4640
+    },
+    {
+      "epoch": 3.627145085803432,
+      "grad_norm": 1.9109569787979126,
+      "learning_rate": 0.00048217071540004457,
+      "loss": 0.1526,
+      "step": 4650
+    },
+    {
+      "epoch": 3.6349453978159127,
+      "grad_norm": 2.045905351638794,
+      "learning_rate": 0.00048105638511254737,
+      "loss": 0.1281,
+      "step": 4660
+    },
+    {
+      "epoch": 3.642745709828393,
+      "grad_norm": 2.0706069469451904,
+      "learning_rate": 0.0004799420548250502,
+      "loss": 0.2016,
+      "step": 4670
+    },
+    {
+      "epoch": 3.6505460218408734,
+      "grad_norm": 1.3242971897125244,
+      "learning_rate": 0.000478827724537553,
+      "loss": 0.1715,
+      "step": 4680
+    },
+    {
+      "epoch": 3.658346333853354,
+      "grad_norm": 1.2207064628601074,
+      "learning_rate": 0.0004777133942500557,
+      "loss": 0.1262,
+      "step": 4690
+    },
+    {
+      "epoch": 3.6661466458658345,
+      "grad_norm": 2.0521817207336426,
+      "learning_rate": 0.0004765990639625585,
+      "loss": 0.1929,
+      "step": 4700
+    },
+    {
+      "epoch": 3.6739469578783153,
+      "grad_norm": 1.009435772895813,
+      "learning_rate": 0.00047548473367506127,
+      "loss": 0.2248,
+      "step": 4710
+    },
+    {
+      "epoch": 3.6817472698907956,
+      "grad_norm": 1.105904221534729,
+      "learning_rate": 0.0004743704033875641,
+      "loss": 0.1554,
+      "step": 4720
+    },
+    {
+      "epoch": 3.689547581903276,
+      "grad_norm": 1.9131145477294922,
+      "learning_rate": 0.0004732560731000669,
+      "loss": 0.1435,
+      "step": 4730
+    },
+    {
+      "epoch": 3.6973478939157567,
+      "grad_norm": 4.082576751708984,
+      "learning_rate": 0.0004721417428125697,
+      "loss": 0.1539,
+      "step": 4740
+    },
+    {
+      "epoch": 3.705148205928237,
+      "grad_norm": 0.8368641138076782,
+      "learning_rate": 0.0004710274125250724,
+      "loss": 0.1296,
+      "step": 4750
+    },
+    {
+      "epoch": 3.712948517940718,
+      "grad_norm": 2.7137794494628906,
+      "learning_rate": 0.00046991308223757523,
+      "loss": 0.1557,
+      "step": 4760
+    },
+    {
+      "epoch": 3.720748829953198,
+      "grad_norm": 1.2786542177200317,
+      "learning_rate": 0.00046879875195007803,
+      "loss": 0.1529,
+      "step": 4770
+    },
+    {
+      "epoch": 3.7285491419656784,
+      "grad_norm": 2.214386463165283,
+      "learning_rate": 0.0004676844216625808,
+      "loss": 0.143,
+      "step": 4780
+    },
+    {
+      "epoch": 3.736349453978159,
+      "grad_norm": 1.2340818643569946,
+      "learning_rate": 0.0004665700913750836,
+      "loss": 0.1481,
+      "step": 4790
+    },
+    {
+      "epoch": 3.7441497659906395,
+      "grad_norm": 1.804880976676941,
+      "learning_rate": 0.0004654557610875864,
+      "loss": 0.1385,
+      "step": 4800
+    },
+    {
+      "epoch": 3.7519500780031203,
+      "grad_norm": 1.0058701038360596,
+      "learning_rate": 0.0004643414308000892,
+      "loss": 0.2446,
+      "step": 4810
+    },
+    {
+      "epoch": 3.7597503900156006,
+      "grad_norm": 2.6180830001831055,
+      "learning_rate": 0.00046322710051259193,
+      "loss": 0.18,
+      "step": 4820
+    },
+    {
+      "epoch": 3.767550702028081,
+      "grad_norm": 0.8924421072006226,
+      "learning_rate": 0.00046211277022509473,
+      "loss": 0.1533,
+      "step": 4830
+    },
+    {
+      "epoch": 3.7753510140405617,
+      "grad_norm": 1.5280059576034546,
+      "learning_rate": 0.0004609984399375975,
+      "loss": 0.1567,
+      "step": 4840
+    },
+    {
+      "epoch": 3.783151326053042,
+      "grad_norm": 2.1860077381134033,
+      "learning_rate": 0.0004598841096501003,
+      "loss": 0.1502,
+      "step": 4850
+    },
+    {
+      "epoch": 3.790951638065523,
+      "grad_norm": 1.2397303581237793,
+      "learning_rate": 0.00045876977936260314,
+      "loss": 0.163,
+      "step": 4860
+    },
+    {
+      "epoch": 3.798751950078003,
+      "grad_norm": 1.8644335269927979,
+      "learning_rate": 0.0004576554490751059,
+      "loss": 0.1843,
+      "step": 4870
+    },
+    {
+      "epoch": 3.8065522620904835,
+      "grad_norm": 4.985640525817871,
+      "learning_rate": 0.0004565411187876087,
+      "loss": 0.2699,
+      "step": 4880
+    },
+    {
+      "epoch": 3.8143525741029642,
+      "grad_norm": 1.4370672702789307,
+      "learning_rate": 0.00045542678850011144,
+      "loss": 0.1285,
+      "step": 4890
+    },
+    {
+      "epoch": 3.8221528861154446,
+      "grad_norm": 1.725677728652954,
+      "learning_rate": 0.00045431245821261424,
+      "loss": 0.1609,
+      "step": 4900
+    },
+    {
+      "epoch": 3.8299531981279253,
+      "grad_norm": 0.9739437699317932,
+      "learning_rate": 0.000453198127925117,
+      "loss": 0.1614,
+      "step": 4910
+    },
+    {
+      "epoch": 3.8377535101404057,
+      "grad_norm": 3.672434091567993,
+      "learning_rate": 0.0004520837976376198,
+      "loss": 0.1747,
+      "step": 4920
+    },
+    {
+      "epoch": 3.845553822152886,
+      "grad_norm": 1.3917264938354492,
+      "learning_rate": 0.0004509694673501226,
+      "loss": 0.159,
+      "step": 4930
+    },
+    {
+      "epoch": 3.8533541341653668,
+      "grad_norm": 2.578519105911255,
+      "learning_rate": 0.0004498551370626254,
+      "loss": 0.1394,
+      "step": 4940
+    },
+    {
+      "epoch": 3.861154446177847,
+      "grad_norm": 1.3072192668914795,
+      "learning_rate": 0.00044874080677512814,
+      "loss": 0.1167,
+      "step": 4950
+    },
+    {
+      "epoch": 3.868954758190328,
+      "grad_norm": 1.1134214401245117,
+      "learning_rate": 0.00044762647648763094,
+      "loss": 0.1445,
+      "step": 4960
+    },
+    {
+      "epoch": 3.876755070202808,
+      "grad_norm": 2.7288825511932373,
+      "learning_rate": 0.00044651214620013374,
+      "loss": 0.1693,
+      "step": 4970
+    },
+    {
+      "epoch": 3.8845553822152885,
+      "grad_norm": 1.1238129138946533,
+      "learning_rate": 0.0004453978159126365,
+      "loss": 0.1455,
+      "step": 4980
+    },
+    {
+      "epoch": 3.892355694227769,
+      "grad_norm": 1.7059211730957031,
+      "learning_rate": 0.0004442834856251393,
+      "loss": 0.1069,
+      "step": 4990
+    },
+    {
+      "epoch": 3.9001560062402496,
+      "grad_norm": 1.0301696062088013,
+      "learning_rate": 0.0004431691553376421,
+      "loss": 0.3035,
+      "step": 5000
+    },
+    {
+      "epoch": 3.9079563182527304,
+      "grad_norm": 1.744986891746521,
+      "learning_rate": 0.0004420548250501449,
+      "loss": 0.1249,
+      "step": 5010
+    },
+    {
+      "epoch": 3.9157566302652107,
+      "grad_norm": 0.8731406331062317,
+      "learning_rate": 0.00044094049476264764,
+      "loss": 0.1714,
+      "step": 5020
+    },
+    {
+      "epoch": 3.923556942277691,
+      "grad_norm": 1.1749571561813354,
+      "learning_rate": 0.00043982616447515045,
+      "loss": 0.264,
+      "step": 5030
+    },
+    {
+      "epoch": 3.9313572542901714,
+      "grad_norm": 2.552839756011963,
+      "learning_rate": 0.0004387118341876532,
+      "loss": 0.2517,
+      "step": 5040
+    },
+    {
+      "epoch": 3.939157566302652,
+      "grad_norm": 1.9722734689712524,
+      "learning_rate": 0.000437597503900156,
+      "loss": 0.1854,
+      "step": 5050
+    },
+    {
+      "epoch": 3.9469578783151325,
+      "grad_norm": 2.063805341720581,
+      "learning_rate": 0.0004364831736126588,
+      "loss": 0.1574,
+      "step": 5060
+    },
+    {
+      "epoch": 3.954758190327613,
+      "grad_norm": 4.838432312011719,
+      "learning_rate": 0.0004353688433251616,
+      "loss": 0.1939,
+      "step": 5070
+    },
+    {
+      "epoch": 3.9625585023400935,
+      "grad_norm": 1.9519829750061035,
+      "learning_rate": 0.0004342545130376644,
+      "loss": 0.1334,
+      "step": 5080
+    },
+    {
+      "epoch": 3.970358814352574,
+      "grad_norm": 1.3953014612197876,
+      "learning_rate": 0.00043314018275016715,
+      "loss": 0.1504,
+      "step": 5090
+    },
+    {
+      "epoch": 3.9781591263650546,
+      "grad_norm": 1.6272999048233032,
+      "learning_rate": 0.00043202585246266995,
+      "loss": 0.153,
+      "step": 5100
+    },
+    {
+      "epoch": 3.985959438377535,
+      "grad_norm": 1.610538125038147,
+      "learning_rate": 0.0004309115221751727,
+      "loss": 0.1727,
+      "step": 5110
+    },
+    {
+      "epoch": 3.9937597503900157,
+      "grad_norm": 1.6635684967041016,
+      "learning_rate": 0.0004297971918876755,
+      "loss": 0.1316,
+      "step": 5120
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 0.31802308559417725,
+      "eval_runtime": 526.9062,
+      "eval_samples_per_second": 1.044,
+      "eval_steps_per_second": 1.044,
+      "eval_wer": 21.84784707472485,
+      "step": 5128
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 8974,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 7,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 4.51799588339712e+18,
+  "train_batch_size": 3,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-5128/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:87bdde76802c6f6975c7a62016872baaa0d2d042f00a6968d4a9fadade014aa7
+size 5432

checkpoint-6410/README.md

@@ -0,0 +1,202 @@
+---
+base_model: openai/whisper-small
+library_name: peft
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.15.1

checkpoint-6410/adapter_config.json

@@ -0,0 +1,37 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": {
+    "base_model_class": "WhisperForConditionalGeneration",
+    "parent_library": "transformers.models.whisper.modeling_whisper"
+  },
+  "base_model_name_or_path": "openai/whisper-small",
+  "bias": "none",
+  "corda_config": null,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "q_proj",
+    "v_proj"
+  ],
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_rslora": false
+}

checkpoint-6410/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7313df1e75f9aca4b1a82d72980785975e3f0e30da87c2b81dff8c577dd213c4
+size 14176064

checkpoint-6410/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3ec451ebc511ef43f8ca418f2248bba15cd78b28d86cc01c9fc53638388a1300
+size 28432570

checkpoint-6410/preprocessor_config.json

@@ -0,0 +1,14 @@
+{
+  "chunk_length": 30,
+  "feature_extractor_type": "WhisperFeatureExtractor",
+  "feature_size": 80,
+  "hop_length": 160,
+  "n_fft": 400,
+  "n_samples": 480000,
+  "nb_max_frames": 3000,
+  "padding_side": "right",
+  "padding_value": 0.0,
+  "processor_class": "WhisperProcessor",
+  "return_attention_mask": false,
+  "sampling_rate": 16000
+}

checkpoint-6410/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6edcbf98f5f90a4551193d8d2d60dae946229979862138b4fb7d659856528da1
+size 14244

checkpoint-6410/scaler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:46213cf2f5a959fe173ea9d1d524a5882aabaa919b9fbc411eb868fcba63adab
+size 988