Upload folder using huggingface_hub

.gitattributes

@@ -33,3 +33,5 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+checkpoint-366/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+tokenizer.json filter=lfs diff=lfs merge=lfs -text

README.md

@@ -0,0 +1,60 @@
+---
+library_name: peft
+license: other
+base_model: Qwen/Qwen2.5-Omni-3B
+tags:
+- llama-factory
+- lora
+- generated_from_trainer
+model-index:
+- name: sft
+  results: []
+---
+
+<!-- This model card has been generated automatically according to the information the Trainer had access to. You
+should probably proofread and complete it, then remove this comment. -->
+
+# sft
+
+This model is a fine-tuned version of [Qwen/Qwen2.5-Omni-3B](https://huggingface.co/Qwen/Qwen2.5-Omni-3B) on the fingu dataset.
+
+## Model description
+
+More information needed
+
+## Intended uses & limitations
+
+More information needed
+
+## Training and evaluation data
+
+More information needed
+
+## Training procedure
+
+### Training hyperparameters
+
+The following hyperparameters were used during training:
+- learning_rate: 0.0001
+- train_batch_size: 2
+- eval_batch_size: 8
+- seed: 42
+- gradient_accumulation_steps: 4
+- total_train_batch_size: 8
+- optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
+- lr_scheduler_type: cosine
+- lr_scheduler_warmup_ratio: 0.1
+- num_epochs: 3.0
+- mixed_precision_training: Native AMP
+
+### Training results
+
+
+
+### Framework versions
+
+- PEFT 0.15.1
+- Transformers 4.52.0.dev0
+- Pytorch 2.2.0a0+81ea7a4
+- Datasets 2.17.1
+- Tokenizers 0.21.1

adapter_config.json

@@ -0,0 +1,249 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "Qwen2.5-Omni-3B/thinker",
+  "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": 16,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 8,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "layers.27.mlp.gate_proj",
+    "model.layers.31.self_attn.k_proj",
+    "model.layers.17.self_attn.k_proj",
+    "layers.7.mlp.up_proj",
+    "model.layers.22.self_attn.q_proj",
+    "model.layers.28.self_attn.q_proj",
+    "layers.12.mlp.up_proj",
+    "layers.14.mlp.gate_proj",
+    "layers.26.mlp.gate_proj",
+    "model.layers.16.self_attn.k_proj",
+    "model.layers.11.self_attn.v_proj",
+    "model.layers.16.self_attn.v_proj",
+    "layers.13.mlp.up_proj",
+    "model.layers.10.self_attn.q_proj",
+    "35.self_attn.v_proj",
+    "layers.17.mlp.up_proj",
+    "model.layers.22.self_attn.k_proj",
+    "layers.4.mlp.down_proj",
+    "layers.5.mlp.gate_proj",
+    "model.layers.6.self_attn.v_proj",
+    "model.layers.23.self_attn.v_proj",
+    "model.layers.26.self_attn.q_proj",
+    "35.self_attn.k_proj",
+    "model.layers.16.self_attn.q_proj",
+    "layers.29.mlp.gate_proj",
+    "layers.16.mlp.up_proj",
+    "layers.27.mlp.up_proj",
+    "layers.3.mlp.gate_proj",
+    "model.layers.4.self_attn.q_proj",
+    "model.layers.10.self_attn.k_proj",
+    "layers.21.mlp.gate_proj",
+    "layers.7.mlp.gate_proj",
+    "model.layers.1.self_attn.k_proj",
+    "model.layers.15.self_attn.q_proj",
+    "model.layers.27.self_attn.q_proj",
+    "model.layers.21.self_attn.v_proj",
+    "model.layers.29.self_attn.q_proj",
+    "model.layers.2.self_attn.v_proj",
+    "35.self_attn.q_proj",
+    "model.layers.5.self_attn.q_proj",
+    "layers.23.mlp.up_proj",
+    "layers.1.mlp.up_proj",
+    "layers.0.mlp.gate_proj",
+    "34.mlp.down_proj",
+    "layers.11.mlp.down_proj",
+    "model.layers.15.self_attn.v_proj",
+    "model.layers.23.self_attn.k_proj",
+    "model.layers.20.self_attn.v_proj",
+    "layers.31.mlp.up_proj",
+    "layers.13.mlp.gate_proj",
+    "layers.2.mlp.gate_proj",
+    "layers.9.mlp.gate_proj",
+    "layers.9.mlp.up_proj",
+    "layers.1.mlp.gate_proj",
+    "layers.1.mlp.down_proj",
+    "layers.7.mlp.down_proj",
+    "model.layers.19.self_attn.k_proj",
+    "model.layers.0.self_attn.k_proj",
+    "model.layers.21.self_attn.q_proj",
+    "model.layers.30.self_attn.v_proj",
+    "33.self_attn.k_proj",
+    "model.layers.3.self_attn.k_proj",
+    "32.self_attn.k_proj",
+    "layers.24.mlp.down_proj",
+    "model.layers.2.self_attn.q_proj",
+    "layers.25.mlp.down_proj",
+    "layers.10.mlp.gate_proj",
+    "32.mlp.gate_proj",
+    "model.layers.1.self_attn.v_proj",
+    "32.self_attn.q_proj",
+    "model.layers.7.self_attn.q_proj",
+    "layers.15.mlp.up_proj",
+    "model.layers.13.self_attn.v_proj",
+    "layers.30.mlp.down_proj",
+    "layers.22.mlp.up_proj",
+    "layers.3.mlp.up_proj",
+    "34.mlp.gate_proj",
+    "layers.31.mlp.gate_proj",
+    "model.layers.11.self_attn.q_proj",
+    "model.layers.27.self_attn.v_proj",
+    "model.layers.20.self_attn.q_proj",
+    "layers.15.mlp.down_proj",
+    "model.layers.29.self_attn.k_proj",
+    "model.layers.3.self_attn.v_proj",
+    "model.layers.18.self_attn.v_proj",
+    "model.layers.6.self_attn.q_proj",
+    "model.layers.18.self_attn.k_proj",
+    "model.layers.14.self_attn.q_proj",
+    "layers.22.mlp.gate_proj",
+    "layers.26.mlp.down_proj",
+    "model.layers.31.self_attn.v_proj",
+    "layers.21.mlp.down_proj",
+    "32.mlp.down_proj",
+    "model.layers.28.self_attn.v_proj",
+    "layers.20.mlp.down_proj",
+    "layers.25.mlp.up_proj",
+    "layers.28.mlp.down_proj",
+    "model.layers.1.self_attn.q_proj",
+    "33.mlp.down_proj",
+    "33.self_attn.q_proj",
+    "model.layers.5.self_attn.v_proj",
+    "35.mlp.gate_proj",
+    "model.layers.25.self_attn.k_proj",
+    "model.layers.24.self_attn.v_proj",
+    "layers.30.mlp.gate_proj",
+    "layers.24.mlp.up_proj",
+    "32.self_attn.v_proj",
+    "model.layers.3.self_attn.q_proj",
+    "layers.16.mlp.down_proj",
+    "model.layers.25.self_attn.q_proj",
+    "layers.6.mlp.up_proj",
+    "model.layers.15.self_attn.k_proj",
+    "35.mlp.up_proj",
+    "model.layers.30.self_attn.q_proj",
+    "model.layers.11.self_attn.k_proj",
+    "model.layers.17.self_attn.v_proj",
+    "model.layers.18.self_attn.q_proj",
+    "layers.15.mlp.gate_proj",
+    "layers.20.mlp.up_proj",
+    "model.layers.28.self_attn.k_proj",
+    "model.layers.12.self_attn.v_proj",
+    "layers.18.mlp.up_proj",
+    "model.layers.29.self_attn.v_proj",
+    "model.layers.7.self_attn.k_proj",
+    "model.layers.17.self_attn.q_proj",
+    "34.self_attn.k_proj",
+    "layers.0.mlp.up_proj",
+    "layers.10.mlp.up_proj",
+    "layers.11.mlp.gate_proj",
+    "34.mlp.up_proj",
+    "model.layers.6.self_attn.k_proj",
+    "layers.29.mlp.down_proj",
+    "layers.3.mlp.down_proj",
+    "model.layers.4.self_attn.k_proj",
+    "layers.26.mlp.up_proj",
+    "33.mlp.gate_proj",
+    "layers.12.mlp.down_proj",
+    "35.mlp.down_proj",
+    "layers.6.mlp.down_proj",
+    "layers.8.mlp.gate_proj",
+    "layers.2.mlp.up_proj",
+    "model.layers.22.self_attn.v_proj",
+    "layers.19.mlp.down_proj",
+    "model.layers.23.self_attn.q_proj",
+    "layers.29.mlp.up_proj",
+    "model.layers.10.self_attn.v_proj",
+    "model.layers.0.self_attn.q_proj",
+    "layers.16.mlp.gate_proj",
+    "layers.14.mlp.up_proj",
+    "layers.31.mlp.down_proj",
+    "model.layers.25.self_attn.v_proj",
+    "32.mlp.up_proj",
+    "layers.9.mlp.down_proj",
+    "layers.6.mlp.gate_proj",
+    "layers.28.mlp.up_proj",
+    "layers.17.mlp.down_proj",
+    "model.layers.20.self_attn.k_proj",
+    "layers.5.mlp.down_proj",
+    "model.layers.0.self_attn.v_proj",
+    "layers.19.mlp.up_proj",
+    "layers.0.mlp.down_proj",
+    "model.layers.9.self_attn.q_proj",
+    "layers.12.mlp.gate_proj",
+    "layers.23.mlp.down_proj",
+    "layers.4.mlp.up_proj",
+    "layers.13.mlp.down_proj",
+    "layers.18.mlp.down_proj",
+    "model.layers.14.self_attn.k_proj",
+    "layers.24.mlp.gate_proj",
+    "model.layers.14.self_attn.v_proj",
+    "model.layers.24.self_attn.q_proj",
+    "model.layers.9.self_attn.k_proj",
+    "model.layers.30.self_attn.k_proj",
+    "layers.8.mlp.up_proj",
+    "layers.23.mlp.gate_proj",
+    "layers.4.mlp.gate_proj",
+    "model.layers.8.self_attn.k_proj",
+    "model.layers.7.self_attn.v_proj",
+    "model.layers.31.self_attn.q_proj",
+    "33.self_attn.v_proj",
+    "model.layers.5.self_attn.k_proj",
+    "model.layers.26.self_attn.k_proj",
+    "model.layers.9.self_attn.v_proj",
+    "layers.5.mlp.up_proj",
+    "model.layers.8.self_attn.q_proj",
+    "layers.27.mlp.down_proj",
+    "layers.25.mlp.gate_proj",
+    "layers.2.mlp.down_proj",
+    "model.layers.27.self_attn.k_proj",
+    "model.layers.12.self_attn.q_proj",
+    "layers.8.mlp.down_proj",
+    "layers.28.mlp.gate_proj",
+    "model.layers.4.self_attn.v_proj",
+    "o_proj",
+    "model.layers.8.self_attn.v_proj",
+    "layers.11.mlp.up_proj",
+    "33.mlp.up_proj",
+    "layers.19.mlp.gate_proj",
+    "model.layers.21.self_attn.k_proj",
+    "model.layers.24.self_attn.k_proj",
+    "model.layers.13.self_attn.q_proj",
+    "model.layers.19.self_attn.q_proj",
+    "model.layers.12.self_attn.k_proj",
+    "layers.22.mlp.down_proj",
+    "34.self_attn.q_proj",
+    "layers.21.mlp.up_proj",
+    "layers.20.mlp.gate_proj",
+    "layers.14.mlp.down_proj",
+    "layers.17.mlp.gate_proj",
+    "model.layers.2.self_attn.k_proj",
+    "layers.30.mlp.up_proj",
+    "model.layers.19.self_attn.v_proj",
+    "layers.18.mlp.gate_proj",
+    "model.layers.13.self_attn.k_proj",
+    "layers.10.mlp.down_proj",
+    "model.layers.26.self_attn.v_proj",
+    "34.self_attn.v_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "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:ca4e8ec3716ba0a46d617444c3880e9ccf3733287bafa3c65499c22f6396c9e1
+size 59933632

added_tokens.json

@@ -0,0 +1,24 @@
+{
+  "</tool_call>": 151658,
+  "<tool_call>": 151657,
+  "<|AUDIO|>": 151646,
+  "<|IMAGE|>": 151655,
+  "<|VIDEO|>": 151656,
+  "<|audio_bos|>": 151647,
+  "<|audio_eos|>": 151648,
+  "<|box_end|>": 151649,
+  "<|endoftext|>": 151643,
+  "<|file_sep|>": 151664,
+  "<|fim_middle|>": 151660,
+  "<|fim_pad|>": 151662,
+  "<|fim_prefix|>": 151659,
+  "<|fim_suffix|>": 151661,
+  "<|im_end|>": 151645,
+  "<|im_start|>": 151644,
+  "<|quad_end|>": 151651,
+  "<|quad_start|>": 151650,
+  "<|repo_name|>": 151663,
+  "<|vision_bos|>": 151652,
+  "<|vision_eos|>": 151653,
+  "<|vision_pad|>": 151654
+}

all_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 2.98159509202454,
+    "total_flos": 3.142935032247091e+16,
+    "train_loss": 0.7803121163341843,
+    "train_runtime": 701.3998,
+    "train_samples_per_second": 4.183,
+    "train_steps_per_second": 0.522
+}

chat_template.jinja

@@ -0,0 +1,7 @@
+{% set audio_count = namespace(value=0) %}{% set image_count = namespace(value=0) %}{% set video_count = namespace(value=0) %}{% for message in messages %}{% if loop.first and message['role'] != 'system' %}<|im_start|>system
+You are a helpful assistant.<|im_end|>
+{% endif %}<|im_start|>{{ message['role'] }}
+{% if message['content'] is string %}{{ message['content'] }}<|im_end|>
+{% else %}{% for content in message['content'] %}{% if content['type'] == 'image' or 'image' in content or 'image_url' in content %}{% set image_count.value = image_count.value + 1 %}{% if add_vision_id %}Picture {{ image_count.value }}: {% endif %}<|vision_bos|><|IMAGE|><|vision_eos|>{% elif content['type'] == 'audio' or 'audio' in content or 'audio_url' in content %}{% set audio_count.value = audio_count.value + 1 %}{% if add_audio_id %}Audio {{ audio_count.value }}: {% endif %}<|audio_bos|><|AUDIO|><|audio_eos|>{% elif content['type'] == 'video' or 'video' in content %}{% set video_count.value = video_count.value + 1 %}{% if add_vision_id %}Video {{ video_count.value }}: {% endif %}<|vision_bos|><|VIDEO|><|vision_eos|>{% elif 'text' in content %}{{ content['text'] }}{% endif %}{% endfor %}<|im_end|>
+{% endif %}{% endfor %}{% if add_generation_prompt %}<|im_start|>assistant
+{% endif %}

checkpoint-366/README.md

@@ -0,0 +1,201 @@
+---
+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-366/adapter_config.json

@@ -0,0 +1,249 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "Qwen2.5-Omni-3B/thinker",
+  "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": 16,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 8,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "layers.27.mlp.gate_proj",
+    "model.layers.31.self_attn.k_proj",
+    "model.layers.17.self_attn.k_proj",
+    "layers.7.mlp.up_proj",
+    "model.layers.22.self_attn.q_proj",
+    "model.layers.28.self_attn.q_proj",
+    "layers.12.mlp.up_proj",
+    "layers.14.mlp.gate_proj",
+    "layers.26.mlp.gate_proj",
+    "model.layers.16.self_attn.k_proj",
+    "model.layers.11.self_attn.v_proj",
+    "model.layers.16.self_attn.v_proj",
+    "layers.13.mlp.up_proj",
+    "model.layers.10.self_attn.q_proj",
+    "35.self_attn.v_proj",
+    "layers.17.mlp.up_proj",
+    "model.layers.22.self_attn.k_proj",
+    "layers.4.mlp.down_proj",
+    "layers.5.mlp.gate_proj",
+    "model.layers.6.self_attn.v_proj",
+    "model.layers.23.self_attn.v_proj",
+    "model.layers.26.self_attn.q_proj",
+    "35.self_attn.k_proj",
+    "model.layers.16.self_attn.q_proj",
+    "layers.29.mlp.gate_proj",
+    "layers.16.mlp.up_proj",
+    "layers.27.mlp.up_proj",
+    "layers.3.mlp.gate_proj",
+    "model.layers.4.self_attn.q_proj",
+    "model.layers.10.self_attn.k_proj",
+    "layers.21.mlp.gate_proj",
+    "layers.7.mlp.gate_proj",
+    "model.layers.1.self_attn.k_proj",
+    "model.layers.15.self_attn.q_proj",
+    "model.layers.27.self_attn.q_proj",
+    "model.layers.21.self_attn.v_proj",
+    "model.layers.29.self_attn.q_proj",
+    "model.layers.2.self_attn.v_proj",
+    "35.self_attn.q_proj",
+    "model.layers.5.self_attn.q_proj",
+    "layers.23.mlp.up_proj",
+    "layers.1.mlp.up_proj",
+    "layers.0.mlp.gate_proj",
+    "34.mlp.down_proj",
+    "layers.11.mlp.down_proj",
+    "model.layers.15.self_attn.v_proj",
+    "model.layers.23.self_attn.k_proj",
+    "model.layers.20.self_attn.v_proj",
+    "layers.31.mlp.up_proj",
+    "layers.13.mlp.gate_proj",
+    "layers.2.mlp.gate_proj",
+    "layers.9.mlp.gate_proj",
+    "layers.9.mlp.up_proj",
+    "layers.1.mlp.gate_proj",
+    "layers.1.mlp.down_proj",
+    "layers.7.mlp.down_proj",
+    "model.layers.19.self_attn.k_proj",
+    "model.layers.0.self_attn.k_proj",
+    "model.layers.21.self_attn.q_proj",
+    "model.layers.30.self_attn.v_proj",
+    "33.self_attn.k_proj",
+    "model.layers.3.self_attn.k_proj",
+    "32.self_attn.k_proj",
+    "layers.24.mlp.down_proj",
+    "model.layers.2.self_attn.q_proj",
+    "layers.25.mlp.down_proj",
+    "layers.10.mlp.gate_proj",
+    "32.mlp.gate_proj",
+    "model.layers.1.self_attn.v_proj",
+    "32.self_attn.q_proj",
+    "model.layers.7.self_attn.q_proj",
+    "layers.15.mlp.up_proj",
+    "model.layers.13.self_attn.v_proj",
+    "layers.30.mlp.down_proj",
+    "layers.22.mlp.up_proj",
+    "layers.3.mlp.up_proj",
+    "34.mlp.gate_proj",
+    "layers.31.mlp.gate_proj",
+    "model.layers.11.self_attn.q_proj",
+    "model.layers.27.self_attn.v_proj",
+    "model.layers.20.self_attn.q_proj",
+    "layers.15.mlp.down_proj",
+    "model.layers.29.self_attn.k_proj",
+    "model.layers.3.self_attn.v_proj",
+    "model.layers.18.self_attn.v_proj",
+    "model.layers.6.self_attn.q_proj",
+    "model.layers.18.self_attn.k_proj",
+    "model.layers.14.self_attn.q_proj",
+    "layers.22.mlp.gate_proj",
+    "layers.26.mlp.down_proj",
+    "model.layers.31.self_attn.v_proj",
+    "layers.21.mlp.down_proj",
+    "32.mlp.down_proj",
+    "model.layers.28.self_attn.v_proj",
+    "layers.20.mlp.down_proj",
+    "layers.25.mlp.up_proj",
+    "layers.28.mlp.down_proj",
+    "model.layers.1.self_attn.q_proj",
+    "33.mlp.down_proj",
+    "33.self_attn.q_proj",
+    "model.layers.5.self_attn.v_proj",
+    "35.mlp.gate_proj",
+    "model.layers.25.self_attn.k_proj",
+    "model.layers.24.self_attn.v_proj",
+    "layers.30.mlp.gate_proj",
+    "layers.24.mlp.up_proj",
+    "32.self_attn.v_proj",
+    "model.layers.3.self_attn.q_proj",
+    "layers.16.mlp.down_proj",
+    "model.layers.25.self_attn.q_proj",
+    "layers.6.mlp.up_proj",
+    "model.layers.15.self_attn.k_proj",
+    "35.mlp.up_proj",
+    "model.layers.30.self_attn.q_proj",
+    "model.layers.11.self_attn.k_proj",
+    "model.layers.17.self_attn.v_proj",
+    "model.layers.18.self_attn.q_proj",
+    "layers.15.mlp.gate_proj",
+    "layers.20.mlp.up_proj",
+    "model.layers.28.self_attn.k_proj",
+    "model.layers.12.self_attn.v_proj",
+    "layers.18.mlp.up_proj",
+    "model.layers.29.self_attn.v_proj",
+    "model.layers.7.self_attn.k_proj",
+    "model.layers.17.self_attn.q_proj",
+    "34.self_attn.k_proj",
+    "layers.0.mlp.up_proj",
+    "layers.10.mlp.up_proj",
+    "layers.11.mlp.gate_proj",
+    "34.mlp.up_proj",
+    "model.layers.6.self_attn.k_proj",
+    "layers.29.mlp.down_proj",
+    "layers.3.mlp.down_proj",
+    "model.layers.4.self_attn.k_proj",
+    "layers.26.mlp.up_proj",
+    "33.mlp.gate_proj",
+    "layers.12.mlp.down_proj",
+    "35.mlp.down_proj",
+    "layers.6.mlp.down_proj",
+    "layers.8.mlp.gate_proj",
+    "layers.2.mlp.up_proj",
+    "model.layers.22.self_attn.v_proj",
+    "layers.19.mlp.down_proj",
+    "model.layers.23.self_attn.q_proj",
+    "layers.29.mlp.up_proj",
+    "model.layers.10.self_attn.v_proj",
+    "model.layers.0.self_attn.q_proj",
+    "layers.16.mlp.gate_proj",
+    "layers.14.mlp.up_proj",
+    "layers.31.mlp.down_proj",
+    "model.layers.25.self_attn.v_proj",
+    "32.mlp.up_proj",
+    "layers.9.mlp.down_proj",
+    "layers.6.mlp.gate_proj",
+    "layers.28.mlp.up_proj",
+    "layers.17.mlp.down_proj",
+    "model.layers.20.self_attn.k_proj",
+    "layers.5.mlp.down_proj",
+    "model.layers.0.self_attn.v_proj",
+    "layers.19.mlp.up_proj",
+    "layers.0.mlp.down_proj",
+    "model.layers.9.self_attn.q_proj",
+    "layers.12.mlp.gate_proj",
+    "layers.23.mlp.down_proj",
+    "layers.4.mlp.up_proj",
+    "layers.13.mlp.down_proj",
+    "layers.18.mlp.down_proj",
+    "model.layers.14.self_attn.k_proj",
+    "layers.24.mlp.gate_proj",
+    "model.layers.14.self_attn.v_proj",
+    "model.layers.24.self_attn.q_proj",
+    "model.layers.9.self_attn.k_proj",
+    "model.layers.30.self_attn.k_proj",
+    "layers.8.mlp.up_proj",
+    "layers.23.mlp.gate_proj",
+    "layers.4.mlp.gate_proj",
+    "model.layers.8.self_attn.k_proj",
+    "model.layers.7.self_attn.v_proj",
+    "model.layers.31.self_attn.q_proj",
+    "33.self_attn.v_proj",
+    "model.layers.5.self_attn.k_proj",
+    "model.layers.26.self_attn.k_proj",
+    "model.layers.9.self_attn.v_proj",
+    "layers.5.mlp.up_proj",
+    "model.layers.8.self_attn.q_proj",
+    "layers.27.mlp.down_proj",
+    "layers.25.mlp.gate_proj",
+    "layers.2.mlp.down_proj",
+    "model.layers.27.self_attn.k_proj",
+    "model.layers.12.self_attn.q_proj",
+    "layers.8.mlp.down_proj",
+    "layers.28.mlp.gate_proj",
+    "model.layers.4.self_attn.v_proj",
+    "o_proj",
+    "model.layers.8.self_attn.v_proj",
+    "layers.11.mlp.up_proj",
+    "33.mlp.up_proj",
+    "layers.19.mlp.gate_proj",
+    "model.layers.21.self_attn.k_proj",
+    "model.layers.24.self_attn.k_proj",
+    "model.layers.13.self_attn.q_proj",
+    "model.layers.19.self_attn.q_proj",
+    "model.layers.12.self_attn.k_proj",
+    "layers.22.mlp.down_proj",
+    "34.self_attn.q_proj",
+    "layers.21.mlp.up_proj",
+    "layers.20.mlp.gate_proj",
+    "layers.14.mlp.down_proj",
+    "layers.17.mlp.gate_proj",
+    "model.layers.2.self_attn.k_proj",
+    "layers.30.mlp.up_proj",
+    "model.layers.19.self_attn.v_proj",
+    "layers.18.mlp.gate_proj",
+    "model.layers.13.self_attn.k_proj",
+    "layers.10.mlp.down_proj",
+    "model.layers.26.self_attn.v_proj",
+    "34.self_attn.v_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_rslora": false
+}

checkpoint-366/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ca4e8ec3716ba0a46d617444c3880e9ccf3733287bafa3c65499c22f6396c9e1
+size 59933632

checkpoint-366/added_tokens.json

@@ -0,0 +1,24 @@
+{
+  "</tool_call>": 151658,
+  "<tool_call>": 151657,
+  "<|AUDIO|>": 151646,
+  "<|IMAGE|>": 151655,
+  "<|VIDEO|>": 151656,
+  "<|audio_bos|>": 151647,
+  "<|audio_eos|>": 151648,
+  "<|box_end|>": 151649,
+  "<|endoftext|>": 151643,
+  "<|file_sep|>": 151664,
+  "<|fim_middle|>": 151660,
+  "<|fim_pad|>": 151662,
+  "<|fim_prefix|>": 151659,
+  "<|fim_suffix|>": 151661,
+  "<|im_end|>": 151645,
+  "<|im_start|>": 151644,
+  "<|quad_end|>": 151651,
+  "<|quad_start|>": 151650,
+  "<|repo_name|>": 151663,
+  "<|vision_bos|>": 151652,
+  "<|vision_eos|>": 151653,
+  "<|vision_pad|>": 151654
+}

checkpoint-366/chat_template.jinja

@@ -0,0 +1,7 @@
+{% set audio_count = namespace(value=0) %}{% set image_count = namespace(value=0) %}{% set video_count = namespace(value=0) %}{% for message in messages %}{% if loop.first and message['role'] != 'system' %}<|im_start|>system
+You are a helpful assistant.<|im_end|>
+{% endif %}<|im_start|>{{ message['role'] }}
+{% if message['content'] is string %}{{ message['content'] }}<|im_end|>
+{% else %}{% for content in message['content'] %}{% if content['type'] == 'image' or 'image' in content or 'image_url' in content %}{% set image_count.value = image_count.value + 1 %}{% if add_vision_id %}Picture {{ image_count.value }}: {% endif %}<|vision_bos|><|IMAGE|><|vision_eos|>{% elif content['type'] == 'audio' or 'audio' in content or 'audio_url' in content %}{% set audio_count.value = audio_count.value + 1 %}{% if add_audio_id %}Audio {{ audio_count.value }}: {% endif %}<|audio_bos|><|AUDIO|><|audio_eos|>{% elif content['type'] == 'video' or 'video' in content %}{% set video_count.value = video_count.value + 1 %}{% if add_vision_id %}Video {{ video_count.value }}: {% endif %}<|vision_bos|><|VIDEO|><|vision_eos|>{% elif 'text' in content %}{{ content['text'] }}{% endif %}{% endfor %}<|im_end|>
+{% endif %}{% endfor %}{% if add_generation_prompt %}<|im_start|>assistant
+{% endif %}

checkpoint-366/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a8e52eb39909a4343a6cbd24219ec4c17d3433473421a6f5e3e19387b228383a
+size 120158586

checkpoint-366/preprocessor_config.json

@@ -0,0 +1,31 @@
+{
+  "chunk_length": 300,
+  "dither": 0.0,
+  "feature_extractor_type": "WhisperFeatureExtractor",
+  "feature_size": 128,
+  "hop_length": 160,
+  "image_mean": [
+    0.48145466,
+    0.4578275,
+    0.40821073
+  ],
+  "image_processor_type": "Qwen2VLImageProcessor",
+  "image_std": [
+    0.26862954,
+    0.26130258,
+    0.27577711
+  ],
+  "max_pixels": 12845056,
+  "merge_size": 2,
+  "min_pixels": 3136,
+  "n_fft": 400,
+  "n_samples": 4800000,
+  "nb_max_frames": 30000,
+  "padding_side": "right",
+  "padding_value": 0.0,
+  "patch_size": 14,
+  "processor_class": "Qwen2_5OmniProcessor",
+  "return_attention_mask": true,
+  "sampling_rate": 16000,
+  "temporal_patch_size": 2
+}

checkpoint-366/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9196a1e708bf24d6abba41cce3f8558820acc3e50f9394c5955e29eb41ffea3d
+size 14244

checkpoint-366/scaler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1dd601c9468d2cb465d756bd9391de63a702847aa0df581564cc01abc771a16e
+size 988

checkpoint-366/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7571fce4428d5a69ef3eae75d67485cce5414971763076e6e6c5a2065ba122b8
+size 1064

checkpoint-366/special_tokens_map.json

@@ -0,0 +1,38 @@
+{
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|AUDIO|>",
+    "<|audio_bos|>",
+    "<|audio_eos|>",
+    "<|box_end|>",
+    "<|quad_start|>",
+    "<|quad_end|>",
+    "<|vision_bos|>",
+    "<|vision_eos|>",
+    "<|vision_pad|>",
+    "<|IMAGE|>",
+    "<|VIDEO|>"
+  ],
+  "audio_bos_token": "<|audio_bos|>",
+  "audio_eos_token": "<|audio_eos|>",
+  "audio_token": "<|AUDIO|>",
+  "eos_token": {
+    "content": "<|im_end|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "image_token": "<|IMAGE|>",
+  "pad_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "video_token": "<|VIDEO|>",
+  "vision_bos_token": "<|vision_bos|>",
+  "vision_eos_token": "<|vision_eos|>"
+}

checkpoint-366/tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8441917e39ae0244e06d704b95b3124795cec478e297f9afac39ba670d7e9d99
+size 11421870

checkpoint-366/tokenizer_config.json

@@ -0,0 +1,223 @@
+{
+  "add_prefix_space": false,
+  "added_tokens_decoder": {
+    "151643": {
+      "content": "<|endoftext|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151644": {
+      "content": "<|im_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151645": {
+      "content": "<|im_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151646": {
+      "content": "<|AUDIO|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151647": {
+      "content": "<|audio_bos|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151648": {
+      "content": "<|audio_eos|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151649": {
+      "content": "<|box_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151650": {
+      "content": "<|quad_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151651": {
+      "content": "<|quad_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151652": {
+      "content": "<|vision_bos|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151653": {
+      "content": "<|vision_eos|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151654": {
+      "content": "<|vision_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151655": {
+      "content": "<|IMAGE|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151656": {
+      "content": "<|VIDEO|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151657": {
+      "content": "<tool_call>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151658": {
+      "content": "</tool_call>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151659": {
+      "content": "<|fim_prefix|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151660": {
+      "content": "<|fim_middle|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151661": {
+      "content": "<|fim_suffix|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151662": {
+      "content": "<|fim_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151663": {
+      "content": "<|repo_name|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151664": {
+      "content": "<|file_sep|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    }
+  },
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|AUDIO|>",
+    "<|audio_bos|>",
+    "<|audio_eos|>",
+    "<|box_end|>",
+    "<|quad_start|>",
+    "<|quad_end|>",
+    "<|vision_bos|>",
+    "<|vision_eos|>",
+    "<|vision_pad|>",
+    "<|IMAGE|>",
+    "<|VIDEO|>"
+  ],
+  "audio_bos_token": "<|audio_bos|>",
+  "audio_eos_token": "<|audio_eos|>",
+  "audio_token": "<|AUDIO|>",
+  "bos_token": null,
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|im_end|>",
+  "errors": "replace",
+  "extra_special_tokens": {
+    "audio_bos_token": "<|audio_bos|>",
+    "audio_eos_token": "<|audio_eos|>",
+    "audio_token": "<|AUDIO|>",
+    "image_token": "<|IMAGE|>",
+    "video_token": "<|VIDEO|>",
+    "vision_bos_token": "<|vision_bos|>",
+    "vision_eos_token": "<|vision_eos|>"
+  },
+  "image_token": "<|IMAGE|>",
+  "model_max_length": 32768,
+  "pad_token": "<|endoftext|>",
+  "padding_side": "right",
+  "processor_class": "Qwen2_5OmniProcessor",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null,
+  "video_token": "<|VIDEO|>",
+  "vision_bos_token": "<|vision_bos|>",
+  "vision_eos_token": "<|vision_eos|>"
+}

checkpoint-366/trainer_state.json

@@ -0,0 +1,2596 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 2.98159509202454,
+  "eval_steps": 500,
+  "global_step": 366,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.0081799591002045,
+      "grad_norm": NaN,
+      "learning_rate": 0.0,
+      "loss": 1.978,
+      "step": 1
+    },
+    {
+      "epoch": 0.016359918200409,
+      "grad_norm": NaN,
+      "learning_rate": 0.0,
+      "loss": 2.0417,
+      "step": 2
+    },
+    {
+      "epoch": 0.024539877300613498,
+      "grad_norm": 3.960280656814575,
+      "learning_rate": 0.0,
+      "loss": 2.4526,
+      "step": 3
+    },
+    {
+      "epoch": 0.032719836400818,
+      "grad_norm": 4.871670722961426,
+      "learning_rate": 2.702702702702703e-06,
+      "loss": 2.3708,
+      "step": 4
+    },
+    {
+      "epoch": 0.0408997955010225,
+      "grad_norm": NaN,
+      "learning_rate": 5.405405405405406e-06,
+      "loss": 2.7402,
+      "step": 5
+    },
+    {
+      "epoch": 0.049079754601226995,
+      "grad_norm": 2.961730718612671,
+      "learning_rate": 5.405405405405406e-06,
+      "loss": 2.6255,
+      "step": 6
+    },
+    {
+      "epoch": 0.05725971370143149,
+      "grad_norm": 2.4267706871032715,
+      "learning_rate": 8.108108108108109e-06,
+      "loss": 2.1672,
+      "step": 7
+    },
+    {
+      "epoch": 0.065439672801636,
+      "grad_norm": 4.815489768981934,
+      "learning_rate": 1.0810810810810812e-05,
+      "loss": 2.6853,
+      "step": 8
+    },
+    {
+      "epoch": 0.0736196319018405,
+      "grad_norm": 5.672784805297852,
+      "learning_rate": 1.3513513513513515e-05,
+      "loss": 2.5256,
+      "step": 9
+    },
+    {
+      "epoch": 0.081799591002045,
+      "grad_norm": 2.97552490234375,
+      "learning_rate": 1.6216216216216218e-05,
+      "loss": 2.1108,
+      "step": 10
+    },
+    {
+      "epoch": 0.08997955010224949,
+      "grad_norm": 2.305542469024658,
+      "learning_rate": 1.891891891891892e-05,
+      "loss": 2.2621,
+      "step": 11
+    },
+    {
+      "epoch": 0.09815950920245399,
+      "grad_norm": 2.330063581466675,
+      "learning_rate": 2.1621621621621624e-05,
+      "loss": 2.2973,
+      "step": 12
+    },
+    {
+      "epoch": 0.10633946830265849,
+      "grad_norm": 2.395848274230957,
+      "learning_rate": 2.4324324324324327e-05,
+      "loss": 2.2853,
+      "step": 13
+    },
+    {
+      "epoch": 0.11451942740286299,
+      "grad_norm": 3.5902745723724365,
+      "learning_rate": 2.702702702702703e-05,
+      "loss": 2.3628,
+      "step": 14
+    },
+    {
+      "epoch": 0.12269938650306748,
+      "grad_norm": 3.785466194152832,
+      "learning_rate": 2.9729729729729733e-05,
+      "loss": 1.6333,
+      "step": 15
+    },
+    {
+      "epoch": 0.130879345603272,
+      "grad_norm": 2.845073699951172,
+      "learning_rate": 3.2432432432432436e-05,
+      "loss": 1.6592,
+      "step": 16
+    },
+    {
+      "epoch": 0.1390593047034765,
+      "grad_norm": 2.9714362621307373,
+      "learning_rate": 3.513513513513514e-05,
+      "loss": 2.4303,
+      "step": 17
+    },
+    {
+      "epoch": 0.147239263803681,
+      "grad_norm": 3.2374515533447266,
+      "learning_rate": 3.783783783783784e-05,
+      "loss": 1.6276,
+      "step": 18
+    },
+    {
+      "epoch": 0.1554192229038855,
+      "grad_norm": 2.4501020908355713,
+      "learning_rate": 4.0540540540540545e-05,
+      "loss": 2.1786,
+      "step": 19
+    },
+    {
+      "epoch": 0.16359918200409,
+      "grad_norm": 1.689795970916748,
+      "learning_rate": 4.324324324324325e-05,
+      "loss": 1.7335,
+      "step": 20
+    },
+    {
+      "epoch": 0.17177914110429449,
+      "grad_norm": 1.5767645835876465,
+      "learning_rate": 4.594594594594595e-05,
+      "loss": 1.9214,
+      "step": 21
+    },
+    {
+      "epoch": 0.17995910020449898,
+      "grad_norm": 2.6853578090667725,
+      "learning_rate": 4.8648648648648654e-05,
+      "loss": 1.9089,
+      "step": 22
+    },
+    {
+      "epoch": 0.18813905930470348,
+      "grad_norm": 3.5681397914886475,
+      "learning_rate": 5.135135135135135e-05,
+      "loss": 2.092,
+      "step": 23
+    },
+    {
+      "epoch": 0.19631901840490798,
+      "grad_norm": 3.208242416381836,
+      "learning_rate": 5.405405405405406e-05,
+      "loss": 1.9184,
+      "step": 24
+    },
+    {
+      "epoch": 0.20449897750511248,
+      "grad_norm": 3.677274227142334,
+      "learning_rate": 5.6756756756756757e-05,
+      "loss": 1.6762,
+      "step": 25
+    },
+    {
+      "epoch": 0.21267893660531698,
+      "grad_norm": 1.9500216245651245,
+      "learning_rate": 5.9459459459459466e-05,
+      "loss": 1.8447,
+      "step": 26
+    },
+    {
+      "epoch": 0.22085889570552147,
+      "grad_norm": 2.72743821144104,
+      "learning_rate": 6.216216216216216e-05,
+      "loss": 1.659,
+      "step": 27
+    },
+    {
+      "epoch": 0.22903885480572597,
+      "grad_norm": 1.4266787767410278,
+      "learning_rate": 6.486486486486487e-05,
+      "loss": 1.5518,
+      "step": 28
+    },
+    {
+      "epoch": 0.23721881390593047,
+      "grad_norm": 1.8338373899459839,
+      "learning_rate": 6.756756756756757e-05,
+      "loss": 1.7725,
+      "step": 29
+    },
+    {
+      "epoch": 0.24539877300613497,
+      "grad_norm": 2.702836751937866,
+      "learning_rate": 7.027027027027028e-05,
+      "loss": 1.3214,
+      "step": 30
+    },
+    {
+      "epoch": 0.25357873210633947,
+      "grad_norm": 3.3664143085479736,
+      "learning_rate": 7.297297297297297e-05,
+      "loss": 1.5599,
+      "step": 31
+    },
+    {
+      "epoch": 0.261758691206544,
+      "grad_norm": 1.7983371019363403,
+      "learning_rate": 7.567567567567568e-05,
+      "loss": 1.4007,
+      "step": 32
+    },
+    {
+      "epoch": 0.26993865030674846,
+      "grad_norm": 1.4321403503417969,
+      "learning_rate": 7.837837837837838e-05,
+      "loss": 1.4209,
+      "step": 33
+    },
+    {
+      "epoch": 0.278118609406953,
+      "grad_norm": 1.7886905670166016,
+      "learning_rate": 8.108108108108109e-05,
+      "loss": 1.1963,
+      "step": 34
+    },
+    {
+      "epoch": 0.28629856850715746,
+      "grad_norm": 2.0502827167510986,
+      "learning_rate": 8.378378378378379e-05,
+      "loss": 1.6071,
+      "step": 35
+    },
+    {
+      "epoch": 0.294478527607362,
+      "grad_norm": 2.351100206375122,
+      "learning_rate": 8.64864864864865e-05,
+      "loss": 1.1902,
+      "step": 36
+    },
+    {
+      "epoch": 0.30265848670756645,
+      "grad_norm": 3.3446481227874756,
+      "learning_rate": 8.918918918918919e-05,
+      "loss": 1.1838,
+      "step": 37
+    },
+    {
+      "epoch": 0.310838445807771,
+      "grad_norm": 1.3906322717666626,
+      "learning_rate": 9.18918918918919e-05,
+      "loss": 0.9948,
+      "step": 38
+    },
+    {
+      "epoch": 0.31901840490797545,
+      "grad_norm": 1.9603602886199951,
+      "learning_rate": 9.45945945945946e-05,
+      "loss": 1.0997,
+      "step": 39
+    },
+    {
+      "epoch": 0.32719836400818,
+      "grad_norm": 1.3630380630493164,
+      "learning_rate": 9.729729729729731e-05,
+      "loss": 0.6781,
+      "step": 40
+    },
+    {
+      "epoch": 0.33537832310838445,
+      "grad_norm": 2.0062906742095947,
+      "learning_rate": 0.0001,
+      "loss": 1.4761,
+      "step": 41
+    },
+    {
+      "epoch": 0.34355828220858897,
+      "grad_norm": 1.2718311548233032,
+      "learning_rate": 9.99977204734326e-05,
+      "loss": 1.0233,
+      "step": 42
+    },
+    {
+      "epoch": 0.35173824130879344,
+      "grad_norm": 1.5002365112304688,
+      "learning_rate": 9.999088210158001e-05,
+      "loss": 0.8199,
+      "step": 43
+    },
+    {
+      "epoch": 0.35991820040899797,
+      "grad_norm": 1.430582880973816,
+      "learning_rate": 9.997948550797227e-05,
+      "loss": 1.1087,
+      "step": 44
+    },
+    {
+      "epoch": 0.36809815950920244,
+      "grad_norm": 2.0703279972076416,
+      "learning_rate": 9.996353173176289e-05,
+      "loss": 0.8149,
+      "step": 45
+    },
+    {
+      "epoch": 0.37627811860940696,
+      "grad_norm": 1.5086464881896973,
+      "learning_rate": 9.994302222763414e-05,
+      "loss": 0.9549,
+      "step": 46
+    },
+    {
+      "epoch": 0.38445807770961143,
+      "grad_norm": 1.5020016431808472,
+      "learning_rate": 9.991795886566441e-05,
+      "loss": 0.9542,
+      "step": 47
+    },
+    {
+      "epoch": 0.39263803680981596,
+      "grad_norm": 1.6791132688522339,
+      "learning_rate": 9.988834393115767e-05,
+      "loss": 0.8864,
+      "step": 48
+    },
+    {
+      "epoch": 0.40081799591002043,
+      "grad_norm": 2.7517454624176025,
+      "learning_rate": 9.98541801244351e-05,
+      "loss": 1.0826,
+      "step": 49
+    },
+    {
+      "epoch": 0.40899795501022496,
+      "grad_norm": 1.6971582174301147,
+      "learning_rate": 9.981547056058893e-05,
+      "loss": 0.9114,
+      "step": 50
+    },
+    {
+      "epoch": 0.4171779141104294,
+      "grad_norm": 1.4928799867630005,
+      "learning_rate": 9.977221876919833e-05,
+      "loss": 0.9755,
+      "step": 51
+    },
+    {
+      "epoch": 0.42535787321063395,
+      "grad_norm": 1.1075421571731567,
+      "learning_rate": 9.972442869400759e-05,
+      "loss": 0.741,
+      "step": 52
+    },
+    {
+      "epoch": 0.4335378323108384,
+      "grad_norm": 1.3488271236419678,
+      "learning_rate": 9.967210469256656e-05,
+      "loss": 1.0441,
+      "step": 53
+    },
+    {
+      "epoch": 0.44171779141104295,
+      "grad_norm": 3.3010079860687256,
+      "learning_rate": 9.961525153583327e-05,
+      "loss": 1.2885,
+      "step": 54
+    },
+    {
+      "epoch": 0.4498977505112474,
+      "grad_norm": 1.24274742603302,
+      "learning_rate": 9.9553874407739e-05,
+      "loss": 0.8107,
+      "step": 55
+    },
+    {
+      "epoch": 0.45807770961145194,
+      "grad_norm": 1.1659082174301147,
+      "learning_rate": 9.948797890471551e-05,
+      "loss": 0.6129,
+      "step": 56
+    },
+    {
+      "epoch": 0.4662576687116564,
+      "grad_norm": 1.5525306463241577,
+      "learning_rate": 9.941757103518478e-05,
+      "loss": 0.9262,
+      "step": 57
+    },
+    {
+      "epoch": 0.47443762781186094,
+      "grad_norm": 1.3923064470291138,
+      "learning_rate": 9.93426572190112e-05,
+      "loss": 0.7552,
+      "step": 58
+    },
+    {
+      "epoch": 0.48261758691206547,
+      "grad_norm": 1.166669487953186,
+      "learning_rate": 9.926324428691611e-05,
+      "loss": 0.6346,
+      "step": 59
+    },
+    {
+      "epoch": 0.49079754601226994,
+      "grad_norm": 2.103994131088257,
+      "learning_rate": 9.917933947985507e-05,
+      "loss": 0.8199,
+      "step": 60
+    },
+    {
+      "epoch": 0.49897750511247446,
+      "grad_norm": 1.633812665939331,
+      "learning_rate": 9.909095044835754e-05,
+      "loss": 0.7485,
+      "step": 61
+    },
+    {
+      "epoch": 0.5071574642126789,
+      "grad_norm": 1.4533647298812866,
+      "learning_rate": 9.899808525182935e-05,
+      "loss": 1.1649,
+      "step": 62
+    },
+    {
+      "epoch": 0.5153374233128835,
+      "grad_norm": 1.6267237663269043,
+      "learning_rate": 9.890075235781779e-05,
+      "loss": 1.1159,
+      "step": 63
+    },
+    {
+      "epoch": 0.523517382413088,
+      "grad_norm": 1.2796165943145752,
+      "learning_rate": 9.879896064123961e-05,
+      "loss": 0.9613,
+      "step": 64
+    },
+    {
+      "epoch": 0.5316973415132924,
+      "grad_norm": 1.3240866661071777,
+      "learning_rate": 9.869271938357167e-05,
+      "loss": 1.047,
+      "step": 65
+    },
+    {
+      "epoch": 0.5398773006134969,
+      "grad_norm": 1.190612554550171,
+      "learning_rate": 9.858203827200476e-05,
+      "loss": 1.1846,
+      "step": 66
+    },
+    {
+      "epoch": 0.5480572597137015,
+      "grad_norm": 1.1655223369598389,
+      "learning_rate": 9.846692739856024e-05,
+      "loss": 0.9566,
+      "step": 67
+    },
+    {
+      "epoch": 0.556237218813906,
+      "grad_norm": 1.2617253065109253,
+      "learning_rate": 9.834739725916988e-05,
+      "loss": 1.1108,
+      "step": 68
+    },
+    {
+      "epoch": 0.5644171779141104,
+      "grad_norm": 1.3576512336730957,
+      "learning_rate": 9.822345875271883e-05,
+      "loss": 1.1265,
+      "step": 69
+    },
+    {
+      "epoch": 0.5725971370143149,
+      "grad_norm": 1.4342156648635864,
+      "learning_rate": 9.809512318005181e-05,
+      "loss": 0.7757,
+      "step": 70
+    },
+    {
+      "epoch": 0.5807770961145194,
+      "grad_norm": 1.0733706951141357,
+      "learning_rate": 9.796240224294271e-05,
+      "loss": 0.9006,
+      "step": 71
+    },
+    {
+      "epoch": 0.588957055214724,
+      "grad_norm": 1.323440432548523,
+      "learning_rate": 9.782530804302763e-05,
+      "loss": 0.9322,
+      "step": 72
+    },
+    {
+      "epoch": 0.5971370143149284,
+      "grad_norm": 1.2899342775344849,
+      "learning_rate": 9.768385308070138e-05,
+      "loss": 0.8403,
+      "step": 73
+    },
+    {
+      "epoch": 0.6053169734151329,
+      "grad_norm": 1.2755167484283447,
+      "learning_rate": 9.753805025397779e-05,
+      "loss": 0.8397,
+      "step": 74
+    },
+    {
+      "epoch": 0.6134969325153374,
+      "grad_norm": 1.265972375869751,
+      "learning_rate": 9.738791285731352e-05,
+      "loss": 0.8143,
+      "step": 75
+    },
+    {
+      "epoch": 0.621676891615542,
+      "grad_norm": 1.1493557691574097,
+      "learning_rate": 9.723345458039594e-05,
+      "loss": 1.059,
+      "step": 76
+    },
+    {
+      "epoch": 0.6298568507157464,
+      "grad_norm": 1.1361910104751587,
+      "learning_rate": 9.707468950689491e-05,
+      "loss": 0.9112,
+      "step": 77
+    },
+    {
+      "epoch": 0.6380368098159509,
+      "grad_norm": 1.4090393781661987,
+      "learning_rate": 9.691163211317853e-05,
+      "loss": 0.7847,
+      "step": 78
+    },
+    {
+      "epoch": 0.6462167689161554,
+      "grad_norm": 1.300688624382019,
+      "learning_rate": 9.674429726699323e-05,
+      "loss": 0.9806,
+      "step": 79
+    },
+    {
+      "epoch": 0.65439672801636,
+      "grad_norm": 1.0143762826919556,
+      "learning_rate": 9.657270022610813e-05,
+      "loss": 0.6648,
+      "step": 80
+    },
+    {
+      "epoch": 0.6625766871165644,
+      "grad_norm": 1.2058460712432861,
+      "learning_rate": 9.63968566369238e-05,
+      "loss": 0.9702,
+      "step": 81
+    },
+    {
+      "epoch": 0.6707566462167689,
+      "grad_norm": 1.0876555442810059,
+      "learning_rate": 9.62167825330455e-05,
+      "loss": 0.7861,
+      "step": 82
+    },
+    {
+      "epoch": 0.6789366053169734,
+      "grad_norm": 1.502429723739624,
+      "learning_rate": 9.603249433382144e-05,
+      "loss": 1.1252,
+      "step": 83
+    },
+    {
+      "epoch": 0.6871165644171779,
+      "grad_norm": 4.1136860847473145,
+      "learning_rate": 9.584400884284545e-05,
+      "loss": 0.7109,
+      "step": 84
+    },
+    {
+      "epoch": 0.6952965235173824,
+      "grad_norm": 0.9980061650276184,
+      "learning_rate": 9.56513432464249e-05,
+      "loss": 0.6421,
+      "step": 85
+    },
+    {
+      "epoch": 0.7034764826175869,
+      "grad_norm": 1.1087136268615723,
+      "learning_rate": 9.545451511201364e-05,
+      "loss": 0.6337,
+      "step": 86
+    },
+    {
+      "epoch": 0.7116564417177914,
+      "grad_norm": 1.4353466033935547,
+      "learning_rate": 9.525354238661009e-05,
+      "loss": 1.0757,
+      "step": 87
+    },
+    {
+      "epoch": 0.7198364008179959,
+      "grad_norm": 1.1413462162017822,
+      "learning_rate": 9.504844339512095e-05,
+      "loss": 0.7056,
+      "step": 88
+    },
+    {
+      "epoch": 0.7280163599182005,
+      "grad_norm": 1.5157971382141113,
+      "learning_rate": 9.483923683869024e-05,
+      "loss": 0.8767,
+      "step": 89
+    },
+    {
+      "epoch": 0.7361963190184049,
+      "grad_norm": 0.999251663684845,
+      "learning_rate": 9.462594179299406e-05,
+      "loss": 0.922,
+      "step": 90
+    },
+    {
+      "epoch": 0.7443762781186094,
+      "grad_norm": 1.2393922805786133,
+      "learning_rate": 9.440857770650138e-05,
+      "loss": 0.8301,
+      "step": 91
+    },
+    {
+      "epoch": 0.7525562372188139,
+      "grad_norm": 1.1513807773590088,
+      "learning_rate": 9.418716439870057e-05,
+      "loss": 0.5308,
+      "step": 92
+    },
+    {
+      "epoch": 0.7607361963190185,
+      "grad_norm": 1.4229981899261475,
+      "learning_rate": 9.396172205829234e-05,
+      "loss": 1.1116,
+      "step": 93
+    },
+    {
+      "epoch": 0.7689161554192229,
+      "grad_norm": 1.4916852712631226,
+      "learning_rate": 9.373227124134888e-05,
+      "loss": 0.8806,
+      "step": 94
+    },
+    {
+      "epoch": 0.7770961145194274,
+      "grad_norm": 1.693434715270996,
+      "learning_rate": 9.34988328694395e-05,
+      "loss": 0.7924,
+      "step": 95
+    },
+    {
+      "epoch": 0.7852760736196319,
+      "grad_norm": 1.1455564498901367,
+      "learning_rate": 9.326142822772302e-05,
+      "loss": 0.7091,
+      "step": 96
+    },
+    {
+      "epoch": 0.7934560327198364,
+      "grad_norm": 1.031115174293518,
+      "learning_rate": 9.302007896300698e-05,
+      "loss": 0.8828,
+      "step": 97
+    },
+    {
+      "epoch": 0.8016359918200409,
+      "grad_norm": 1.5678527355194092,
+      "learning_rate": 9.27748070817738e-05,
+      "loss": 1.0531,
+      "step": 98
+    },
+    {
+      "epoch": 0.8098159509202454,
+      "grad_norm": 1.2558964490890503,
+      "learning_rate": 9.252563494817425e-05,
+      "loss": 0.9222,
+      "step": 99
+    },
+    {
+      "epoch": 0.8179959100204499,
+      "grad_norm": 1.2573111057281494,
+      "learning_rate": 9.227258528198831e-05,
+      "loss": 0.8988,
+      "step": 100
+    },
+    {
+      "epoch": 0.8261758691206544,
+      "grad_norm": 1.3229575157165527,
+      "learning_rate": 9.201568115655342e-05,
+      "loss": 1.0139,
+      "step": 101
+    },
+    {
+      "epoch": 0.8343558282208589,
+      "grad_norm": 0.913388729095459,
+      "learning_rate": 9.175494599666077e-05,
+      "loss": 0.6802,
+      "step": 102
+    },
+    {
+      "epoch": 0.8425357873210634,
+      "grad_norm": 3.0073134899139404,
+      "learning_rate": 9.149040357641929e-05,
+      "loss": 0.8834,
+      "step": 103
+    },
+    {
+      "epoch": 0.8507157464212679,
+      "grad_norm": 0.9436314105987549,
+      "learning_rate": 9.122207801708802e-05,
+      "loss": 0.6559,
+      "step": 104
+    },
+    {
+      "epoch": 0.8588957055214724,
+      "grad_norm": 1.1913410425186157,
+      "learning_rate": 9.094999378487659e-05,
+      "loss": 0.7427,
+      "step": 105
+    },
+    {
+      "epoch": 0.8670756646216768,
+      "grad_norm": 1.03123140335083,
+      "learning_rate": 9.067417568871445e-05,
+      "loss": 0.6253,
+      "step": 106
+    },
+    {
+      "epoch": 0.8752556237218814,
+      "grad_norm": 0.9424878358840942,
+      "learning_rate": 9.03946488779887e-05,
+      "loss": 0.7601,
+      "step": 107
+    },
+    {
+      "epoch": 0.8834355828220859,
+      "grad_norm": 1.2202138900756836,
+      "learning_rate": 9.011143884025101e-05,
+      "loss": 1.024,
+      "step": 108
+    },
+    {
+      "epoch": 0.8916155419222904,
+      "grad_norm": 1.3849170207977295,
+      "learning_rate": 8.982457139889357e-05,
+      "loss": 0.8283,
+      "step": 109
+    },
+    {
+      "epoch": 0.8997955010224948,
+      "grad_norm": 1.2288554906845093,
+      "learning_rate": 8.953407271079455e-05,
+      "loss": 0.8297,
+      "step": 110
+    },
+    {
+      "epoch": 0.9079754601226994,
+      "grad_norm": 0.9206739664077759,
+      "learning_rate": 8.923996926393305e-05,
+      "loss": 0.5966,
+      "step": 111
+    },
+    {
+      "epoch": 0.9161554192229039,
+      "grad_norm": 0.9282044172286987,
+      "learning_rate": 8.894228787497389e-05,
+      "loss": 0.7775,
+      "step": 112
+    },
+    {
+      "epoch": 0.9243353783231084,
+      "grad_norm": 1.2421815395355225,
+      "learning_rate": 8.864105568682244e-05,
+      "loss": 0.7838,
+      "step": 113
+    },
+    {
+      "epoch": 0.9325153374233128,
+      "grad_norm": 1.1172124147415161,
+      "learning_rate": 8.833630016614976e-05,
+      "loss": 0.5921,
+      "step": 114
+    },
+    {
+      "epoch": 0.9406952965235174,
+      "grad_norm": 1.310473918914795,
+      "learning_rate": 8.802804910088809e-05,
+      "loss": 1.0578,
+      "step": 115
+    },
+    {
+      "epoch": 0.9488752556237219,
+      "grad_norm": 1.1656848192214966,
+      "learning_rate": 8.771633059769711e-05,
+      "loss": 0.8205,
+      "step": 116
+    },
+    {
+      "epoch": 0.9570552147239264,
+      "grad_norm": 1.0159672498703003,
+      "learning_rate": 8.740117307940123e-05,
+      "loss": 0.8237,
+      "step": 117
+    },
+    {
+      "epoch": 0.9652351738241309,
+      "grad_norm": 1.2568827867507935,
+      "learning_rate": 8.708260528239788e-05,
+      "loss": 1.0473,
+      "step": 118
+    },
+    {
+      "epoch": 0.9734151329243353,
+      "grad_norm": 1.2711869478225708,
+      "learning_rate": 8.676065625403733e-05,
+      "loss": 1.0789,
+      "step": 119
+    },
+    {
+      "epoch": 0.9815950920245399,
+      "grad_norm": 1.2562803030014038,
+      "learning_rate": 8.64353553499741e-05,
+      "loss": 0.5315,
+      "step": 120
+    },
+    {
+      "epoch": 0.9897750511247444,
+      "grad_norm": 1.4107680320739746,
+      "learning_rate": 8.610673223149034e-05,
+      "loss": 0.9738,
+      "step": 121
+    },
+    {
+      "epoch": 0.9979550102249489,
+      "grad_norm": 1.0771377086639404,
+      "learning_rate": 8.577481686279123e-05,
+      "loss": 0.6114,
+      "step": 122
+    },
+    {
+      "epoch": 1.0,
+      "grad_norm": 0.6168379187583923,
+      "learning_rate": 8.543963950827279e-05,
+      "loss": 0.1857,
+      "step": 123
+    },
+    {
+      "epoch": 1.0081799591002045,
+      "grad_norm": 1.0099263191223145,
+      "learning_rate": 8.510123072976239e-05,
+      "loss": 0.77,
+      "step": 124
+    },
+    {
+      "epoch": 1.016359918200409,
+      "grad_norm": 1.0608845949172974,
+      "learning_rate": 8.475962138373213e-05,
+      "loss": 0.588,
+      "step": 125
+    },
+    {
+      "epoch": 1.0245398773006136,
+      "grad_norm": 1.2696729898452759,
+      "learning_rate": 8.441484261848514e-05,
+      "loss": 0.8879,
+      "step": 126
+    },
+    {
+      "epoch": 1.032719836400818,
+      "grad_norm": 1.556289792060852,
+      "learning_rate": 8.406692587131568e-05,
+      "loss": 1.1292,
+      "step": 127
+    },
+    {
+      "epoch": 1.0408997955010224,
+      "grad_norm": 1.027779459953308,
+      "learning_rate": 8.371590286564247e-05,
+      "loss": 0.6589,
+      "step": 128
+    },
+    {
+      "epoch": 1.049079754601227,
+      "grad_norm": 1.1132937669754028,
+      "learning_rate": 8.336180560811619e-05,
+      "loss": 0.6267,
+      "step": 129
+    },
+    {
+      "epoch": 1.0572597137014315,
+      "grad_norm": 1.7684708833694458,
+      "learning_rate": 8.30046663857011e-05,
+      "loss": 0.7399,
+      "step": 130
+    },
+    {
+      "epoch": 1.065439672801636,
+      "grad_norm": 1.372917890548706,
+      "learning_rate": 8.264451776273104e-05,
+      "loss": 0.8849,
+      "step": 131
+    },
+    {
+      "epoch": 1.0736196319018405,
+      "grad_norm": 1.1678389310836792,
+      "learning_rate": 8.228139257794012e-05,
+      "loss": 0.901,
+      "step": 132
+    },
+    {
+      "epoch": 1.081799591002045,
+      "grad_norm": 1.0002321004867554,
+      "learning_rate": 8.191532394146865e-05,
+      "loss": 0.3923,
+      "step": 133
+    },
+    {
+      "epoch": 1.0899795501022496,
+      "grad_norm": 1.0493892431259155,
+      "learning_rate": 8.154634523184388e-05,
+      "loss": 0.77,
+      "step": 134
+    },
+    {
+      "epoch": 1.098159509202454,
+      "grad_norm": 1.4020309448242188,
+      "learning_rate": 8.117449009293668e-05,
+      "loss": 0.7849,
+      "step": 135
+    },
+    {
+      "epoch": 1.1063394683026584,
+      "grad_norm": 1.3133962154388428,
+      "learning_rate": 8.07997924308938e-05,
+      "loss": 0.7912,
+      "step": 136
+    },
+    {
+      "epoch": 1.114519427402863,
+      "grad_norm": 1.3940467834472656,
+      "learning_rate": 8.042228641104622e-05,
+      "loss": 0.7142,
+      "step": 137
+    },
+    {
+      "epoch": 1.1226993865030674,
+      "grad_norm": 1.3877304792404175,
+      "learning_rate": 8.004200645479403e-05,
+      "loss": 0.5454,
+      "step": 138
+    },
+    {
+      "epoch": 1.130879345603272,
+      "grad_norm": 1.1132571697235107,
+      "learning_rate": 7.965898723646776e-05,
+      "loss": 0.7387,
+      "step": 139
+    },
+    {
+      "epoch": 1.1390593047034765,
+      "grad_norm": 1.206382155418396,
+      "learning_rate": 7.927326368016677e-05,
+      "loss": 0.7271,
+      "step": 140
+    },
+    {
+      "epoch": 1.147239263803681,
+      "grad_norm": 1.2813421487808228,
+      "learning_rate": 7.888487095657484e-05,
+      "loss": 0.8452,
+      "step": 141
+    },
+    {
+      "epoch": 1.1554192229038855,
+      "grad_norm": 1.1103274822235107,
+      "learning_rate": 7.849384447975321e-05,
+      "loss": 0.5735,
+      "step": 142
+    },
+    {
+      "epoch": 1.16359918200409,
+      "grad_norm": 1.1904572248458862,
+      "learning_rate": 7.810021990391164e-05,
+      "loss": 0.486,
+      "step": 143
+    },
+    {
+      "epoch": 1.1717791411042944,
+      "grad_norm": 1.361222743988037,
+      "learning_rate": 7.770403312015721e-05,
+      "loss": 0.9265,
+      "step": 144
+    },
+    {
+      "epoch": 1.179959100204499,
+      "grad_norm": 1.1453652381896973,
+      "learning_rate": 7.73053202532219e-05,
+      "loss": 0.6186,
+      "step": 145
+    },
+    {
+      "epoch": 1.1881390593047034,
+      "grad_norm": 1.2070741653442383,
+      "learning_rate": 7.690411765816864e-05,
+      "loss": 0.7012,
+      "step": 146
+    },
+    {
+      "epoch": 1.196319018404908,
+      "grad_norm": 1.4246371984481812,
+      "learning_rate": 7.650046191707641e-05,
+      "loss": 0.7644,
+      "step": 147
+    },
+    {
+      "epoch": 1.2044989775051125,
+      "grad_norm": 1.2275187969207764,
+      "learning_rate": 7.60943898357046e-05,
+      "loss": 0.614,
+      "step": 148
+    },
+    {
+      "epoch": 1.212678936605317,
+      "grad_norm": 1.292330265045166,
+      "learning_rate": 7.568593844013718e-05,
+      "loss": 0.6722,
+      "step": 149
+    },
+    {
+      "epoch": 1.2208588957055215,
+      "grad_norm": 1.54197359085083,
+      "learning_rate": 7.527514497340642e-05,
+      "loss": 0.6981,
+      "step": 150
+    },
+    {
+      "epoch": 1.229038854805726,
+      "grad_norm": 1.605914831161499,
+      "learning_rate": 7.48620468920972e-05,
+      "loss": 0.7524,
+      "step": 151
+    },
+    {
+      "epoch": 1.2372188139059306,
+      "grad_norm": 1.2442176342010498,
+      "learning_rate": 7.444668186293153e-05,
+      "loss": 0.6238,
+      "step": 152
+    },
+    {
+      "epoch": 1.2453987730061349,
+      "grad_norm": 1.4838721752166748,
+      "learning_rate": 7.402908775933419e-05,
+      "loss": 0.7599,
+      "step": 153
+    },
+    {
+      "epoch": 1.2535787321063394,
+      "grad_norm": 1.8454984426498413,
+      "learning_rate": 7.360930265797935e-05,
+      "loss": 1.1331,
+      "step": 154
+    },
+    {
+      "epoch": 1.261758691206544,
+      "grad_norm": 1.3571646213531494,
+      "learning_rate": 7.31873648353186e-05,
+      "loss": 0.6468,
+      "step": 155
+    },
+    {
+      "epoch": 1.2699386503067485,
+      "grad_norm": 1.3795866966247559,
+      "learning_rate": 7.276331276409106e-05,
+      "loss": 0.7253,
+      "step": 156
+    },
+    {
+      "epoch": 1.278118609406953,
+      "grad_norm": 1.4821308851242065,
+      "learning_rate": 7.23371851098152e-05,
+      "loss": 0.842,
+      "step": 157
+    },
+    {
+      "epoch": 1.2862985685071575,
+      "grad_norm": 1.0921138525009155,
+      "learning_rate": 7.190902072726335e-05,
+      "loss": 0.5379,
+      "step": 158
+    },
+    {
+      "epoch": 1.294478527607362,
+      "grad_norm": 1.5662935972213745,
+      "learning_rate": 7.147885865691899e-05,
+      "loss": 0.918,
+      "step": 159
+    },
+    {
+      "epoch": 1.3026584867075663,
+      "grad_norm": 1.3333555459976196,
+      "learning_rate": 7.104673812141675e-05,
+      "loss": 0.6727,
+      "step": 160
+    },
+    {
+      "epoch": 1.310838445807771,
+      "grad_norm": 1.1487497091293335,
+      "learning_rate": 7.061269852196632e-05,
+      "loss": 0.4279,
+      "step": 161
+    },
+    {
+      "epoch": 1.3190184049079754,
+      "grad_norm": 1.1033565998077393,
+      "learning_rate": 7.017677943475961e-05,
+      "loss": 0.6372,
+      "step": 162
+    },
+    {
+      "epoch": 1.32719836400818,
+      "grad_norm": 1.2100588083267212,
+      "learning_rate": 6.973902060736226e-05,
+      "loss": 0.7071,
+      "step": 163
+    },
+    {
+      "epoch": 1.3353783231083844,
+      "grad_norm": 1.421066403388977,
+      "learning_rate": 6.929946195508932e-05,
+      "loss": 0.767,
+      "step": 164
+    },
+    {
+      "epoch": 1.343558282208589,
+      "grad_norm": 1.2306902408599854,
+      "learning_rate": 6.885814355736586e-05,
+      "loss": 0.5587,
+      "step": 165
+    },
+    {
+      "epoch": 1.3517382413087935,
+      "grad_norm": 1.5315287113189697,
+      "learning_rate": 6.841510565407235e-05,
+      "loss": 0.7519,
+      "step": 166
+    },
+    {
+      "epoch": 1.359918200408998,
+      "grad_norm": 1.2497670650482178,
+      "learning_rate": 6.797038864187564e-05,
+      "loss": 0.5612,
+      "step": 167
+    },
+    {
+      "epoch": 1.3680981595092025,
+      "grad_norm": 1.6106078624725342,
+      "learning_rate": 6.752403307054549e-05,
+      "loss": 0.7194,
+      "step": 168
+    },
+    {
+      "epoch": 1.3762781186094069,
+      "grad_norm": 1.2407530546188354,
+      "learning_rate": 6.707607963925724e-05,
+      "loss": 0.531,
+      "step": 169
+    },
+    {
+      "epoch": 1.3844580777096114,
+      "grad_norm": 1.663898229598999,
+      "learning_rate": 6.66265691928808e-05,
+      "loss": 0.7906,
+      "step": 170
+    },
+    {
+      "epoch": 1.392638036809816,
+      "grad_norm": 1.3650121688842773,
+      "learning_rate": 6.617554271825636e-05,
+      "loss": 0.7207,
+      "step": 171
+    },
+    {
+      "epoch": 1.4008179959100204,
+      "grad_norm": 1.1001300811767578,
+      "learning_rate": 6.572304134045717e-05,
+      "loss": 0.5145,
+      "step": 172
+    },
+    {
+      "epoch": 1.408997955010225,
+      "grad_norm": 1.0687707662582397,
+      "learning_rate": 6.526910631903973e-05,
+      "loss": 0.3521,
+      "step": 173
+    },
+    {
+      "epoch": 1.4171779141104295,
+      "grad_norm": 1.2442213296890259,
+      "learning_rate": 6.481377904428171e-05,
+      "loss": 0.7026,
+      "step": 174
+    },
+    {
+      "epoch": 1.425357873210634,
+      "grad_norm": 1.31452214717865,
+      "learning_rate": 6.435710103340786e-05,
+      "loss": 0.7313,
+      "step": 175
+    },
+    {
+      "epoch": 1.4335378323108383,
+      "grad_norm": 1.5573769807815552,
+      "learning_rate": 6.389911392680456e-05,
+      "loss": 0.7659,
+      "step": 176
+    },
+    {
+      "epoch": 1.441717791411043,
+      "grad_norm": 1.2089431285858154,
+      "learning_rate": 6.343985948422287e-05,
+      "loss": 0.6916,
+      "step": 177
+    },
+    {
+      "epoch": 1.4498977505112474,
+      "grad_norm": 1.5785194635391235,
+      "learning_rate": 6.297937958097094e-05,
+      "loss": 0.8101,
+      "step": 178
+    },
+    {
+      "epoch": 1.4580777096114519,
+      "grad_norm": 1.4134269952774048,
+      "learning_rate": 6.251771620409563e-05,
+      "loss": 0.7504,
+      "step": 179
+    },
+    {
+      "epoch": 1.4662576687116564,
+      "grad_norm": 1.4751485586166382,
+      "learning_rate": 6.205491144855432e-05,
+      "loss": 0.5948,
+      "step": 180
+    },
+    {
+      "epoch": 1.474437627811861,
+      "grad_norm": 1.31548273563385,
+      "learning_rate": 6.159100751337642e-05,
+      "loss": 0.7057,
+      "step": 181
+    },
+    {
+      "epoch": 1.4826175869120655,
+      "grad_norm": 1.8151648044586182,
+      "learning_rate": 6.112604669781572e-05,
+      "loss": 0.9037,
+      "step": 182
+    },
+    {
+      "epoch": 1.49079754601227,
+      "grad_norm": 1.3681972026824951,
+      "learning_rate": 6.0660071397493514e-05,
+      "loss": 0.7223,
+      "step": 183
+    },
+    {
+      "epoch": 1.4989775051124745,
+      "grad_norm": 1.6292760372161865,
+      "learning_rate": 6.019312410053286e-05,
+      "loss": 0.6083,
+      "step": 184
+    },
+    {
+      "epoch": 1.5071574642126788,
+      "grad_norm": 1.8144514560699463,
+      "learning_rate": 5.972524738368452e-05,
+      "loss": 0.7662,
+      "step": 185
+    },
+    {
+      "epoch": 1.5153374233128836,
+      "grad_norm": 1.650654911994934,
+      "learning_rate": 5.925648390844476e-05,
+      "loss": 0.902,
+      "step": 186
+    },
+    {
+      "epoch": 1.5235173824130879,
+      "grad_norm": 1.4780257940292358,
+      "learning_rate": 5.878687641716538e-05,
+      "loss": 0.6566,
+      "step": 187
+    },
+    {
+      "epoch": 1.5316973415132924,
+      "grad_norm": 1.1706862449645996,
+      "learning_rate": 5.831646772915651e-05,
+      "loss": 0.4189,
+      "step": 188
+    },
+    {
+      "epoch": 1.539877300613497,
+      "grad_norm": 1.287718653678894,
+      "learning_rate": 5.7845300736782204e-05,
+      "loss": 0.5549,
+      "step": 189
+    },
+    {
+      "epoch": 1.5480572597137015,
+      "grad_norm": 1.3776918649673462,
+      "learning_rate": 5.737341840154956e-05,
+      "loss": 0.5456,
+      "step": 190
+    },
+    {
+      "epoch": 1.556237218813906,
+      "grad_norm": 1.2569301128387451,
+      "learning_rate": 5.6900863750191347e-05,
+      "loss": 0.6808,
+      "step": 191
+    },
+    {
+      "epoch": 1.5644171779141103,
+      "grad_norm": 1.7013508081436157,
+      "learning_rate": 5.642767987074288e-05,
+      "loss": 0.7974,
+      "step": 192
+    },
+    {
+      "epoch": 1.572597137014315,
+      "grad_norm": 1.5190353393554688,
+      "learning_rate": 5.5953909908613114e-05,
+      "loss": 0.5416,
+      "step": 193
+    },
+    {
+      "epoch": 1.5807770961145193,
+      "grad_norm": 1.4736334085464478,
+      "learning_rate": 5.547959706265068e-05,
+      "loss": 0.6788,
+      "step": 194
+    },
+    {
+      "epoch": 1.588957055214724,
+      "grad_norm": 2.006303548812866,
+      "learning_rate": 5.5004784581204927e-05,
+      "loss": 0.9634,
+      "step": 195
+    },
+    {
+      "epoch": 1.5971370143149284,
+      "grad_norm": 1.3578423261642456,
+      "learning_rate": 5.4529515758182506e-05,
+      "loss": 0.6563,
+      "step": 196
+    },
+    {
+      "epoch": 1.605316973415133,
+      "grad_norm": 1.4116990566253662,
+      "learning_rate": 5.405383392909973e-05,
+      "loss": 0.6062,
+      "step": 197
+    },
+    {
+      "epoch": 1.6134969325153374,
+      "grad_norm": 1.2362536191940308,
+      "learning_rate": 5.357778246713131e-05,
+      "loss": 0.4829,
+      "step": 198
+    },
+    {
+      "epoch": 1.621676891615542,
+      "grad_norm": 1.1780372858047485,
+      "learning_rate": 5.310140477915544e-05,
+      "loss": 0.465,
+      "step": 199
+    },
+    {
+      "epoch": 1.6298568507157465,
+      "grad_norm": 1.3919291496276855,
+      "learning_rate": 5.262474430179597e-05,
+      "loss": 0.6967,
+      "step": 200
+    },
+    {
+      "epoch": 1.6380368098159508,
+      "grad_norm": 1.7452629804611206,
+      "learning_rate": 5.214784449746174e-05,
+      "loss": 0.9096,
+      "step": 201
+    },
+    {
+      "epoch": 1.6462167689161555,
+      "grad_norm": 1.4730846881866455,
+      "learning_rate": 5.167074885038373e-05,
+      "loss": 0.7473,
+      "step": 202
+    },
+    {
+      "epoch": 1.6543967280163598,
+      "grad_norm": 1.5404870510101318,
+      "learning_rate": 5.119350086265004e-05,
+      "loss": 0.6233,
+      "step": 203
+    },
+    {
+      "epoch": 1.6625766871165644,
+      "grad_norm": 1.4780898094177246,
+      "learning_rate": 5.0716144050239375e-05,
+      "loss": 0.7599,
+      "step": 204
+    },
+    {
+      "epoch": 1.670756646216769,
+      "grad_norm": 1.194542407989502,
+      "learning_rate": 5.023872193905316e-05,
+      "loss": 0.5638,
+      "step": 205
+    },
+    {
+      "epoch": 1.6789366053169734,
+      "grad_norm": 1.3504347801208496,
+      "learning_rate": 4.976127806094684e-05,
+      "loss": 0.4701,
+      "step": 206
+    },
+    {
+      "epoch": 1.687116564417178,
+      "grad_norm": 2.1446099281311035,
+      "learning_rate": 4.928385594976063e-05,
+      "loss": 0.9383,
+      "step": 207
+    },
+    {
+      "epoch": 1.6952965235173822,
+      "grad_norm": 1.4745142459869385,
+      "learning_rate": 4.880649913734996e-05,
+      "loss": 0.5817,
+      "step": 208
+    },
+    {
+      "epoch": 1.703476482617587,
+      "grad_norm": 1.3029444217681885,
+      "learning_rate": 4.832925114961629e-05,
+      "loss": 0.3481,
+      "step": 209
+    },
+    {
+      "epoch": 1.7116564417177913,
+      "grad_norm": 1.1414580345153809,
+      "learning_rate": 4.785215550253826e-05,
+      "loss": 0.4348,
+      "step": 210
+    },
+    {
+      "epoch": 1.719836400817996,
+      "grad_norm": 1.4996669292449951,
+      "learning_rate": 4.7375255698204045e-05,
+      "loss": 0.653,
+      "step": 211
+    },
+    {
+      "epoch": 1.7280163599182004,
+      "grad_norm": 1.66719388961792,
+      "learning_rate": 4.6898595220844574e-05,
+      "loss": 0.7181,
+      "step": 212
+    },
+    {
+      "epoch": 1.7361963190184049,
+      "grad_norm": 1.476560354232788,
+      "learning_rate": 4.64222175328687e-05,
+      "loss": 0.6183,
+      "step": 213
+    },
+    {
+      "epoch": 1.7443762781186094,
+      "grad_norm": 1.7405219078063965,
+      "learning_rate": 4.594616607090028e-05,
+      "loss": 0.689,
+      "step": 214
+    },
+    {
+      "epoch": 1.752556237218814,
+      "grad_norm": 1.1866732835769653,
+      "learning_rate": 4.547048424181751e-05,
+      "loss": 0.4616,
+      "step": 215
+    },
+    {
+      "epoch": 1.7607361963190185,
+      "grad_norm": 1.7068077325820923,
+      "learning_rate": 4.4995215418795085e-05,
+      "loss": 0.6318,
+      "step": 216
+    },
+    {
+      "epoch": 1.7689161554192228,
+      "grad_norm": 1.4736443758010864,
+      "learning_rate": 4.452040293734934e-05,
+      "loss": 0.4611,
+      "step": 217
+    },
+    {
+      "epoch": 1.7770961145194275,
+      "grad_norm": 0.8084559440612793,
+      "learning_rate": 4.404609009138689e-05,
+      "loss": 0.1962,
+      "step": 218
+    },
+    {
+      "epoch": 1.7852760736196318,
+      "grad_norm": 1.1126220226287842,
+      "learning_rate": 4.357232012925714e-05,
+      "loss": 0.3804,
+      "step": 219
+    },
+    {
+      "epoch": 1.7934560327198366,
+      "grad_norm": 1.4977810382843018,
+      "learning_rate": 4.3099136249808665e-05,
+      "loss": 0.5431,
+      "step": 220
+    },
+    {
+      "epoch": 1.8016359918200409,
+      "grad_norm": 1.47788405418396,
+      "learning_rate": 4.262658159845046e-05,
+      "loss": 0.6498,
+      "step": 221
+    },
+    {
+      "epoch": 1.8098159509202454,
+      "grad_norm": 1.2339309453964233,
+      "learning_rate": 4.215469926321779e-05,
+      "loss": 0.4812,
+      "step": 222
+    },
+    {
+      "epoch": 1.81799591002045,
+      "grad_norm": 1.4342414140701294,
+      "learning_rate": 4.1683532270843504e-05,
+      "loss": 0.5703,
+      "step": 223
+    },
+    {
+      "epoch": 1.8261758691206544,
+      "grad_norm": 1.795954942703247,
+      "learning_rate": 4.121312358283463e-05,
+      "loss": 0.992,
+      "step": 224
+    },
+    {
+      "epoch": 1.834355828220859,
+      "grad_norm": 1.3253310918807983,
+      "learning_rate": 4.074351609155527e-05,
+      "loss": 0.5907,
+      "step": 225
+    },
+    {
+      "epoch": 1.8425357873210633,
+      "grad_norm": 1.4935518503189087,
+      "learning_rate": 4.027475261631548e-05,
+      "loss": 0.7448,
+      "step": 226
+    },
+    {
+      "epoch": 1.850715746421268,
+      "grad_norm": 1.8565064668655396,
+      "learning_rate": 3.980687589946715e-05,
+      "loss": 0.8506,
+      "step": 227
+    },
+    {
+      "epoch": 1.8588957055214723,
+      "grad_norm": 2.0860605239868164,
+      "learning_rate": 3.9339928602506505e-05,
+      "loss": 0.4935,
+      "step": 228
+    },
+    {
+      "epoch": 1.8670756646216768,
+      "grad_norm": 1.6324408054351807,
+      "learning_rate": 3.887395330218429e-05,
+      "loss": 0.586,
+      "step": 229
+    },
+    {
+      "epoch": 1.8752556237218814,
+      "grad_norm": 1.8466233015060425,
+      "learning_rate": 3.840899248662358e-05,
+      "loss": 0.7596,
+      "step": 230
+    },
+    {
+      "epoch": 1.883435582822086,
+      "grad_norm": 1.867876648902893,
+      "learning_rate": 3.7945088551445693e-05,
+      "loss": 0.7946,
+      "step": 231
+    },
+    {
+      "epoch": 1.8916155419222904,
+      "grad_norm": 1.3713316917419434,
+      "learning_rate": 3.748228379590438e-05,
+      "loss": 0.5414,
+      "step": 232
+    },
+    {
+      "epoch": 1.8997955010224947,
+      "grad_norm": 1.6689939498901367,
+      "learning_rate": 3.7020620419029094e-05,
+      "loss": 0.6574,
+      "step": 233
+    },
+    {
+      "epoch": 1.9079754601226995,
+      "grad_norm": 1.4076114892959595,
+      "learning_rate": 3.656014051577713e-05,
+      "loss": 0.5052,
+      "step": 234
+    },
+    {
+      "epoch": 1.9161554192229038,
+      "grad_norm": 1.6957688331604004,
+      "learning_rate": 3.610088607319544e-05,
+      "loss": 0.4209,
+      "step": 235
+    },
+    {
+      "epoch": 1.9243353783231085,
+      "grad_norm": 1.465134620666504,
+      "learning_rate": 3.564289896659214e-05,
+      "loss": 0.562,
+      "step": 236
+    },
+    {
+      "epoch": 1.9325153374233128,
+      "grad_norm": 1.626769781112671,
+      "learning_rate": 3.5186220955718306e-05,
+      "loss": 0.6494,
+      "step": 237
+    },
+    {
+      "epoch": 1.9406952965235174,
+      "grad_norm": 1.2987111806869507,
+      "learning_rate": 3.473089368096026e-05,
+      "loss": 0.5365,
+      "step": 238
+    },
+    {
+      "epoch": 1.9488752556237219,
+      "grad_norm": 1.7133764028549194,
+      "learning_rate": 3.427695865954284e-05,
+      "loss": 0.7972,
+      "step": 239
+    },
+    {
+      "epoch": 1.9570552147239264,
+      "grad_norm": 1.067958116531372,
+      "learning_rate": 3.3824457281743646e-05,
+      "loss": 0.2413,
+      "step": 240
+    },
+    {
+      "epoch": 1.965235173824131,
+      "grad_norm": 1.5035715103149414,
+      "learning_rate": 3.337343080711921e-05,
+      "loss": 0.655,
+      "step": 241
+    },
+    {
+      "epoch": 1.9734151329243352,
+      "grad_norm": 1.9790688753128052,
+      "learning_rate": 3.2923920360742774e-05,
+      "loss": 0.7517,
+      "step": 242
+    },
+    {
+      "epoch": 1.98159509202454,
+      "grad_norm": 1.79633367061615,
+      "learning_rate": 3.2475966929454504e-05,
+      "loss": 0.527,
+      "step": 243
+    },
+    {
+      "epoch": 1.9897750511247443,
+      "grad_norm": 1.59013032913208,
+      "learning_rate": 3.202961135812437e-05,
+      "loss": 0.5922,
+      "step": 244
+    },
+    {
+      "epoch": 1.997955010224949,
+      "grad_norm": 1.6466726064682007,
+      "learning_rate": 3.158489434592766e-05,
+      "loss": 0.6738,
+      "step": 245
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 0.8072463274002075,
+      "learning_rate": 3.114185644263415e-05,
+      "loss": 0.1228,
+      "step": 246
+    },
+    {
+      "epoch": 2.0081799591002043,
+      "grad_norm": 1.412455439567566,
+      "learning_rate": 3.070053804491068e-05,
+      "loss": 0.5372,
+      "step": 247
+    },
+    {
+      "epoch": 2.016359918200409,
+      "grad_norm": 1.15187406539917,
+      "learning_rate": 3.026097939263775e-05,
+      "loss": 0.3056,
+      "step": 248
+    },
+    {
+      "epoch": 2.0245398773006134,
+      "grad_norm": 1.3967753648757935,
+      "learning_rate": 2.9823220565240394e-05,
+      "loss": 0.5469,
+      "step": 249
+    },
+    {
+      "epoch": 2.032719836400818,
+      "grad_norm": 1.5053709745407104,
+      "learning_rate": 2.938730147803369e-05,
+      "loss": 0.5333,
+      "step": 250
+    },
+    {
+      "epoch": 2.0408997955010224,
+      "grad_norm": 1.2719367742538452,
+      "learning_rate": 2.895326187858326e-05,
+      "loss": 0.4873,
+      "step": 251
+    },
+    {
+      "epoch": 2.049079754601227,
+      "grad_norm": 1.3638321161270142,
+      "learning_rate": 2.852114134308104e-05,
+      "loss": 0.4676,
+      "step": 252
+    },
+    {
+      "epoch": 2.0572597137014315,
+      "grad_norm": 1.4079426527023315,
+      "learning_rate": 2.8090979272736662e-05,
+      "loss": 0.5474,
+      "step": 253
+    },
+    {
+      "epoch": 2.065439672801636,
+      "grad_norm": 1.3648539781570435,
+      "learning_rate": 2.7662814890184818e-05,
+      "loss": 0.3774,
+      "step": 254
+    },
+    {
+      "epoch": 2.0736196319018405,
+      "grad_norm": 1.411365032196045,
+      "learning_rate": 2.7236687235908953e-05,
+      "loss": 0.5021,
+      "step": 255
+    },
+    {
+      "epoch": 2.081799591002045,
+      "grad_norm": 0.8350080251693726,
+      "learning_rate": 2.6812635164681386e-05,
+      "loss": 0.295,
+      "step": 256
+    },
+    {
+      "epoch": 2.0899795501022496,
+      "grad_norm": 1.4837121963500977,
+      "learning_rate": 2.6390697342020665e-05,
+      "loss": 0.4359,
+      "step": 257
+    },
+    {
+      "epoch": 2.098159509202454,
+      "grad_norm": 1.447041392326355,
+      "learning_rate": 2.5970912240665813e-05,
+      "loss": 0.4699,
+      "step": 258
+    },
+    {
+      "epoch": 2.1063394683026586,
+      "grad_norm": 1.5087660551071167,
+      "learning_rate": 2.555331813706847e-05,
+      "loss": 0.5016,
+      "step": 259
+    },
+    {
+      "epoch": 2.114519427402863,
+      "grad_norm": 1.4970585107803345,
+      "learning_rate": 2.5137953107902813e-05,
+      "loss": 0.4827,
+      "step": 260
+    },
+    {
+      "epoch": 2.1226993865030677,
+      "grad_norm": 1.5823018550872803,
+      "learning_rate": 2.472485502659358e-05,
+      "loss": 0.3951,
+      "step": 261
+    },
+    {
+      "epoch": 2.130879345603272,
+      "grad_norm": 1.208630919456482,
+      "learning_rate": 2.4314061559862833e-05,
+      "loss": 0.3384,
+      "step": 262
+    },
+    {
+      "epoch": 2.1390593047034763,
+      "grad_norm": 1.6956486701965332,
+      "learning_rate": 2.3905610164295394e-05,
+      "loss": 0.4982,
+      "step": 263
+    },
+    {
+      "epoch": 2.147239263803681,
+      "grad_norm": 1.4397342205047607,
+      "learning_rate": 2.3499538082923606e-05,
+      "loss": 0.4574,
+      "step": 264
+    },
+    {
+      "epoch": 2.1554192229038853,
+      "grad_norm": 1.3102678060531616,
+      "learning_rate": 2.3095882341831372e-05,
+      "loss": 0.3559,
+      "step": 265
+    },
+    {
+      "epoch": 2.16359918200409,
+      "grad_norm": 1.2937331199645996,
+      "learning_rate": 2.2694679746778115e-05,
+      "loss": 0.3721,
+      "step": 266
+    },
+    {
+      "epoch": 2.1717791411042944,
+      "grad_norm": 1.5506526231765747,
+      "learning_rate": 2.22959668798428e-05,
+      "loss": 0.4909,
+      "step": 267
+    },
+    {
+      "epoch": 2.179959100204499,
+      "grad_norm": 1.8627556562423706,
+      "learning_rate": 2.1899780096088375e-05,
+      "loss": 0.7858,
+      "step": 268
+    },
+    {
+      "epoch": 2.1881390593047034,
+      "grad_norm": 1.7848111391067505,
+      "learning_rate": 2.1506155520246797e-05,
+      "loss": 0.6337,
+      "step": 269
+    },
+    {
+      "epoch": 2.196319018404908,
+      "grad_norm": 1.2659337520599365,
+      "learning_rate": 2.1115129043425187e-05,
+      "loss": 0.2693,
+      "step": 270
+    },
+    {
+      "epoch": 2.2044989775051125,
+      "grad_norm": 1.6412032842636108,
+      "learning_rate": 2.0726736319833228e-05,
+      "loss": 0.5306,
+      "step": 271
+    },
+    {
+      "epoch": 2.212678936605317,
+      "grad_norm": 1.611624002456665,
+      "learning_rate": 2.0341012763532243e-05,
+      "loss": 0.3388,
+      "step": 272
+    },
+    {
+      "epoch": 2.2208588957055215,
+      "grad_norm": 1.1925326585769653,
+      "learning_rate": 1.995799354520598e-05,
+      "loss": 0.3615,
+      "step": 273
+    },
+    {
+      "epoch": 2.229038854805726,
+      "grad_norm": 1.7512476444244385,
+      "learning_rate": 1.9577713588953795e-05,
+      "loss": 0.5129,
+      "step": 274
+    },
+    {
+      "epoch": 2.2372188139059306,
+      "grad_norm": 1.5006930828094482,
+      "learning_rate": 1.9200207569106216e-05,
+      "loss": 0.4129,
+      "step": 275
+    },
+    {
+      "epoch": 2.245398773006135,
+      "grad_norm": 1.8680585622787476,
+      "learning_rate": 1.8825509907063327e-05,
+      "loss": 0.5374,
+      "step": 276
+    },
+    {
+      "epoch": 2.2535787321063396,
+      "grad_norm": 1.8856024742126465,
+      "learning_rate": 1.8453654768156138e-05,
+      "loss": 0.562,
+      "step": 277
+    },
+    {
+      "epoch": 2.261758691206544,
+      "grad_norm": 1.9243358373641968,
+      "learning_rate": 1.8084676058531373e-05,
+      "loss": 0.6637,
+      "step": 278
+    },
+    {
+      "epoch": 2.2699386503067487,
+      "grad_norm": 2.3150854110717773,
+      "learning_rate": 1.771860742205988e-05,
+      "loss": 0.5932,
+      "step": 279
+    },
+    {
+      "epoch": 2.278118609406953,
+      "grad_norm": 1.2950345277786255,
+      "learning_rate": 1.7355482237268983e-05,
+      "loss": 0.341,
+      "step": 280
+    },
+    {
+      "epoch": 2.2862985685071573,
+      "grad_norm": 1.5685244798660278,
+      "learning_rate": 1.699533361429891e-05,
+      "loss": 0.4248,
+      "step": 281
+    },
+    {
+      "epoch": 2.294478527607362,
+      "grad_norm": 1.7234948873519897,
+      "learning_rate": 1.663819439188382e-05,
+      "loss": 0.7139,
+      "step": 282
+    },
+    {
+      "epoch": 2.3026584867075663,
+      "grad_norm": 1.5493229627609253,
+      "learning_rate": 1.6284097134357536e-05,
+      "loss": 0.4609,
+      "step": 283
+    },
+    {
+      "epoch": 2.310838445807771,
+      "grad_norm": 1.262978196144104,
+      "learning_rate": 1.5933074128684332e-05,
+      "loss": 0.3572,
+      "step": 284
+    },
+    {
+      "epoch": 2.3190184049079754,
+      "grad_norm": 1.7874940633773804,
+      "learning_rate": 1.5585157381514875e-05,
+      "loss": 0.5078,
+      "step": 285
+    },
+    {
+      "epoch": 2.32719836400818,
+      "grad_norm": 1.7057137489318848,
+      "learning_rate": 1.5240378616267886e-05,
+      "loss": 0.5262,
+      "step": 286
+    },
+    {
+      "epoch": 2.3353783231083844,
+      "grad_norm": 1.5174486637115479,
+      "learning_rate": 1.489876927023761e-05,
+      "loss": 0.4075,
+      "step": 287
+    },
+    {
+      "epoch": 2.3435582822085887,
+      "grad_norm": 1.473712682723999,
+      "learning_rate": 1.4560360491727231e-05,
+      "loss": 0.4237,
+      "step": 288
+    },
+    {
+      "epoch": 2.3517382413087935,
+      "grad_norm": 2.0275111198425293,
+      "learning_rate": 1.4225183137208776e-05,
+      "loss": 0.7344,
+      "step": 289
+    },
+    {
+      "epoch": 2.359918200408998,
+      "grad_norm": 1.5504990816116333,
+      "learning_rate": 1.389326776850966e-05,
+      "loss": 0.5226,
+      "step": 290
+    },
+    {
+      "epoch": 2.3680981595092025,
+      "grad_norm": 0.9763877987861633,
+      "learning_rate": 1.3564644650025893e-05,
+      "loss": 0.2004,
+      "step": 291
+    },
+    {
+      "epoch": 2.376278118609407,
+      "grad_norm": 1.6431723833084106,
+      "learning_rate": 1.3239343745962679e-05,
+      "loss": 0.5426,
+      "step": 292
+    },
+    {
+      "epoch": 2.3844580777096116,
+      "grad_norm": 1.4661204814910889,
+      "learning_rate": 1.2917394717602121e-05,
+      "loss": 0.3689,
+      "step": 293
+    },
+    {
+      "epoch": 2.392638036809816,
+      "grad_norm": 1.3995070457458496,
+      "learning_rate": 1.2598826920598772e-05,
+      "loss": 0.3994,
+      "step": 294
+    },
+    {
+      "epoch": 2.40081799591002,
+      "grad_norm": 1.6375926733016968,
+      "learning_rate": 1.2283669402302878e-05,
+      "loss": 0.4635,
+      "step": 295
+    },
+    {
+      "epoch": 2.408997955010225,
+      "grad_norm": 1.6579980850219727,
+      "learning_rate": 1.197195089911191e-05,
+      "loss": 0.44,
+      "step": 296
+    },
+    {
+      "epoch": 2.4171779141104293,
+      "grad_norm": 2.057859420776367,
+      "learning_rate": 1.1663699833850238e-05,
+      "loss": 0.809,
+      "step": 297
+    },
+    {
+      "epoch": 2.425357873210634,
+      "grad_norm": 1.9846243858337402,
+      "learning_rate": 1.1358944313177567e-05,
+      "loss": 0.526,
+      "step": 298
+    },
+    {
+      "epoch": 2.4335378323108383,
+      "grad_norm": 1.8454967737197876,
+      "learning_rate": 1.1057712125026116e-05,
+      "loss": 0.4943,
+      "step": 299
+    },
+    {
+      "epoch": 2.441717791411043,
+      "grad_norm": 1.3751471042633057,
+      "learning_rate": 1.0760030736066951e-05,
+      "loss": 0.2973,
+      "step": 300
+    },
+    {
+      "epoch": 2.4498977505112474,
+      "grad_norm": 1.7352081537246704,
+      "learning_rate": 1.0465927289205452e-05,
+      "loss": 0.4647,
+      "step": 301
+    },
+    {
+      "epoch": 2.458077709611452,
+      "grad_norm": 1.6583192348480225,
+      "learning_rate": 1.017542860110644e-05,
+      "loss": 0.5614,
+      "step": 302
+    },
+    {
+      "epoch": 2.4662576687116564,
+      "grad_norm": 1.1086567640304565,
+      "learning_rate": 9.888561159748993e-06,
+      "loss": 0.2343,
+      "step": 303
+    },
+    {
+      "epoch": 2.474437627811861,
+      "grad_norm": 1.2182183265686035,
+      "learning_rate": 9.605351122011309e-06,
+      "loss": 0.5084,
+      "step": 304
+    },
+    {
+      "epoch": 2.4826175869120655,
+      "grad_norm": 1.5897687673568726,
+      "learning_rate": 9.325824311285564e-06,
+      "loss": 0.4916,
+      "step": 305
+    },
+    {
+      "epoch": 2.4907975460122698,
+      "grad_norm": 1.7576637268066406,
+      "learning_rate": 9.050006215123419e-06,
+      "loss": 0.5896,
+      "step": 306
+    },
+    {
+      "epoch": 2.4989775051124745,
+      "grad_norm": 1.3375118970870972,
+      "learning_rate": 8.777921982911996e-06,
+      "loss": 0.3472,
+      "step": 307
+    },
+    {
+      "epoch": 2.507157464212679,
+      "grad_norm": 1.643762230873108,
+      "learning_rate": 8.509596423580712e-06,
+      "loss": 0.6561,
+      "step": 308
+    },
+    {
+      "epoch": 2.5153374233128836,
+      "grad_norm": 1.8207759857177734,
+      "learning_rate": 8.245054003339247e-06,
+      "loss": 0.446,
+      "step": 309
+    },
+    {
+      "epoch": 2.523517382413088,
+      "grad_norm": 1.7931218147277832,
+      "learning_rate": 7.984318843446593e-06,
+      "loss": 0.6626,
+      "step": 310
+    },
+    {
+      "epoch": 2.5316973415132926,
+      "grad_norm": 1.5871256589889526,
+      "learning_rate": 7.727414718011704e-06,
+      "loss": 0.6779,
+      "step": 311
+    },
+    {
+      "epoch": 2.539877300613497,
+      "grad_norm": 1.6045511960983276,
+      "learning_rate": 7.474365051825749e-06,
+      "loss": 0.4369,
+      "step": 312
+    },
+    {
+      "epoch": 2.5480572597137012,
+      "grad_norm": 1.9614039659500122,
+      "learning_rate": 7.225192918226214e-06,
+      "loss": 0.5339,
+      "step": 313
+    },
+    {
+      "epoch": 2.556237218813906,
+      "grad_norm": 1.6761356592178345,
+      "learning_rate": 6.979921036993042e-06,
+      "loss": 0.4714,
+      "step": 314
+    },
+    {
+      "epoch": 2.5644171779141103,
+      "grad_norm": 1.268598198890686,
+      "learning_rate": 6.738571772276997e-06,
+      "loss": 0.3589,
+      "step": 315
+    },
+    {
+      "epoch": 2.572597137014315,
+      "grad_norm": 1.9515974521636963,
+      "learning_rate": 6.501167130560515e-06,
+      "loss": 0.7677,
+      "step": 316
+    },
+    {
+      "epoch": 2.5807770961145193,
+      "grad_norm": 1.752503514289856,
+      "learning_rate": 6.267728758651132e-06,
+      "loss": 0.6019,
+      "step": 317
+    },
+    {
+      "epoch": 2.588957055214724,
+      "grad_norm": 1.6404023170471191,
+      "learning_rate": 6.03827794170767e-06,
+      "loss": 0.3813,
+      "step": 318
+    },
+    {
+      "epoch": 2.5971370143149284,
+      "grad_norm": 1.6431866884231567,
+      "learning_rate": 5.8128356012994375e-06,
+      "loss": 0.5397,
+      "step": 319
+    },
+    {
+      "epoch": 2.6053169734151327,
+      "grad_norm": 1.604200005531311,
+      "learning_rate": 5.591422293498633e-06,
+      "loss": 0.5326,
+      "step": 320
+    },
+    {
+      "epoch": 2.6134969325153374,
+      "grad_norm": 1.955712080001831,
+      "learning_rate": 5.374058207005944e-06,
+      "loss": 0.6279,
+      "step": 321
+    },
+    {
+      "epoch": 2.621676891615542,
+      "grad_norm": 1.9583613872528076,
+      "learning_rate": 5.160763161309767e-06,
+      "loss": 0.7064,
+      "step": 322
+    },
+    {
+      "epoch": 2.6298568507157465,
+      "grad_norm": 1.465756893157959,
+      "learning_rate": 4.951556604879048e-06,
+      "loss": 0.3176,
+      "step": 323
+    },
+    {
+      "epoch": 2.638036809815951,
+      "grad_norm": 1.0220084190368652,
+      "learning_rate": 4.746457613389904e-06,
+      "loss": 0.1989,
+      "step": 324
+    },
+    {
+      "epoch": 2.6462167689161555,
+      "grad_norm": 1.9900139570236206,
+      "learning_rate": 4.545484887986368e-06,
+      "loss": 0.4488,
+      "step": 325
+    },
+    {
+      "epoch": 2.65439672801636,
+      "grad_norm": 1.8389681577682495,
+      "learning_rate": 4.348656753575092e-06,
+      "loss": 0.8159,
+      "step": 326
+    },
+    {
+      "epoch": 2.662576687116564,
+      "grad_norm": 1.8046656847000122,
+      "learning_rate": 4.155991157154554e-06,
+      "loss": 0.5941,
+      "step": 327
+    },
+    {
+      "epoch": 2.670756646216769,
+      "grad_norm": 1.5946298837661743,
+      "learning_rate": 3.967505666178556e-06,
+      "loss": 0.6167,
+      "step": 328
+    },
+    {
+      "epoch": 2.6789366053169736,
+      "grad_norm": 1.6215424537658691,
+      "learning_rate": 3.783217466954503e-06,
+      "loss": 0.5432,
+      "step": 329
+    },
+    {
+      "epoch": 2.687116564417178,
+      "grad_norm": 1.5136370658874512,
+      "learning_rate": 3.603143363076217e-06,
+      "loss": 0.2688,
+      "step": 330
+    },
+    {
+      "epoch": 2.6952965235173822,
+      "grad_norm": 2.0225648880004883,
+      "learning_rate": 3.427299773891868e-06,
+      "loss": 0.3968,
+      "step": 331
+    },
+    {
+      "epoch": 2.703476482617587,
+      "grad_norm": 1.170069694519043,
+      "learning_rate": 3.2557027330067658e-06,
+      "loss": 0.3143,
+      "step": 332
+    },
+    {
+      "epoch": 2.7116564417177913,
+      "grad_norm": 1.2336766719818115,
+      "learning_rate": 3.0883678868214806e-06,
+      "loss": 0.4023,
+      "step": 333
+    },
+    {
+      "epoch": 2.719836400817996,
+      "grad_norm": 1.8785996437072754,
+      "learning_rate": 2.925310493105099e-06,
+      "loss": 0.6501,
+      "step": 334
+    },
+    {
+      "epoch": 2.7280163599182004,
+      "grad_norm": 1.7136589288711548,
+      "learning_rate": 2.7665454196040664e-06,
+      "loss": 0.3418,
+      "step": 335
+    },
+    {
+      "epoch": 2.736196319018405,
+      "grad_norm": 1.5453672409057617,
+      "learning_rate": 2.612087142686487e-06,
+      "loss": 0.4047,
+      "step": 336
+    },
+    {
+      "epoch": 2.7443762781186094,
+      "grad_norm": 1.5091831684112549,
+      "learning_rate": 2.4619497460222184e-06,
+      "loss": 0.3707,
+      "step": 337
+    },
+    {
+      "epoch": 2.7525562372188137,
+      "grad_norm": 1.996533751487732,
+      "learning_rate": 2.316146919298623e-06,
+      "loss": 0.7776,
+      "step": 338
+    },
+    {
+      "epoch": 2.7607361963190185,
+      "grad_norm": 2.2293291091918945,
+      "learning_rate": 2.1746919569723855e-06,
+      "loss": 0.7055,
+      "step": 339
+    },
+    {
+      "epoch": 2.7689161554192228,
+      "grad_norm": 1.906553864479065,
+      "learning_rate": 2.0375977570572967e-06,
+      "loss": 0.6423,
+      "step": 340
+    },
+    {
+      "epoch": 2.7770961145194275,
+      "grad_norm": 1.684910535812378,
+      "learning_rate": 1.9048768199481982e-06,
+      "loss": 0.5679,
+      "step": 341
+    },
+    {
+      "epoch": 2.785276073619632,
+      "grad_norm": 1.3118062019348145,
+      "learning_rate": 1.7765412472811771e-06,
+      "loss": 0.3036,
+      "step": 342
+    },
+    {
+      "epoch": 2.7934560327198366,
+      "grad_norm": 1.9178974628448486,
+      "learning_rate": 1.6526027408301226e-06,
+      "loss": 0.5829,
+      "step": 343
+    },
+    {
+      "epoch": 2.801635991820041,
+      "grad_norm": 1.860939860343933,
+      "learning_rate": 1.5330726014397668e-06,
+      "loss": 0.4617,
+      "step": 344
+    },
+    {
+      "epoch": 2.809815950920245,
+      "grad_norm": 1.7959818840026855,
+      "learning_rate": 1.417961727995254e-06,
+      "loss": 0.4604,
+      "step": 345
+    },
+    {
+      "epoch": 2.81799591002045,
+      "grad_norm": 1.414788842201233,
+      "learning_rate": 1.3072806164283358e-06,
+      "loss": 0.3398,
+      "step": 346
+    },
+    {
+      "epoch": 2.8261758691206547,
+      "grad_norm": 1.316179633140564,
+      "learning_rate": 1.2010393587603974e-06,
+      "loss": 0.3707,
+      "step": 347
+    },
+    {
+      "epoch": 2.834355828220859,
+      "grad_norm": 2.140214443206787,
+      "learning_rate": 1.099247642182205e-06,
+      "loss": 0.6991,
+      "step": 348
+    },
+    {
+      "epoch": 2.8425357873210633,
+      "grad_norm": 1.6871670484542847,
+      "learning_rate": 1.0019147481706625e-06,
+      "loss": 0.6069,
+      "step": 349
+    },
+    {
+      "epoch": 2.850715746421268,
+      "grad_norm": 1.7937754392623901,
+      "learning_rate": 9.090495516424713e-07,
+      "loss": 0.3841,
+      "step": 350
+    },
+    {
+      "epoch": 2.8588957055214723,
+      "grad_norm": 1.3567599058151245,
+      "learning_rate": 8.206605201449447e-07,
+      "loss": 0.3186,
+      "step": 351
+    },
+    {
+      "epoch": 2.8670756646216766,
+      "grad_norm": 1.0344539880752563,
+      "learning_rate": 7.36755713083892e-07,
+      "loss": 0.1676,
+      "step": 352
+    },
+    {
+      "epoch": 2.8752556237218814,
+      "grad_norm": 1.908477783203125,
+      "learning_rate": 6.573427809888067e-07,
+      "loss": 0.7295,
+      "step": 353
+    },
+    {
+      "epoch": 2.883435582822086,
+      "grad_norm": 2.14554500579834,
+      "learning_rate": 5.824289648152126e-07,
+      "loss": 0.8187,
+      "step": 354
+    },
+    {
+      "epoch": 2.8916155419222904,
+      "grad_norm": 1.6814268827438354,
+      "learning_rate": 5.120210952844872e-07,
+      "loss": 0.5205,
+      "step": 355
+    },
+    {
+      "epoch": 2.8997955010224947,
+      "grad_norm": 1.6498082876205444,
+      "learning_rate": 4.461255922609986e-07,
+      "loss": 0.4557,
+      "step": 356
+    },
+    {
+      "epoch": 2.9079754601226995,
+      "grad_norm": 1.4337708950042725,
+      "learning_rate": 3.8474846416672874e-07,
+      "loss": 0.3251,
+      "step": 357
+    },
+    {
+      "epoch": 2.9161554192229038,
+      "grad_norm": 1.875313401222229,
+      "learning_rate": 3.278953074334512e-07,
+      "loss": 0.5001,
+      "step": 358
+    },
+    {
+      "epoch": 2.9243353783231085,
+      "grad_norm": 1.350846529006958,
+      "learning_rate": 2.75571305992417e-07,
+      "loss": 0.2414,
+      "step": 359
+    },
+    {
+      "epoch": 2.932515337423313,
+      "grad_norm": 1.5978336334228516,
+      "learning_rate": 2.2778123080167135e-07,
+      "loss": 0.4585,
+      "step": 360
+    },
+    {
+      "epoch": 2.9406952965235176,
+      "grad_norm": 1.672541856765747,
+      "learning_rate": 1.8452943941106859e-07,
+      "loss": 0.5382,
+      "step": 361
+    },
+    {
+      "epoch": 2.948875255623722,
+      "grad_norm": 1.3987324237823486,
+      "learning_rate": 1.4581987556490095e-07,
+      "loss": 0.3326,
+      "step": 362
+    },
+    {
+      "epoch": 2.957055214723926,
+      "grad_norm": 1.4565430879592896,
+      "learning_rate": 1.1165606884234181e-07,
+      "loss": 0.4901,
+      "step": 363
+    },
+    {
+      "epoch": 2.965235173824131,
+      "grad_norm": 1.3861486911773682,
+      "learning_rate": 8.204113433559201e-08,
+      "loss": 0.2756,
+      "step": 364
+    },
+    {
+      "epoch": 2.9734151329243352,
+      "grad_norm": 1.4839295148849487,
+      "learning_rate": 5.697777236585711e-08,
+      "loss": 0.3303,
+      "step": 365
+    },
+    {
+      "epoch": 2.98159509202454,
+      "grad_norm": 1.6138904094696045,
+      "learning_rate": 3.6468268237105366e-08,
+      "loss": 0.558,
+      "step": 366
+    }
+  ],
+  "logging_steps": 1,
+  "max_steps": 366,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 3,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 3.142935032247091e+16,
+  "train_batch_size": 2,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-366/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2d8fa4ccfbbafc0e16806b535f96bbe57bf08f0c48d862e66bc54bb36f8a41a4
+size 5688

preprocessor_config.json

@@ -0,0 +1,31 @@
+{
+  "chunk_length": 300,
+  "dither": 0.0,
+  "feature_extractor_type": "WhisperFeatureExtractor",
+  "feature_size": 128,
+  "hop_length": 160,
+  "image_mean": [
+    0.48145466,
+    0.4578275,
+    0.40821073
+  ],
+  "image_processor_type": "Qwen2VLImageProcessor",
+  "image_std": [
+    0.26862954,
+    0.26130258,
+    0.27577711
+  ],
+  "max_pixels": 12845056,
+  "merge_size": 2,
+  "min_pixels": 3136,
+  "n_fft": 400,
+  "n_samples": 4800000,
+  "nb_max_frames": 30000,
+  "padding_side": "right",
+  "padding_value": 0.0,
+  "patch_size": 14,
+  "processor_class": "Qwen2_5OmniProcessor",
+  "return_attention_mask": true,
+  "sampling_rate": 16000,
+  "temporal_patch_size": 2
+}

runs/May06_06-58-26_main1/events.out.tfevents.1746514759.main1.1186.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f262264ad9e5027274f59f6b61f89ab3688ba2d38b69579f4ce7e4d8fd134054
+size 84385

special_tokens_map.json

@@ -0,0 +1,38 @@
+{
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|AUDIO|>",
+    "<|audio_bos|>",
+    "<|audio_eos|>",
+    "<|box_end|>",
+    "<|quad_start|>",
+    "<|quad_end|>",
+    "<|vision_bos|>",
+    "<|vision_eos|>",
+    "<|vision_pad|>",
+    "<|IMAGE|>",
+    "<|VIDEO|>"
+  ],
+  "audio_bos_token": "<|audio_bos|>",
+  "audio_eos_token": "<|audio_eos|>",
+  "audio_token": "<|AUDIO|>",
+  "eos_token": {
+    "content": "<|im_end|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "image_token": "<|IMAGE|>",
+  "pad_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "video_token": "<|VIDEO|>",
+  "vision_bos_token": "<|vision_bos|>",
+  "vision_eos_token": "<|vision_eos|>"
+}

tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8441917e39ae0244e06d704b95b3124795cec478e297f9afac39ba670d7e9d99
+size 11421870

tokenizer_config.json

@@ -0,0 +1,223 @@
+{
+  "add_prefix_space": false,
+  "added_tokens_decoder": {
+    "151643": {
+      "content": "<|endoftext|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151644": {
+      "content": "<|im_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151645": {
+      "content": "<|im_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151646": {
+      "content": "<|AUDIO|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151647": {
+      "content": "<|audio_bos|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151648": {
+      "content": "<|audio_eos|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151649": {
+      "content": "<|box_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151650": {
+      "content": "<|quad_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151651": {
+      "content": "<|quad_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151652": {
+      "content": "<|vision_bos|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151653": {
+      "content": "<|vision_eos|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151654": {
+      "content": "<|vision_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151655": {
+      "content": "<|IMAGE|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151656": {
+      "content": "<|VIDEO|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151657": {
+      "content": "<tool_call>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151658": {
+      "content": "</tool_call>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151659": {
+      "content": "<|fim_prefix|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151660": {
+      "content": "<|fim_middle|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151661": {
+      "content": "<|fim_suffix|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151662": {
+      "content": "<|fim_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151663": {
+      "content": "<|repo_name|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151664": {
+      "content": "<|file_sep|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    }
+  },
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|AUDIO|>",
+    "<|audio_bos|>",
+    "<|audio_eos|>",
+    "<|box_end|>",
+    "<|quad_start|>",
+    "<|quad_end|>",
+    "<|vision_bos|>",
+    "<|vision_eos|>",
+    "<|vision_pad|>",
+    "<|IMAGE|>",
+    "<|VIDEO|>"
+  ],
+  "audio_bos_token": "<|audio_bos|>",
+  "audio_eos_token": "<|audio_eos|>",
+  "audio_token": "<|AUDIO|>",
+  "bos_token": null,
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|im_end|>",
+  "errors": "replace",
+  "extra_special_tokens": {
+    "audio_bos_token": "<|audio_bos|>",
+    "audio_eos_token": "<|audio_eos|>",
+    "audio_token": "<|AUDIO|>",
+    "image_token": "<|IMAGE|>",
+    "video_token": "<|VIDEO|>",
+    "vision_bos_token": "<|vision_bos|>",
+    "vision_eos_token": "<|vision_eos|>"
+  },
+  "image_token": "<|IMAGE|>",
+  "model_max_length": 32768,
+  "pad_token": "<|endoftext|>",
+  "padding_side": "right",
+  "processor_class": "Qwen2_5OmniProcessor",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null,
+  "video_token": "<|VIDEO|>",
+  "vision_bos_token": "<|vision_bos|>",
+  "vision_eos_token": "<|vision_eos|>"
+}

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 2.98159509202454,
+    "total_flos": 3.142935032247091e+16,
+    "train_loss": 0.7803121163341843,
+    "train_runtime": 701.3998,
+    "train_samples_per_second": 4.183,
+    "train_steps_per_second": 0.522
+}

trainer_log.jsonl

@@ -0,0 +1,367 @@
+{"current_steps": 1, "total_steps": 366, "loss": 1.978, "lr": 0.0, "epoch": 0.0081799591002045, "percentage": 0.27, "elapsed_time": "0:00:05", "remaining_time": "0:31:02"}
+{"current_steps": 2, "total_steps": 366, "loss": 2.0417, "lr": 0.0, "epoch": 0.016359918200409, "percentage": 0.55, "elapsed_time": "0:00:07", "remaining_time": "0:21:28"}
+{"current_steps": 3, "total_steps": 366, "loss": 2.4526, "lr": 0.0, "epoch": 0.024539877300613498, "percentage": 0.82, "elapsed_time": "0:00:09", "remaining_time": "0:19:00"}
+{"current_steps": 4, "total_steps": 366, "loss": 2.3708, "lr": 2.702702702702703e-06, "epoch": 0.032719836400818, "percentage": 1.09, "elapsed_time": "0:00:11", "remaining_time": "0:17:07"}
+{"current_steps": 5, "total_steps": 366, "loss": 2.7402, "lr": 5.405405405405406e-06, "epoch": 0.0408997955010225, "percentage": 1.37, "elapsed_time": "0:00:13", "remaining_time": "0:16:00"}
+{"current_steps": 6, "total_steps": 366, "loss": 2.6255, "lr": 5.405405405405406e-06, "epoch": 0.049079754601226995, "percentage": 1.64, "elapsed_time": "0:00:15", "remaining_time": "0:15:18"}
+{"current_steps": 7, "total_steps": 366, "loss": 2.1672, "lr": 8.108108108108109e-06, "epoch": 0.05725971370143149, "percentage": 1.91, "elapsed_time": "0:00:17", "remaining_time": "0:14:54"}
+{"current_steps": 8, "total_steps": 366, "loss": 2.6853, "lr": 1.0810810810810812e-05, "epoch": 0.065439672801636, "percentage": 2.19, "elapsed_time": "0:00:19", "remaining_time": "0:14:23"}
+{"current_steps": 9, "total_steps": 366, "loss": 2.5256, "lr": 1.3513513513513515e-05, "epoch": 0.0736196319018405, "percentage": 2.46, "elapsed_time": "0:00:21", "remaining_time": "0:14:00"}
+{"current_steps": 10, "total_steps": 366, "loss": 2.1108, "lr": 1.6216216216216218e-05, "epoch": 0.081799591002045, "percentage": 2.73, "elapsed_time": "0:00:23", "remaining_time": "0:13:42"}
+{"current_steps": 11, "total_steps": 366, "loss": 2.2621, "lr": 1.891891891891892e-05, "epoch": 0.08997955010224949, "percentage": 3.01, "elapsed_time": "0:00:24", "remaining_time": "0:13:25"}
+{"current_steps": 12, "total_steps": 366, "loss": 2.2973, "lr": 2.1621621621621624e-05, "epoch": 0.09815950920245399, "percentage": 3.28, "elapsed_time": "0:00:26", "remaining_time": "0:13:10"}
+{"current_steps": 13, "total_steps": 366, "loss": 2.2853, "lr": 2.4324324324324327e-05, "epoch": 0.10633946830265849, "percentage": 3.55, "elapsed_time": "0:00:28", "remaining_time": "0:12:57"}
+{"current_steps": 14, "total_steps": 366, "loss": 2.3628, "lr": 2.702702702702703e-05, "epoch": 0.11451942740286299, "percentage": 3.83, "elapsed_time": "0:00:30", "remaining_time": "0:12:48"}
+{"current_steps": 15, "total_steps": 366, "loss": 1.6333, "lr": 2.9729729729729733e-05, "epoch": 0.12269938650306748, "percentage": 4.1, "elapsed_time": "0:00:32", "remaining_time": "0:12:39"}
+{"current_steps": 16, "total_steps": 366, "loss": 1.6592, "lr": 3.2432432432432436e-05, "epoch": 0.130879345603272, "percentage": 4.37, "elapsed_time": "0:00:34", "remaining_time": "0:12:31"}
+{"current_steps": 17, "total_steps": 366, "loss": 2.4303, "lr": 3.513513513513514e-05, "epoch": 0.1390593047034765, "percentage": 4.64, "elapsed_time": "0:00:36", "remaining_time": "0:12:25"}
+{"current_steps": 18, "total_steps": 366, "loss": 1.6276, "lr": 3.783783783783784e-05, "epoch": 0.147239263803681, "percentage": 4.92, "elapsed_time": "0:00:38", "remaining_time": "0:12:19"}
+{"current_steps": 19, "total_steps": 366, "loss": 2.1786, "lr": 4.0540540540540545e-05, "epoch": 0.1554192229038855, "percentage": 5.19, "elapsed_time": "0:00:40", "remaining_time": "0:12:12"}
+{"current_steps": 20, "total_steps": 366, "loss": 1.7335, "lr": 4.324324324324325e-05, "epoch": 0.16359918200409, "percentage": 5.46, "elapsed_time": "0:00:42", "remaining_time": "0:12:07"}
+{"current_steps": 21, "total_steps": 366, "loss": 1.9214, "lr": 4.594594594594595e-05, "epoch": 0.17177914110429449, "percentage": 5.74, "elapsed_time": "0:00:43", "remaining_time": "0:12:01"}
+{"current_steps": 22, "total_steps": 366, "loss": 1.9089, "lr": 4.8648648648648654e-05, "epoch": 0.17995910020449898, "percentage": 6.01, "elapsed_time": "0:00:45", "remaining_time": "0:11:56"}
+{"current_steps": 23, "total_steps": 366, "loss": 2.092, "lr": 5.135135135135135e-05, "epoch": 0.18813905930470348, "percentage": 6.28, "elapsed_time": "0:00:47", "remaining_time": "0:11:50"}
+{"current_steps": 24, "total_steps": 366, "loss": 1.9184, "lr": 5.405405405405406e-05, "epoch": 0.19631901840490798, "percentage": 6.56, "elapsed_time": "0:00:49", "remaining_time": "0:11:47"}
+{"current_steps": 25, "total_steps": 366, "loss": 1.6762, "lr": 5.6756756756756757e-05, "epoch": 0.20449897750511248, "percentage": 6.83, "elapsed_time": "0:00:51", "remaining_time": "0:11:42"}
+{"current_steps": 26, "total_steps": 366, "loss": 1.8447, "lr": 5.9459459459459466e-05, "epoch": 0.21267893660531698, "percentage": 7.1, "elapsed_time": "0:00:53", "remaining_time": "0:11:38"}
+{"current_steps": 27, "total_steps": 366, "loss": 1.659, "lr": 6.216216216216216e-05, "epoch": 0.22085889570552147, "percentage": 7.38, "elapsed_time": "0:00:55", "remaining_time": "0:11:35"}
+{"current_steps": 28, "total_steps": 366, "loss": 1.5518, "lr": 6.486486486486487e-05, "epoch": 0.22903885480572597, "percentage": 7.65, "elapsed_time": "0:00:57", "remaining_time": "0:11:33"}
+{"current_steps": 29, "total_steps": 366, "loss": 1.7725, "lr": 6.756756756756757e-05, "epoch": 0.23721881390593047, "percentage": 7.92, "elapsed_time": "0:00:59", "remaining_time": "0:11:29"}
+{"current_steps": 30, "total_steps": 366, "loss": 1.3214, "lr": 7.027027027027028e-05, "epoch": 0.24539877300613497, "percentage": 8.2, "elapsed_time": "0:01:01", "remaining_time": "0:11:24"}
+{"current_steps": 31, "total_steps": 366, "loss": 1.5599, "lr": 7.297297297297297e-05, "epoch": 0.25357873210633947, "percentage": 8.47, "elapsed_time": "0:01:03", "remaining_time": "0:11:21"}
+{"current_steps": 32, "total_steps": 366, "loss": 1.4007, "lr": 7.567567567567568e-05, "epoch": 0.261758691206544, "percentage": 8.74, "elapsed_time": "0:01:04", "remaining_time": "0:11:17"}
+{"current_steps": 33, "total_steps": 366, "loss": 1.4209, "lr": 7.837837837837838e-05, "epoch": 0.26993865030674846, "percentage": 9.02, "elapsed_time": "0:01:06", "remaining_time": "0:11:13"}
+{"current_steps": 34, "total_steps": 366, "loss": 1.1963, "lr": 8.108108108108109e-05, "epoch": 0.278118609406953, "percentage": 9.29, "elapsed_time": "0:01:08", "remaining_time": "0:11:10"}
+{"current_steps": 35, "total_steps": 366, "loss": 1.6071, "lr": 8.378378378378379e-05, "epoch": 0.28629856850715746, "percentage": 9.56, "elapsed_time": "0:01:10", "remaining_time": "0:11:07"}
+{"current_steps": 36, "total_steps": 366, "loss": 1.1902, "lr": 8.64864864864865e-05, "epoch": 0.294478527607362, "percentage": 9.84, "elapsed_time": "0:01:12", "remaining_time": "0:11:04"}
+{"current_steps": 37, "total_steps": 366, "loss": 1.1838, "lr": 8.918918918918919e-05, "epoch": 0.30265848670756645, "percentage": 10.11, "elapsed_time": "0:01:14", "remaining_time": "0:11:01"}
+{"current_steps": 38, "total_steps": 366, "loss": 0.9948, "lr": 9.18918918918919e-05, "epoch": 0.310838445807771, "percentage": 10.38, "elapsed_time": "0:01:16", "remaining_time": "0:10:58"}
+{"current_steps": 39, "total_steps": 366, "loss": 1.0997, "lr": 9.45945945945946e-05, "epoch": 0.31901840490797545, "percentage": 10.66, "elapsed_time": "0:01:18", "remaining_time": "0:10:55"}
+{"current_steps": 40, "total_steps": 366, "loss": 0.6781, "lr": 9.729729729729731e-05, "epoch": 0.32719836400818, "percentage": 10.93, "elapsed_time": "0:01:20", "remaining_time": "0:10:52"}
+{"current_steps": 41, "total_steps": 366, "loss": 1.4761, "lr": 0.0001, "epoch": 0.33537832310838445, "percentage": 11.2, "elapsed_time": "0:01:21", "remaining_time": "0:10:49"}
+{"current_steps": 42, "total_steps": 366, "loss": 1.0233, "lr": 9.99977204734326e-05, "epoch": 0.34355828220858897, "percentage": 11.48, "elapsed_time": "0:01:24", "remaining_time": "0:10:48"}
+{"current_steps": 43, "total_steps": 366, "loss": 0.8199, "lr": 9.999088210158001e-05, "epoch": 0.35173824130879344, "percentage": 11.75, "elapsed_time": "0:01:25", "remaining_time": "0:10:45"}
+{"current_steps": 44, "total_steps": 366, "loss": 1.1087, "lr": 9.997948550797227e-05, "epoch": 0.35991820040899797, "percentage": 12.02, "elapsed_time": "0:01:27", "remaining_time": "0:10:42"}
+{"current_steps": 45, "total_steps": 366, "loss": 0.8149, "lr": 9.996353173176289e-05, "epoch": 0.36809815950920244, "percentage": 12.3, "elapsed_time": "0:01:29", "remaining_time": "0:10:39"}
+{"current_steps": 46, "total_steps": 366, "loss": 0.9549, "lr": 9.994302222763414e-05, "epoch": 0.37627811860940696, "percentage": 12.57, "elapsed_time": "0:01:31", "remaining_time": "0:10:36"}
+{"current_steps": 47, "total_steps": 366, "loss": 0.9542, "lr": 9.991795886566441e-05, "epoch": 0.38445807770961143, "percentage": 12.84, "elapsed_time": "0:01:33", "remaining_time": "0:10:33"}
+{"current_steps": 48, "total_steps": 366, "loss": 0.8864, "lr": 9.988834393115767e-05, "epoch": 0.39263803680981596, "percentage": 13.11, "elapsed_time": "0:01:35", "remaining_time": "0:10:30"}
+{"current_steps": 49, "total_steps": 366, "loss": 1.0826, "lr": 9.98541801244351e-05, "epoch": 0.40081799591002043, "percentage": 13.39, "elapsed_time": "0:01:37", "remaining_time": "0:10:28"}
+{"current_steps": 50, "total_steps": 366, "loss": 0.9114, "lr": 9.981547056058893e-05, "epoch": 0.40899795501022496, "percentage": 13.66, "elapsed_time": "0:01:38", "remaining_time": "0:10:24"}
+{"current_steps": 51, "total_steps": 366, "loss": 0.9755, "lr": 9.977221876919833e-05, "epoch": 0.4171779141104294, "percentage": 13.93, "elapsed_time": "0:01:40", "remaining_time": "0:10:22"}
+{"current_steps": 52, "total_steps": 366, "loss": 0.741, "lr": 9.972442869400759e-05, "epoch": 0.42535787321063395, "percentage": 14.21, "elapsed_time": "0:01:42", "remaining_time": "0:10:20"}
+{"current_steps": 53, "total_steps": 366, "loss": 1.0441, "lr": 9.967210469256656e-05, "epoch": 0.4335378323108384, "percentage": 14.48, "elapsed_time": "0:01:44", "remaining_time": "0:10:17"}
+{"current_steps": 54, "total_steps": 366, "loss": 1.2885, "lr": 9.961525153583327e-05, "epoch": 0.44171779141104295, "percentage": 14.75, "elapsed_time": "0:01:46", "remaining_time": "0:10:15"}
+{"current_steps": 55, "total_steps": 366, "loss": 0.8107, "lr": 9.9553874407739e-05, "epoch": 0.4498977505112474, "percentage": 15.03, "elapsed_time": "0:01:48", "remaining_time": "0:10:12"}
+{"current_steps": 56, "total_steps": 366, "loss": 0.6129, "lr": 9.948797890471551e-05, "epoch": 0.45807770961145194, "percentage": 15.3, "elapsed_time": "0:01:50", "remaining_time": "0:10:10"}
+{"current_steps": 57, "total_steps": 366, "loss": 0.9262, "lr": 9.941757103518478e-05, "epoch": 0.4662576687116564, "percentage": 15.57, "elapsed_time": "0:01:52", "remaining_time": "0:10:07"}
+{"current_steps": 58, "total_steps": 366, "loss": 0.7552, "lr": 9.93426572190112e-05, "epoch": 0.47443762781186094, "percentage": 15.85, "elapsed_time": "0:01:53", "remaining_time": "0:10:05"}
+{"current_steps": 59, "total_steps": 366, "loss": 0.6346, "lr": 9.926324428691611e-05, "epoch": 0.48261758691206547, "percentage": 16.12, "elapsed_time": "0:01:56", "remaining_time": "0:10:03"}
+{"current_steps": 60, "total_steps": 366, "loss": 0.8199, "lr": 9.917933947985507e-05, "epoch": 0.49079754601226994, "percentage": 16.39, "elapsed_time": "0:01:57", "remaining_time": "0:10:01"}
+{"current_steps": 61, "total_steps": 366, "loss": 0.7485, "lr": 9.909095044835754e-05, "epoch": 0.49897750511247446, "percentage": 16.67, "elapsed_time": "0:01:59", "remaining_time": "0:09:58"}
+{"current_steps": 62, "total_steps": 366, "loss": 1.1649, "lr": 9.899808525182935e-05, "epoch": 0.5071574642126789, "percentage": 16.94, "elapsed_time": "0:02:01", "remaining_time": "0:09:56"}
+{"current_steps": 63, "total_steps": 366, "loss": 1.1159, "lr": 9.890075235781779e-05, "epoch": 0.5153374233128835, "percentage": 17.21, "elapsed_time": "0:02:03", "remaining_time": "0:09:54"}
+{"current_steps": 64, "total_steps": 366, "loss": 0.9613, "lr": 9.879896064123961e-05, "epoch": 0.523517382413088, "percentage": 17.49, "elapsed_time": "0:02:05", "remaining_time": "0:09:51"}
+{"current_steps": 65, "total_steps": 366, "loss": 1.047, "lr": 9.869271938357167e-05, "epoch": 0.5316973415132924, "percentage": 17.76, "elapsed_time": "0:02:07", "remaining_time": "0:09:49"}
+{"current_steps": 66, "total_steps": 366, "loss": 1.1846, "lr": 9.858203827200476e-05, "epoch": 0.5398773006134969, "percentage": 18.03, "elapsed_time": "0:02:09", "remaining_time": "0:09:47"}
+{"current_steps": 67, "total_steps": 366, "loss": 0.9566, "lr": 9.846692739856024e-05, "epoch": 0.5480572597137015, "percentage": 18.31, "elapsed_time": "0:02:11", "remaining_time": "0:09:44"}
+{"current_steps": 68, "total_steps": 366, "loss": 1.1108, "lr": 9.834739725916988e-05, "epoch": 0.556237218813906, "percentage": 18.58, "elapsed_time": "0:02:12", "remaining_time": "0:09:42"}
+{"current_steps": 69, "total_steps": 366, "loss": 1.1265, "lr": 9.822345875271883e-05, "epoch": 0.5644171779141104, "percentage": 18.85, "elapsed_time": "0:02:14", "remaining_time": "0:09:40"}
+{"current_steps": 70, "total_steps": 366, "loss": 0.7757, "lr": 9.809512318005181e-05, "epoch": 0.5725971370143149, "percentage": 19.13, "elapsed_time": "0:02:16", "remaining_time": "0:09:38"}
+{"current_steps": 71, "total_steps": 366, "loss": 0.9006, "lr": 9.796240224294271e-05, "epoch": 0.5807770961145194, "percentage": 19.4, "elapsed_time": "0:02:18", "remaining_time": "0:09:36"}
+{"current_steps": 72, "total_steps": 366, "loss": 0.9322, "lr": 9.782530804302763e-05, "epoch": 0.588957055214724, "percentage": 19.67, "elapsed_time": "0:02:20", "remaining_time": "0:09:33"}
+{"current_steps": 73, "total_steps": 366, "loss": 0.8403, "lr": 9.768385308070138e-05, "epoch": 0.5971370143149284, "percentage": 19.95, "elapsed_time": "0:02:22", "remaining_time": "0:09:31"}
+{"current_steps": 74, "total_steps": 366, "loss": 0.8397, "lr": 9.753805025397779e-05, "epoch": 0.6053169734151329, "percentage": 20.22, "elapsed_time": "0:02:24", "remaining_time": "0:09:29"}
+{"current_steps": 75, "total_steps": 366, "loss": 0.8143, "lr": 9.738791285731352e-05, "epoch": 0.6134969325153374, "percentage": 20.49, "elapsed_time": "0:02:26", "remaining_time": "0:09:26"}
+{"current_steps": 76, "total_steps": 366, "loss": 1.059, "lr": 9.723345458039594e-05, "epoch": 0.621676891615542, "percentage": 20.77, "elapsed_time": "0:02:28", "remaining_time": "0:09:24"}
+{"current_steps": 77, "total_steps": 366, "loss": 0.9112, "lr": 9.707468950689491e-05, "epoch": 0.6298568507157464, "percentage": 21.04, "elapsed_time": "0:02:29", "remaining_time": "0:09:22"}
+{"current_steps": 78, "total_steps": 366, "loss": 0.7847, "lr": 9.691163211317853e-05, "epoch": 0.6380368098159509, "percentage": 21.31, "elapsed_time": "0:02:31", "remaining_time": "0:09:20"}
+{"current_steps": 79, "total_steps": 366, "loss": 0.9806, "lr": 9.674429726699323e-05, "epoch": 0.6462167689161554, "percentage": 21.58, "elapsed_time": "0:02:33", "remaining_time": "0:09:18"}
+{"current_steps": 80, "total_steps": 366, "loss": 0.6648, "lr": 9.657270022610813e-05, "epoch": 0.65439672801636, "percentage": 21.86, "elapsed_time": "0:02:35", "remaining_time": "0:09:16"}
+{"current_steps": 81, "total_steps": 366, "loss": 0.9702, "lr": 9.63968566369238e-05, "epoch": 0.6625766871165644, "percentage": 22.13, "elapsed_time": "0:02:37", "remaining_time": "0:09:14"}
+{"current_steps": 82, "total_steps": 366, "loss": 0.7861, "lr": 9.62167825330455e-05, "epoch": 0.6707566462167689, "percentage": 22.4, "elapsed_time": "0:02:39", "remaining_time": "0:09:11"}
+{"current_steps": 83, "total_steps": 366, "loss": 1.1252, "lr": 9.603249433382144e-05, "epoch": 0.6789366053169734, "percentage": 22.68, "elapsed_time": "0:02:41", "remaining_time": "0:09:10"}
+{"current_steps": 84, "total_steps": 366, "loss": 0.7109, "lr": 9.584400884284545e-05, "epoch": 0.6871165644171779, "percentage": 22.95, "elapsed_time": "0:02:43", "remaining_time": "0:09:08"}
+{"current_steps": 85, "total_steps": 366, "loss": 0.6421, "lr": 9.56513432464249e-05, "epoch": 0.6952965235173824, "percentage": 23.22, "elapsed_time": "0:02:45", "remaining_time": "0:09:06"}
+{"current_steps": 86, "total_steps": 366, "loss": 0.6337, "lr": 9.545451511201364e-05, "epoch": 0.7034764826175869, "percentage": 23.5, "elapsed_time": "0:02:47", "remaining_time": "0:09:04"}
+{"current_steps": 87, "total_steps": 366, "loss": 1.0757, "lr": 9.525354238661009e-05, "epoch": 0.7116564417177914, "percentage": 23.77, "elapsed_time": "0:02:49", "remaining_time": "0:09:01"}
+{"current_steps": 88, "total_steps": 366, "loss": 0.7056, "lr": 9.504844339512095e-05, "epoch": 0.7198364008179959, "percentage": 24.04, "elapsed_time": "0:02:50", "remaining_time": "0:08:59"}
+{"current_steps": 89, "total_steps": 366, "loss": 0.8767, "lr": 9.483923683869024e-05, "epoch": 0.7280163599182005, "percentage": 24.32, "elapsed_time": "0:02:52", "remaining_time": "0:08:57"}
+{"current_steps": 90, "total_steps": 366, "loss": 0.922, "lr": 9.462594179299406e-05, "epoch": 0.7361963190184049, "percentage": 24.59, "elapsed_time": "0:02:54", "remaining_time": "0:08:55"}
+{"current_steps": 91, "total_steps": 366, "loss": 0.8301, "lr": 9.440857770650138e-05, "epoch": 0.7443762781186094, "percentage": 24.86, "elapsed_time": "0:02:56", "remaining_time": "0:08:53"}
+{"current_steps": 92, "total_steps": 366, "loss": 0.5308, "lr": 9.418716439870057e-05, "epoch": 0.7525562372188139, "percentage": 25.14, "elapsed_time": "0:02:58", "remaining_time": "0:08:51"}
+{"current_steps": 93, "total_steps": 366, "loss": 1.1116, "lr": 9.396172205829234e-05, "epoch": 0.7607361963190185, "percentage": 25.41, "elapsed_time": "0:03:00", "remaining_time": "0:08:48"}
+{"current_steps": 94, "total_steps": 366, "loss": 0.8806, "lr": 9.373227124134888e-05, "epoch": 0.7689161554192229, "percentage": 25.68, "elapsed_time": "0:03:02", "remaining_time": "0:08:46"}
+{"current_steps": 95, "total_steps": 366, "loss": 0.7924, "lr": 9.34988328694395e-05, "epoch": 0.7770961145194274, "percentage": 25.96, "elapsed_time": "0:03:03", "remaining_time": "0:08:44"}
+{"current_steps": 96, "total_steps": 366, "loss": 0.7091, "lr": 9.326142822772302e-05, "epoch": 0.7852760736196319, "percentage": 26.23, "elapsed_time": "0:03:05", "remaining_time": "0:08:42"}
+{"current_steps": 97, "total_steps": 366, "loss": 0.8828, "lr": 9.302007896300698e-05, "epoch": 0.7934560327198364, "percentage": 26.5, "elapsed_time": "0:03:07", "remaining_time": "0:08:40"}
+{"current_steps": 98, "total_steps": 366, "loss": 1.0531, "lr": 9.27748070817738e-05, "epoch": 0.8016359918200409, "percentage": 26.78, "elapsed_time": "0:03:09", "remaining_time": "0:08:38"}
+{"current_steps": 99, "total_steps": 366, "loss": 0.9222, "lr": 9.252563494817425e-05, "epoch": 0.8098159509202454, "percentage": 27.05, "elapsed_time": "0:03:11", "remaining_time": "0:08:36"}
+{"current_steps": 100, "total_steps": 366, "loss": 0.8988, "lr": 9.227258528198831e-05, "epoch": 0.8179959100204499, "percentage": 27.32, "elapsed_time": "0:03:13", "remaining_time": "0:08:34"}
+{"current_steps": 101, "total_steps": 366, "loss": 1.0139, "lr": 9.201568115655342e-05, "epoch": 0.8261758691206544, "percentage": 27.6, "elapsed_time": "0:03:15", "remaining_time": "0:08:32"}
+{"current_steps": 102, "total_steps": 366, "loss": 0.6802, "lr": 9.175494599666077e-05, "epoch": 0.8343558282208589, "percentage": 27.87, "elapsed_time": "0:03:17", "remaining_time": "0:08:30"}
+{"current_steps": 103, "total_steps": 366, "loss": 0.8834, "lr": 9.149040357641929e-05, "epoch": 0.8425357873210634, "percentage": 28.14, "elapsed_time": "0:03:19", "remaining_time": "0:08:28"}
+{"current_steps": 104, "total_steps": 366, "loss": 0.6559, "lr": 9.122207801708802e-05, "epoch": 0.8507157464212679, "percentage": 28.42, "elapsed_time": "0:03:20", "remaining_time": "0:08:26"}
+{"current_steps": 105, "total_steps": 366, "loss": 0.7427, "lr": 9.094999378487659e-05, "epoch": 0.8588957055214724, "percentage": 28.69, "elapsed_time": "0:03:22", "remaining_time": "0:08:24"}
+{"current_steps": 106, "total_steps": 366, "loss": 0.6253, "lr": 9.067417568871445e-05, "epoch": 0.8670756646216768, "percentage": 28.96, "elapsed_time": "0:03:24", "remaining_time": "0:08:22"}
+{"current_steps": 107, "total_steps": 366, "loss": 0.7601, "lr": 9.03946488779887e-05, "epoch": 0.8752556237218814, "percentage": 29.23, "elapsed_time": "0:03:26", "remaining_time": "0:08:20"}
+{"current_steps": 108, "total_steps": 366, "loss": 1.024, "lr": 9.011143884025101e-05, "epoch": 0.8834355828220859, "percentage": 29.51, "elapsed_time": "0:03:28", "remaining_time": "0:08:18"}
+{"current_steps": 109, "total_steps": 366, "loss": 0.8283, "lr": 8.982457139889357e-05, "epoch": 0.8916155419222904, "percentage": 29.78, "elapsed_time": "0:03:30", "remaining_time": "0:08:15"}
+{"current_steps": 110, "total_steps": 366, "loss": 0.8297, "lr": 8.953407271079455e-05, "epoch": 0.8997955010224948, "percentage": 30.05, "elapsed_time": "0:03:32", "remaining_time": "0:08:13"}
+{"current_steps": 111, "total_steps": 366, "loss": 0.5966, "lr": 8.923996926393305e-05, "epoch": 0.9079754601226994, "percentage": 30.33, "elapsed_time": "0:03:34", "remaining_time": "0:08:12"}
+{"current_steps": 112, "total_steps": 366, "loss": 0.7775, "lr": 8.894228787497389e-05, "epoch": 0.9161554192229039, "percentage": 30.6, "elapsed_time": "0:03:36", "remaining_time": "0:08:10"}
+{"current_steps": 113, "total_steps": 366, "loss": 0.7838, "lr": 8.864105568682244e-05, "epoch": 0.9243353783231084, "percentage": 30.87, "elapsed_time": "0:03:38", "remaining_time": "0:08:08"}
+{"current_steps": 114, "total_steps": 366, "loss": 0.5921, "lr": 8.833630016614976e-05, "epoch": 0.9325153374233128, "percentage": 31.15, "elapsed_time": "0:03:39", "remaining_time": "0:08:06"}
+{"current_steps": 115, "total_steps": 366, "loss": 1.0578, "lr": 8.802804910088809e-05, "epoch": 0.9406952965235174, "percentage": 31.42, "elapsed_time": "0:03:41", "remaining_time": "0:08:04"}
+{"current_steps": 116, "total_steps": 366, "loss": 0.8205, "lr": 8.771633059769711e-05, "epoch": 0.9488752556237219, "percentage": 31.69, "elapsed_time": "0:03:43", "remaining_time": "0:08:02"}
+{"current_steps": 117, "total_steps": 366, "loss": 0.8237, "lr": 8.740117307940123e-05, "epoch": 0.9570552147239264, "percentage": 31.97, "elapsed_time": "0:03:45", "remaining_time": "0:08:00"}
+{"current_steps": 118, "total_steps": 366, "loss": 1.0473, "lr": 8.708260528239788e-05, "epoch": 0.9652351738241309, "percentage": 32.24, "elapsed_time": "0:03:47", "remaining_time": "0:07:58"}
+{"current_steps": 119, "total_steps": 366, "loss": 1.0789, "lr": 8.676065625403733e-05, "epoch": 0.9734151329243353, "percentage": 32.51, "elapsed_time": "0:03:49", "remaining_time": "0:07:56"}
+{"current_steps": 120, "total_steps": 366, "loss": 0.5315, "lr": 8.64353553499741e-05, "epoch": 0.9815950920245399, "percentage": 32.79, "elapsed_time": "0:03:51", "remaining_time": "0:07:54"}
+{"current_steps": 121, "total_steps": 366, "loss": 0.9738, "lr": 8.610673223149034e-05, "epoch": 0.9897750511247444, "percentage": 33.06, "elapsed_time": "0:03:53", "remaining_time": "0:07:52"}
+{"current_steps": 122, "total_steps": 366, "loss": 0.6114, "lr": 8.577481686279123e-05, "epoch": 0.9979550102249489, "percentage": 33.33, "elapsed_time": "0:03:55", "remaining_time": "0:07:50"}
+{"current_steps": 123, "total_steps": 366, "loss": 0.1857, "lr": 8.543963950827279e-05, "epoch": 1.0, "percentage": 33.61, "elapsed_time": "0:03:55", "remaining_time": "0:07:45"}
+{"current_steps": 124, "total_steps": 366, "loss": 0.77, "lr": 8.510123072976239e-05, "epoch": 1.0081799591002045, "percentage": 33.88, "elapsed_time": "0:03:58", "remaining_time": "0:07:45"}
+{"current_steps": 125, "total_steps": 366, "loss": 0.588, "lr": 8.475962138373213e-05, "epoch": 1.016359918200409, "percentage": 34.15, "elapsed_time": "0:04:00", "remaining_time": "0:07:43"}
+{"current_steps": 126, "total_steps": 366, "loss": 0.8879, "lr": 8.441484261848514e-05, "epoch": 1.0245398773006136, "percentage": 34.43, "elapsed_time": "0:04:02", "remaining_time": "0:07:42"}
+{"current_steps": 127, "total_steps": 366, "loss": 1.1292, "lr": 8.406692587131568e-05, "epoch": 1.032719836400818, "percentage": 34.7, "elapsed_time": "0:04:04", "remaining_time": "0:07:40"}
+{"current_steps": 128, "total_steps": 366, "loss": 0.6589, "lr": 8.371590286564247e-05, "epoch": 1.0408997955010224, "percentage": 34.97, "elapsed_time": "0:04:06", "remaining_time": "0:07:38"}
+{"current_steps": 129, "total_steps": 366, "loss": 0.6267, "lr": 8.336180560811619e-05, "epoch": 1.049079754601227, "percentage": 35.25, "elapsed_time": "0:04:08", "remaining_time": "0:07:36"}
+{"current_steps": 130, "total_steps": 366, "loss": 0.7399, "lr": 8.30046663857011e-05, "epoch": 1.0572597137014315, "percentage": 35.52, "elapsed_time": "0:04:10", "remaining_time": "0:07:34"}
+{"current_steps": 131, "total_steps": 366, "loss": 0.8849, "lr": 8.264451776273104e-05, "epoch": 1.065439672801636, "percentage": 35.79, "elapsed_time": "0:04:12", "remaining_time": "0:07:32"}
+{"current_steps": 132, "total_steps": 366, "loss": 0.901, "lr": 8.228139257794012e-05, "epoch": 1.0736196319018405, "percentage": 36.07, "elapsed_time": "0:04:14", "remaining_time": "0:07:30"}
+{"current_steps": 133, "total_steps": 366, "loss": 0.3923, "lr": 8.191532394146865e-05, "epoch": 1.081799591002045, "percentage": 36.34, "elapsed_time": "0:04:15", "remaining_time": "0:07:28"}
+{"current_steps": 134, "total_steps": 366, "loss": 0.77, "lr": 8.154634523184388e-05, "epoch": 1.0899795501022496, "percentage": 36.61, "elapsed_time": "0:04:17", "remaining_time": "0:07:26"}
+{"current_steps": 135, "total_steps": 366, "loss": 0.7849, "lr": 8.117449009293668e-05, "epoch": 1.098159509202454, "percentage": 36.89, "elapsed_time": "0:04:19", "remaining_time": "0:07:24"}
+{"current_steps": 136, "total_steps": 366, "loss": 0.7912, "lr": 8.07997924308938e-05, "epoch": 1.1063394683026584, "percentage": 37.16, "elapsed_time": "0:04:21", "remaining_time": "0:07:22"}
+{"current_steps": 137, "total_steps": 366, "loss": 0.7142, "lr": 8.042228641104622e-05, "epoch": 1.114519427402863, "percentage": 37.43, "elapsed_time": "0:04:23", "remaining_time": "0:07:20"}
+{"current_steps": 138, "total_steps": 366, "loss": 0.5454, "lr": 8.004200645479403e-05, "epoch": 1.1226993865030674, "percentage": 37.7, "elapsed_time": "0:04:25", "remaining_time": "0:07:18"}
+{"current_steps": 139, "total_steps": 366, "loss": 0.7387, "lr": 7.965898723646776e-05, "epoch": 1.130879345603272, "percentage": 37.98, "elapsed_time": "0:04:27", "remaining_time": "0:07:16"}
+{"current_steps": 140, "total_steps": 366, "loss": 0.7271, "lr": 7.927326368016677e-05, "epoch": 1.1390593047034765, "percentage": 38.25, "elapsed_time": "0:04:29", "remaining_time": "0:07:14"}
+{"current_steps": 141, "total_steps": 366, "loss": 0.8452, "lr": 7.888487095657484e-05, "epoch": 1.147239263803681, "percentage": 38.52, "elapsed_time": "0:04:31", "remaining_time": "0:07:12"}
+{"current_steps": 142, "total_steps": 366, "loss": 0.5735, "lr": 7.849384447975321e-05, "epoch": 1.1554192229038855, "percentage": 38.8, "elapsed_time": "0:04:33", "remaining_time": "0:07:10"}
+{"current_steps": 143, "total_steps": 366, "loss": 0.486, "lr": 7.810021990391164e-05, "epoch": 1.16359918200409, "percentage": 39.07, "elapsed_time": "0:04:35", "remaining_time": "0:07:09"}
+{"current_steps": 144, "total_steps": 366, "loss": 0.9265, "lr": 7.770403312015721e-05, "epoch": 1.1717791411042944, "percentage": 39.34, "elapsed_time": "0:04:37", "remaining_time": "0:07:07"}
+{"current_steps": 145, "total_steps": 366, "loss": 0.6186, "lr": 7.73053202532219e-05, "epoch": 1.179959100204499, "percentage": 39.62, "elapsed_time": "0:04:39", "remaining_time": "0:07:05"}
+{"current_steps": 146, "total_steps": 366, "loss": 0.7012, "lr": 7.690411765816864e-05, "epoch": 1.1881390593047034, "percentage": 39.89, "elapsed_time": "0:04:40", "remaining_time": "0:07:03"}
+{"current_steps": 147, "total_steps": 366, "loss": 0.7644, "lr": 7.650046191707641e-05, "epoch": 1.196319018404908, "percentage": 40.16, "elapsed_time": "0:04:42", "remaining_time": "0:07:01"}
+{"current_steps": 148, "total_steps": 366, "loss": 0.614, "lr": 7.60943898357046e-05, "epoch": 1.2044989775051125, "percentage": 40.44, "elapsed_time": "0:04:44", "remaining_time": "0:06:59"}
+{"current_steps": 149, "total_steps": 366, "loss": 0.6722, "lr": 7.568593844013718e-05, "epoch": 1.212678936605317, "percentage": 40.71, "elapsed_time": "0:04:46", "remaining_time": "0:06:57"}
+{"current_steps": 150, "total_steps": 366, "loss": 0.6981, "lr": 7.527514497340642e-05, "epoch": 1.2208588957055215, "percentage": 40.98, "elapsed_time": "0:04:48", "remaining_time": "0:06:55"}
+{"current_steps": 151, "total_steps": 366, "loss": 0.7524, "lr": 7.48620468920972e-05, "epoch": 1.229038854805726, "percentage": 41.26, "elapsed_time": "0:04:50", "remaining_time": "0:06:53"}
+{"current_steps": 152, "total_steps": 366, "loss": 0.6238, "lr": 7.444668186293153e-05, "epoch": 1.2372188139059306, "percentage": 41.53, "elapsed_time": "0:04:52", "remaining_time": "0:06:51"}
+{"current_steps": 153, "total_steps": 366, "loss": 0.7599, "lr": 7.402908775933419e-05, "epoch": 1.2453987730061349, "percentage": 41.8, "elapsed_time": "0:04:54", "remaining_time": "0:06:49"}
+{"current_steps": 154, "total_steps": 366, "loss": 1.1331, "lr": 7.360930265797935e-05, "epoch": 1.2535787321063394, "percentage": 42.08, "elapsed_time": "0:04:56", "remaining_time": "0:06:47"}
+{"current_steps": 155, "total_steps": 366, "loss": 0.6468, "lr": 7.31873648353186e-05, "epoch": 1.261758691206544, "percentage": 42.35, "elapsed_time": "0:04:58", "remaining_time": "0:06:45"}
+{"current_steps": 156, "total_steps": 366, "loss": 0.7253, "lr": 7.276331276409106e-05, "epoch": 1.2699386503067485, "percentage": 42.62, "elapsed_time": "0:05:00", "remaining_time": "0:06:43"}
+{"current_steps": 157, "total_steps": 366, "loss": 0.842, "lr": 7.23371851098152e-05, "epoch": 1.278118609406953, "percentage": 42.9, "elapsed_time": "0:05:01", "remaining_time": "0:06:41"}
+{"current_steps": 158, "total_steps": 366, "loss": 0.5379, "lr": 7.190902072726335e-05, "epoch": 1.2862985685071575, "percentage": 43.17, "elapsed_time": "0:05:03", "remaining_time": "0:06:40"}
+{"current_steps": 159, "total_steps": 366, "loss": 0.918, "lr": 7.147885865691899e-05, "epoch": 1.294478527607362, "percentage": 43.44, "elapsed_time": "0:05:05", "remaining_time": "0:06:38"}
+{"current_steps": 160, "total_steps": 366, "loss": 0.6727, "lr": 7.104673812141675e-05, "epoch": 1.3026584867075663, "percentage": 43.72, "elapsed_time": "0:05:07", "remaining_time": "0:06:36"}
+{"current_steps": 161, "total_steps": 366, "loss": 0.4279, "lr": 7.061269852196632e-05, "epoch": 1.310838445807771, "percentage": 43.99, "elapsed_time": "0:05:09", "remaining_time": "0:06:34"}
+{"current_steps": 162, "total_steps": 366, "loss": 0.6372, "lr": 7.017677943475961e-05, "epoch": 1.3190184049079754, "percentage": 44.26, "elapsed_time": "0:05:11", "remaining_time": "0:06:32"}
+{"current_steps": 163, "total_steps": 366, "loss": 0.7071, "lr": 6.973902060736226e-05, "epoch": 1.32719836400818, "percentage": 44.54, "elapsed_time": "0:05:13", "remaining_time": "0:06:30"}
+{"current_steps": 164, "total_steps": 366, "loss": 0.767, "lr": 6.929946195508932e-05, "epoch": 1.3353783231083844, "percentage": 44.81, "elapsed_time": "0:05:15", "remaining_time": "0:06:28"}
+{"current_steps": 165, "total_steps": 366, "loss": 0.5587, "lr": 6.885814355736586e-05, "epoch": 1.343558282208589, "percentage": 45.08, "elapsed_time": "0:05:17", "remaining_time": "0:06:26"}
+{"current_steps": 166, "total_steps": 366, "loss": 0.7519, "lr": 6.841510565407235e-05, "epoch": 1.3517382413087935, "percentage": 45.36, "elapsed_time": "0:05:19", "remaining_time": "0:06:24"}
+{"current_steps": 167, "total_steps": 366, "loss": 0.5612, "lr": 6.797038864187564e-05, "epoch": 1.359918200408998, "percentage": 45.63, "elapsed_time": "0:05:20", "remaining_time": "0:06:22"}
+{"current_steps": 168, "total_steps": 366, "loss": 0.7194, "lr": 6.752403307054549e-05, "epoch": 1.3680981595092025, "percentage": 45.9, "elapsed_time": "0:05:22", "remaining_time": "0:06:20"}
+{"current_steps": 169, "total_steps": 366, "loss": 0.531, "lr": 6.707607963925724e-05, "epoch": 1.3762781186094069, "percentage": 46.17, "elapsed_time": "0:05:24", "remaining_time": "0:06:18"}
+{"current_steps": 170, "total_steps": 366, "loss": 0.7906, "lr": 6.66265691928808e-05, "epoch": 1.3844580777096114, "percentage": 46.45, "elapsed_time": "0:05:26", "remaining_time": "0:06:16"}
+{"current_steps": 171, "total_steps": 366, "loss": 0.7207, "lr": 6.617554271825636e-05, "epoch": 1.392638036809816, "percentage": 46.72, "elapsed_time": "0:05:28", "remaining_time": "0:06:14"}
+{"current_steps": 172, "total_steps": 366, "loss": 0.5145, "lr": 6.572304134045717e-05, "epoch": 1.4008179959100204, "percentage": 46.99, "elapsed_time": "0:05:30", "remaining_time": "0:06:12"}
+{"current_steps": 173, "total_steps": 366, "loss": 0.3521, "lr": 6.526910631903973e-05, "epoch": 1.408997955010225, "percentage": 47.27, "elapsed_time": "0:05:32", "remaining_time": "0:06:10"}
+{"current_steps": 174, "total_steps": 366, "loss": 0.7026, "lr": 6.481377904428171e-05, "epoch": 1.4171779141104295, "percentage": 47.54, "elapsed_time": "0:05:34", "remaining_time": "0:06:08"}
+{"current_steps": 175, "total_steps": 366, "loss": 0.7313, "lr": 6.435710103340786e-05, "epoch": 1.425357873210634, "percentage": 47.81, "elapsed_time": "0:05:36", "remaining_time": "0:06:06"}
+{"current_steps": 176, "total_steps": 366, "loss": 0.7659, "lr": 6.389911392680456e-05, "epoch": 1.4335378323108383, "percentage": 48.09, "elapsed_time": "0:05:37", "remaining_time": "0:06:04"}
+{"current_steps": 177, "total_steps": 366, "loss": 0.6916, "lr": 6.343985948422287e-05, "epoch": 1.441717791411043, "percentage": 48.36, "elapsed_time": "0:05:40", "remaining_time": "0:06:03"}
+{"current_steps": 178, "total_steps": 366, "loss": 0.8101, "lr": 6.297937958097094e-05, "epoch": 1.4498977505112474, "percentage": 48.63, "elapsed_time": "0:05:42", "remaining_time": "0:06:01"}
+{"current_steps": 179, "total_steps": 366, "loss": 0.7504, "lr": 6.251771620409563e-05, "epoch": 1.4580777096114519, "percentage": 48.91, "elapsed_time": "0:05:44", "remaining_time": "0:05:59"}
+{"current_steps": 180, "total_steps": 366, "loss": 0.5948, "lr": 6.205491144855432e-05, "epoch": 1.4662576687116564, "percentage": 49.18, "elapsed_time": "0:05:45", "remaining_time": "0:05:57"}
+{"current_steps": 181, "total_steps": 366, "loss": 0.7057, "lr": 6.159100751337642e-05, "epoch": 1.474437627811861, "percentage": 49.45, "elapsed_time": "0:05:47", "remaining_time": "0:05:55"}
+{"current_steps": 182, "total_steps": 366, "loss": 0.9037, "lr": 6.112604669781572e-05, "epoch": 1.4826175869120655, "percentage": 49.73, "elapsed_time": "0:05:49", "remaining_time": "0:05:53"}
+{"current_steps": 183, "total_steps": 366, "loss": 0.7223, "lr": 6.0660071397493514e-05, "epoch": 1.49079754601227, "percentage": 50.0, "elapsed_time": "0:05:51", "remaining_time": "0:05:51"}
+{"current_steps": 184, "total_steps": 366, "loss": 0.6083, "lr": 6.019312410053286e-05, "epoch": 1.4989775051124745, "percentage": 50.27, "elapsed_time": "0:05:53", "remaining_time": "0:05:49"}
+{"current_steps": 185, "total_steps": 366, "loss": 0.7662, "lr": 5.972524738368452e-05, "epoch": 1.5071574642126788, "percentage": 50.55, "elapsed_time": "0:05:55", "remaining_time": "0:05:47"}
+{"current_steps": 186, "total_steps": 366, "loss": 0.902, "lr": 5.925648390844476e-05, "epoch": 1.5153374233128836, "percentage": 50.82, "elapsed_time": "0:05:57", "remaining_time": "0:05:45"}
+{"current_steps": 187, "total_steps": 366, "loss": 0.6566, "lr": 5.878687641716538e-05, "epoch": 1.5235173824130879, "percentage": 51.09, "elapsed_time": "0:05:59", "remaining_time": "0:05:43"}
+{"current_steps": 188, "total_steps": 366, "loss": 0.4189, "lr": 5.831646772915651e-05, "epoch": 1.5316973415132924, "percentage": 51.37, "elapsed_time": "0:06:01", "remaining_time": "0:05:41"}
+{"current_steps": 189, "total_steps": 366, "loss": 0.5549, "lr": 5.7845300736782204e-05, "epoch": 1.539877300613497, "percentage": 51.64, "elapsed_time": "0:06:03", "remaining_time": "0:05:40"}
+{"current_steps": 190, "total_steps": 366, "loss": 0.5456, "lr": 5.737341840154956e-05, "epoch": 1.5480572597137015, "percentage": 51.91, "elapsed_time": "0:06:05", "remaining_time": "0:05:38"}
+{"current_steps": 191, "total_steps": 366, "loss": 0.6808, "lr": 5.6900863750191347e-05, "epoch": 1.556237218813906, "percentage": 52.19, "elapsed_time": "0:06:06", "remaining_time": "0:05:36"}
+{"current_steps": 192, "total_steps": 366, "loss": 0.7974, "lr": 5.642767987074288e-05, "epoch": 1.5644171779141103, "percentage": 52.46, "elapsed_time": "0:06:08", "remaining_time": "0:05:34"}
+{"current_steps": 193, "total_steps": 366, "loss": 0.5416, "lr": 5.5953909908613114e-05, "epoch": 1.572597137014315, "percentage": 52.73, "elapsed_time": "0:06:10", "remaining_time": "0:05:32"}
+{"current_steps": 194, "total_steps": 366, "loss": 0.6788, "lr": 5.547959706265068e-05, "epoch": 1.5807770961145193, "percentage": 53.01, "elapsed_time": "0:06:12", "remaining_time": "0:05:30"}
+{"current_steps": 195, "total_steps": 366, "loss": 0.9634, "lr": 5.5004784581204927e-05, "epoch": 1.588957055214724, "percentage": 53.28, "elapsed_time": "0:06:14", "remaining_time": "0:05:28"}
+{"current_steps": 196, "total_steps": 366, "loss": 0.6563, "lr": 5.4529515758182506e-05, "epoch": 1.5971370143149284, "percentage": 53.55, "elapsed_time": "0:06:16", "remaining_time": "0:05:26"}
+{"current_steps": 197, "total_steps": 366, "loss": 0.6062, "lr": 5.405383392909973e-05, "epoch": 1.605316973415133, "percentage": 53.83, "elapsed_time": "0:06:18", "remaining_time": "0:05:24"}
+{"current_steps": 198, "total_steps": 366, "loss": 0.4829, "lr": 5.357778246713131e-05, "epoch": 1.6134969325153374, "percentage": 54.1, "elapsed_time": "0:06:20", "remaining_time": "0:05:22"}
+{"current_steps": 199, "total_steps": 366, "loss": 0.465, "lr": 5.310140477915544e-05, "epoch": 1.621676891615542, "percentage": 54.37, "elapsed_time": "0:06:22", "remaining_time": "0:05:20"}
+{"current_steps": 200, "total_steps": 366, "loss": 0.6967, "lr": 5.262474430179597e-05, "epoch": 1.6298568507157465, "percentage": 54.64, "elapsed_time": "0:06:24", "remaining_time": "0:05:18"}
+{"current_steps": 201, "total_steps": 366, "loss": 0.9096, "lr": 5.214784449746174e-05, "epoch": 1.6380368098159508, "percentage": 54.92, "elapsed_time": "0:06:26", "remaining_time": "0:05:16"}
+{"current_steps": 202, "total_steps": 366, "loss": 0.7473, "lr": 5.167074885038373e-05, "epoch": 1.6462167689161555, "percentage": 55.19, "elapsed_time": "0:06:27", "remaining_time": "0:05:14"}
+{"current_steps": 203, "total_steps": 366, "loss": 0.6233, "lr": 5.119350086265004e-05, "epoch": 1.6543967280163598, "percentage": 55.46, "elapsed_time": "0:06:29", "remaining_time": "0:05:12"}
+{"current_steps": 204, "total_steps": 366, "loss": 0.7599, "lr": 5.0716144050239375e-05, "epoch": 1.6625766871165644, "percentage": 55.74, "elapsed_time": "0:06:31", "remaining_time": "0:05:11"}
+{"current_steps": 205, "total_steps": 366, "loss": 0.5638, "lr": 5.023872193905316e-05, "epoch": 1.670756646216769, "percentage": 56.01, "elapsed_time": "0:06:33", "remaining_time": "0:05:09"}
+{"current_steps": 206, "total_steps": 366, "loss": 0.4701, "lr": 4.976127806094684e-05, "epoch": 1.6789366053169734, "percentage": 56.28, "elapsed_time": "0:06:35", "remaining_time": "0:05:07"}
+{"current_steps": 207, "total_steps": 366, "loss": 0.9383, "lr": 4.928385594976063e-05, "epoch": 1.687116564417178, "percentage": 56.56, "elapsed_time": "0:06:37", "remaining_time": "0:05:05"}
+{"current_steps": 208, "total_steps": 366, "loss": 0.5817, "lr": 4.880649913734996e-05, "epoch": 1.6952965235173822, "percentage": 56.83, "elapsed_time": "0:06:39", "remaining_time": "0:05:03"}
+{"current_steps": 209, "total_steps": 366, "loss": 0.3481, "lr": 4.832925114961629e-05, "epoch": 1.703476482617587, "percentage": 57.1, "elapsed_time": "0:06:41", "remaining_time": "0:05:01"}
+{"current_steps": 210, "total_steps": 366, "loss": 0.4348, "lr": 4.785215550253826e-05, "epoch": 1.7116564417177913, "percentage": 57.38, "elapsed_time": "0:06:43", "remaining_time": "0:04:59"}
+{"current_steps": 211, "total_steps": 366, "loss": 0.653, "lr": 4.7375255698204045e-05, "epoch": 1.719836400817996, "percentage": 57.65, "elapsed_time": "0:06:45", "remaining_time": "0:04:57"}
+{"current_steps": 212, "total_steps": 366, "loss": 0.7181, "lr": 4.6898595220844574e-05, "epoch": 1.7280163599182004, "percentage": 57.92, "elapsed_time": "0:06:46", "remaining_time": "0:04:55"}
+{"current_steps": 213, "total_steps": 366, "loss": 0.6183, "lr": 4.64222175328687e-05, "epoch": 1.7361963190184049, "percentage": 58.2, "elapsed_time": "0:06:48", "remaining_time": "0:04:53"}
+{"current_steps": 214, "total_steps": 366, "loss": 0.689, "lr": 4.594616607090028e-05, "epoch": 1.7443762781186094, "percentage": 58.47, "elapsed_time": "0:06:50", "remaining_time": "0:04:51"}
+{"current_steps": 215, "total_steps": 366, "loss": 0.4616, "lr": 4.547048424181751e-05, "epoch": 1.752556237218814, "percentage": 58.74, "elapsed_time": "0:06:52", "remaining_time": "0:04:49"}
+{"current_steps": 216, "total_steps": 366, "loss": 0.6318, "lr": 4.4995215418795085e-05, "epoch": 1.7607361963190185, "percentage": 59.02, "elapsed_time": "0:06:54", "remaining_time": "0:04:47"}
+{"current_steps": 217, "total_steps": 366, "loss": 0.4611, "lr": 4.452040293734934e-05, "epoch": 1.7689161554192228, "percentage": 59.29, "elapsed_time": "0:06:56", "remaining_time": "0:04:45"}
+{"current_steps": 218, "total_steps": 366, "loss": 0.1962, "lr": 4.404609009138689e-05, "epoch": 1.7770961145194275, "percentage": 59.56, "elapsed_time": "0:06:58", "remaining_time": "0:04:44"}
+{"current_steps": 219, "total_steps": 366, "loss": 0.3804, "lr": 4.357232012925714e-05, "epoch": 1.7852760736196318, "percentage": 59.84, "elapsed_time": "0:07:00", "remaining_time": "0:04:42"}
+{"current_steps": 220, "total_steps": 366, "loss": 0.5431, "lr": 4.3099136249808665e-05, "epoch": 1.7934560327198366, "percentage": 60.11, "elapsed_time": "0:07:02", "remaining_time": "0:04:40"}
+{"current_steps": 221, "total_steps": 366, "loss": 0.6498, "lr": 4.262658159845046e-05, "epoch": 1.8016359918200409, "percentage": 60.38, "elapsed_time": "0:07:04", "remaining_time": "0:04:38"}
+{"current_steps": 222, "total_steps": 366, "loss": 0.4812, "lr": 4.215469926321779e-05, "epoch": 1.8098159509202454, "percentage": 60.66, "elapsed_time": "0:07:05", "remaining_time": "0:04:36"}
+{"current_steps": 223, "total_steps": 366, "loss": 0.5703, "lr": 4.1683532270843504e-05, "epoch": 1.81799591002045, "percentage": 60.93, "elapsed_time": "0:07:07", "remaining_time": "0:04:34"}
+{"current_steps": 224, "total_steps": 366, "loss": 0.992, "lr": 4.121312358283463e-05, "epoch": 1.8261758691206544, "percentage": 61.2, "elapsed_time": "0:07:09", "remaining_time": "0:04:32"}
+{"current_steps": 225, "total_steps": 366, "loss": 0.5907, "lr": 4.074351609155527e-05, "epoch": 1.834355828220859, "percentage": 61.48, "elapsed_time": "0:07:11", "remaining_time": "0:04:30"}
+{"current_steps": 226, "total_steps": 366, "loss": 0.7448, "lr": 4.027475261631548e-05, "epoch": 1.8425357873210633, "percentage": 61.75, "elapsed_time": "0:07:13", "remaining_time": "0:04:28"}
+{"current_steps": 227, "total_steps": 366, "loss": 0.8506, "lr": 3.980687589946715e-05, "epoch": 1.850715746421268, "percentage": 62.02, "elapsed_time": "0:07:15", "remaining_time": "0:04:26"}
+{"current_steps": 228, "total_steps": 366, "loss": 0.4935, "lr": 3.9339928602506505e-05, "epoch": 1.8588957055214723, "percentage": 62.3, "elapsed_time": "0:07:17", "remaining_time": "0:04:24"}
+{"current_steps": 229, "total_steps": 366, "loss": 0.586, "lr": 3.887395330218429e-05, "epoch": 1.8670756646216768, "percentage": 62.57, "elapsed_time": "0:07:19", "remaining_time": "0:04:22"}
+{"current_steps": 230, "total_steps": 366, "loss": 0.7596, "lr": 3.840899248662358e-05, "epoch": 1.8752556237218814, "percentage": 62.84, "elapsed_time": "0:07:21", "remaining_time": "0:04:20"}
+{"current_steps": 231, "total_steps": 366, "loss": 0.7946, "lr": 3.7945088551445693e-05, "epoch": 1.883435582822086, "percentage": 63.11, "elapsed_time": "0:07:23", "remaining_time": "0:04:18"}
+{"current_steps": 232, "total_steps": 366, "loss": 0.5414, "lr": 3.748228379590438e-05, "epoch": 1.8916155419222904, "percentage": 63.39, "elapsed_time": "0:07:25", "remaining_time": "0:04:17"}
+{"current_steps": 233, "total_steps": 366, "loss": 0.6574, "lr": 3.7020620419029094e-05, "epoch": 1.8997955010224947, "percentage": 63.66, "elapsed_time": "0:07:26", "remaining_time": "0:04:15"}
+{"current_steps": 234, "total_steps": 366, "loss": 0.5052, "lr": 3.656014051577713e-05, "epoch": 1.9079754601226995, "percentage": 63.93, "elapsed_time": "0:07:28", "remaining_time": "0:04:13"}
+{"current_steps": 235, "total_steps": 366, "loss": 0.4209, "lr": 3.610088607319544e-05, "epoch": 1.9161554192229038, "percentage": 64.21, "elapsed_time": "0:07:31", "remaining_time": "0:04:11"}
+{"current_steps": 236, "total_steps": 366, "loss": 0.562, "lr": 3.564289896659214e-05, "epoch": 1.9243353783231085, "percentage": 64.48, "elapsed_time": "0:07:32", "remaining_time": "0:04:09"}
+{"current_steps": 237, "total_steps": 366, "loss": 0.6494, "lr": 3.5186220955718306e-05, "epoch": 1.9325153374233128, "percentage": 64.75, "elapsed_time": "0:07:34", "remaining_time": "0:04:07"}
+{"current_steps": 238, "total_steps": 366, "loss": 0.5365, "lr": 3.473089368096026e-05, "epoch": 1.9406952965235174, "percentage": 65.03, "elapsed_time": "0:07:36", "remaining_time": "0:04:05"}
+{"current_steps": 239, "total_steps": 366, "loss": 0.7972, "lr": 3.427695865954284e-05, "epoch": 1.9488752556237219, "percentage": 65.3, "elapsed_time": "0:07:38", "remaining_time": "0:04:03"}
+{"current_steps": 240, "total_steps": 366, "loss": 0.2413, "lr": 3.3824457281743646e-05, "epoch": 1.9570552147239264, "percentage": 65.57, "elapsed_time": "0:07:40", "remaining_time": "0:04:01"}
+{"current_steps": 241, "total_steps": 366, "loss": 0.655, "lr": 3.337343080711921e-05, "epoch": 1.965235173824131, "percentage": 65.85, "elapsed_time": "0:07:42", "remaining_time": "0:03:59"}
+{"current_steps": 242, "total_steps": 366, "loss": 0.7517, "lr": 3.2923920360742774e-05, "epoch": 1.9734151329243352, "percentage": 66.12, "elapsed_time": "0:07:44", "remaining_time": "0:03:57"}
+{"current_steps": 243, "total_steps": 366, "loss": 0.527, "lr": 3.2475966929454504e-05, "epoch": 1.98159509202454, "percentage": 66.39, "elapsed_time": "0:07:45", "remaining_time": "0:03:55"}
+{"current_steps": 244, "total_steps": 366, "loss": 0.5922, "lr": 3.202961135812437e-05, "epoch": 1.9897750511247443, "percentage": 66.67, "elapsed_time": "0:07:47", "remaining_time": "0:03:53"}
+{"current_steps": 245, "total_steps": 366, "loss": 0.6738, "lr": 3.158489434592766e-05, "epoch": 1.997955010224949, "percentage": 66.94, "elapsed_time": "0:07:49", "remaining_time": "0:03:52"}
+{"current_steps": 246, "total_steps": 366, "loss": 0.1228, "lr": 3.114185644263415e-05, "epoch": 2.0, "percentage": 67.21, "elapsed_time": "0:07:50", "remaining_time": "0:03:49"}
+{"current_steps": 247, "total_steps": 366, "loss": 0.5372, "lr": 3.070053804491068e-05, "epoch": 2.0081799591002043, "percentage": 67.49, "elapsed_time": "0:07:53", "remaining_time": "0:03:48"}
+{"current_steps": 248, "total_steps": 366, "loss": 0.3056, "lr": 3.026097939263775e-05, "epoch": 2.016359918200409, "percentage": 67.76, "elapsed_time": "0:07:55", "remaining_time": "0:03:46"}
+{"current_steps": 249, "total_steps": 366, "loss": 0.5469, "lr": 2.9823220565240394e-05, "epoch": 2.0245398773006134, "percentage": 68.03, "elapsed_time": "0:07:57", "remaining_time": "0:03:44"}
+{"current_steps": 250, "total_steps": 366, "loss": 0.5333, "lr": 2.938730147803369e-05, "epoch": 2.032719836400818, "percentage": 68.31, "elapsed_time": "0:07:59", "remaining_time": "0:03:42"}
+{"current_steps": 251, "total_steps": 366, "loss": 0.4873, "lr": 2.895326187858326e-05, "epoch": 2.0408997955010224, "percentage": 68.58, "elapsed_time": "0:08:01", "remaining_time": "0:03:40"}
+{"current_steps": 252, "total_steps": 366, "loss": 0.4676, "lr": 2.852114134308104e-05, "epoch": 2.049079754601227, "percentage": 68.85, "elapsed_time": "0:08:03", "remaining_time": "0:03:38"}
+{"current_steps": 253, "total_steps": 366, "loss": 0.5474, "lr": 2.8090979272736662e-05, "epoch": 2.0572597137014315, "percentage": 69.13, "elapsed_time": "0:08:05", "remaining_time": "0:03:36"}
+{"current_steps": 254, "total_steps": 366, "loss": 0.3774, "lr": 2.7662814890184818e-05, "epoch": 2.065439672801636, "percentage": 69.4, "elapsed_time": "0:08:06", "remaining_time": "0:03:34"}
+{"current_steps": 255, "total_steps": 366, "loss": 0.5021, "lr": 2.7236687235908953e-05, "epoch": 2.0736196319018405, "percentage": 69.67, "elapsed_time": "0:08:08", "remaining_time": "0:03:32"}
+{"current_steps": 256, "total_steps": 366, "loss": 0.295, "lr": 2.6812635164681386e-05, "epoch": 2.081799591002045, "percentage": 69.95, "elapsed_time": "0:08:10", "remaining_time": "0:03:30"}
+{"current_steps": 257, "total_steps": 366, "loss": 0.4359, "lr": 2.6390697342020665e-05, "epoch": 2.0899795501022496, "percentage": 70.22, "elapsed_time": "0:08:12", "remaining_time": "0:03:29"}
+{"current_steps": 258, "total_steps": 366, "loss": 0.4699, "lr": 2.5970912240665813e-05, "epoch": 2.098159509202454, "percentage": 70.49, "elapsed_time": "0:08:14", "remaining_time": "0:03:27"}
+{"current_steps": 259, "total_steps": 366, "loss": 0.5016, "lr": 2.555331813706847e-05, "epoch": 2.1063394683026586, "percentage": 70.77, "elapsed_time": "0:08:16", "remaining_time": "0:03:25"}
+{"current_steps": 260, "total_steps": 366, "loss": 0.4827, "lr": 2.5137953107902813e-05, "epoch": 2.114519427402863, "percentage": 71.04, "elapsed_time": "0:08:18", "remaining_time": "0:03:23"}
+{"current_steps": 261, "total_steps": 366, "loss": 0.3951, "lr": 2.472485502659358e-05, "epoch": 2.1226993865030677, "percentage": 71.31, "elapsed_time": "0:08:20", "remaining_time": "0:03:21"}
+{"current_steps": 262, "total_steps": 366, "loss": 0.3384, "lr": 2.4314061559862833e-05, "epoch": 2.130879345603272, "percentage": 71.58, "elapsed_time": "0:08:22", "remaining_time": "0:03:19"}
+{"current_steps": 263, "total_steps": 366, "loss": 0.4982, "lr": 2.3905610164295394e-05, "epoch": 2.1390593047034763, "percentage": 71.86, "elapsed_time": "0:08:24", "remaining_time": "0:03:17"}
+{"current_steps": 264, "total_steps": 366, "loss": 0.4574, "lr": 2.3499538082923606e-05, "epoch": 2.147239263803681, "percentage": 72.13, "elapsed_time": "0:08:26", "remaining_time": "0:03:15"}
+{"current_steps": 265, "total_steps": 366, "loss": 0.3559, "lr": 2.3095882341831372e-05, "epoch": 2.1554192229038853, "percentage": 72.4, "elapsed_time": "0:08:28", "remaining_time": "0:03:13"}
+{"current_steps": 266, "total_steps": 366, "loss": 0.3721, "lr": 2.2694679746778115e-05, "epoch": 2.16359918200409, "percentage": 72.68, "elapsed_time": "0:08:29", "remaining_time": "0:03:11"}
+{"current_steps": 267, "total_steps": 366, "loss": 0.4909, "lr": 2.22959668798428e-05, "epoch": 2.1717791411042944, "percentage": 72.95, "elapsed_time": "0:08:31", "remaining_time": "0:03:09"}
+{"current_steps": 268, "total_steps": 366, "loss": 0.7858, "lr": 2.1899780096088375e-05, "epoch": 2.179959100204499, "percentage": 73.22, "elapsed_time": "0:08:33", "remaining_time": "0:03:07"}
+{"current_steps": 269, "total_steps": 366, "loss": 0.6337, "lr": 2.1506155520246797e-05, "epoch": 2.1881390593047034, "percentage": 73.5, "elapsed_time": "0:08:35", "remaining_time": "0:03:05"}
+{"current_steps": 270, "total_steps": 366, "loss": 0.2693, "lr": 2.1115129043425187e-05, "epoch": 2.196319018404908, "percentage": 73.77, "elapsed_time": "0:08:37", "remaining_time": "0:03:04"}
+{"current_steps": 271, "total_steps": 366, "loss": 0.5306, "lr": 2.0726736319833228e-05, "epoch": 2.2044989775051125, "percentage": 74.04, "elapsed_time": "0:08:39", "remaining_time": "0:03:02"}
+{"current_steps": 272, "total_steps": 366, "loss": 0.3388, "lr": 2.0341012763532243e-05, "epoch": 2.212678936605317, "percentage": 74.32, "elapsed_time": "0:08:41", "remaining_time": "0:03:00"}
+{"current_steps": 273, "total_steps": 366, "loss": 0.3615, "lr": 1.995799354520598e-05, "epoch": 2.2208588957055215, "percentage": 74.59, "elapsed_time": "0:08:43", "remaining_time": "0:02:58"}
+{"current_steps": 274, "total_steps": 366, "loss": 0.5129, "lr": 1.9577713588953795e-05, "epoch": 2.229038854805726, "percentage": 74.86, "elapsed_time": "0:08:45", "remaining_time": "0:02:56"}
+{"current_steps": 275, "total_steps": 366, "loss": 0.4129, "lr": 1.9200207569106216e-05, "epoch": 2.2372188139059306, "percentage": 75.14, "elapsed_time": "0:08:47", "remaining_time": "0:02:54"}
+{"current_steps": 276, "total_steps": 366, "loss": 0.5374, "lr": 1.8825509907063327e-05, "epoch": 2.245398773006135, "percentage": 75.41, "elapsed_time": "0:08:48", "remaining_time": "0:02:52"}
+{"current_steps": 277, "total_steps": 366, "loss": 0.562, "lr": 1.8453654768156138e-05, "epoch": 2.2535787321063396, "percentage": 75.68, "elapsed_time": "0:08:50", "remaining_time": "0:02:50"}
+{"current_steps": 278, "total_steps": 366, "loss": 0.6637, "lr": 1.8084676058531373e-05, "epoch": 2.261758691206544, "percentage": 75.96, "elapsed_time": "0:08:52", "remaining_time": "0:02:48"}
+{"current_steps": 279, "total_steps": 366, "loss": 0.5932, "lr": 1.771860742205988e-05, "epoch": 2.2699386503067487, "percentage": 76.23, "elapsed_time": "0:08:54", "remaining_time": "0:02:46"}
+{"current_steps": 280, "total_steps": 366, "loss": 0.341, "lr": 1.7355482237268983e-05, "epoch": 2.278118609406953, "percentage": 76.5, "elapsed_time": "0:08:56", "remaining_time": "0:02:44"}
+{"current_steps": 281, "total_steps": 366, "loss": 0.4248, "lr": 1.699533361429891e-05, "epoch": 2.2862985685071573, "percentage": 76.78, "elapsed_time": "0:08:58", "remaining_time": "0:02:42"}
+{"current_steps": 282, "total_steps": 366, "loss": 0.7139, "lr": 1.663819439188382e-05, "epoch": 2.294478527607362, "percentage": 77.05, "elapsed_time": "0:09:00", "remaining_time": "0:02:40"}
+{"current_steps": 283, "total_steps": 366, "loss": 0.4609, "lr": 1.6284097134357536e-05, "epoch": 2.3026584867075663, "percentage": 77.32, "elapsed_time": "0:09:02", "remaining_time": "0:02:39"}
+{"current_steps": 284, "total_steps": 366, "loss": 0.3572, "lr": 1.5933074128684332e-05, "epoch": 2.310838445807771, "percentage": 77.6, "elapsed_time": "0:09:04", "remaining_time": "0:02:37"}
+{"current_steps": 285, "total_steps": 366, "loss": 0.5078, "lr": 1.5585157381514875e-05, "epoch": 2.3190184049079754, "percentage": 77.87, "elapsed_time": "0:09:06", "remaining_time": "0:02:35"}
+{"current_steps": 286, "total_steps": 366, "loss": 0.5262, "lr": 1.5240378616267886e-05, "epoch": 2.32719836400818, "percentage": 78.14, "elapsed_time": "0:09:08", "remaining_time": "0:02:33"}
+{"current_steps": 287, "total_steps": 366, "loss": 0.4075, "lr": 1.489876927023761e-05, "epoch": 2.3353783231083844, "percentage": 78.42, "elapsed_time": "0:09:09", "remaining_time": "0:02:31"}
+{"current_steps": 288, "total_steps": 366, "loss": 0.4237, "lr": 1.4560360491727231e-05, "epoch": 2.3435582822085887, "percentage": 78.69, "elapsed_time": "0:09:11", "remaining_time": "0:02:29"}
+{"current_steps": 289, "total_steps": 366, "loss": 0.7344, "lr": 1.4225183137208776e-05, "epoch": 2.3517382413087935, "percentage": 78.96, "elapsed_time": "0:09:13", "remaining_time": "0:02:27"}
+{"current_steps": 290, "total_steps": 366, "loss": 0.5226, "lr": 1.389326776850966e-05, "epoch": 2.359918200408998, "percentage": 79.23, "elapsed_time": "0:09:15", "remaining_time": "0:02:25"}
+{"current_steps": 291, "total_steps": 366, "loss": 0.2004, "lr": 1.3564644650025893e-05, "epoch": 2.3680981595092025, "percentage": 79.51, "elapsed_time": "0:09:17", "remaining_time": "0:02:23"}
+{"current_steps": 292, "total_steps": 366, "loss": 0.5426, "lr": 1.3239343745962679e-05, "epoch": 2.376278118609407, "percentage": 79.78, "elapsed_time": "0:09:19", "remaining_time": "0:02:21"}
+{"current_steps": 293, "total_steps": 366, "loss": 0.3689, "lr": 1.2917394717602121e-05, "epoch": 2.3844580777096116, "percentage": 80.05, "elapsed_time": "0:09:21", "remaining_time": "0:02:19"}
+{"current_steps": 294, "total_steps": 366, "loss": 0.3994, "lr": 1.2598826920598772e-05, "epoch": 2.392638036809816, "percentage": 80.33, "elapsed_time": "0:09:23", "remaining_time": "0:02:17"}
+{"current_steps": 295, "total_steps": 366, "loss": 0.4635, "lr": 1.2283669402302878e-05, "epoch": 2.40081799591002, "percentage": 80.6, "elapsed_time": "0:09:25", "remaining_time": "0:02:16"}
+{"current_steps": 296, "total_steps": 366, "loss": 0.44, "lr": 1.197195089911191e-05, "epoch": 2.408997955010225, "percentage": 80.87, "elapsed_time": "0:09:27", "remaining_time": "0:02:14"}
+{"current_steps": 297, "total_steps": 366, "loss": 0.809, "lr": 1.1663699833850238e-05, "epoch": 2.4171779141104293, "percentage": 81.15, "elapsed_time": "0:09:28", "remaining_time": "0:02:12"}
+{"current_steps": 298, "total_steps": 366, "loss": 0.526, "lr": 1.1358944313177567e-05, "epoch": 2.425357873210634, "percentage": 81.42, "elapsed_time": "0:09:30", "remaining_time": "0:02:10"}
+{"current_steps": 299, "total_steps": 366, "loss": 0.4943, "lr": 1.1057712125026116e-05, "epoch": 2.4335378323108383, "percentage": 81.69, "elapsed_time": "0:09:32", "remaining_time": "0:02:08"}
+{"current_steps": 300, "total_steps": 366, "loss": 0.2973, "lr": 1.0760030736066951e-05, "epoch": 2.441717791411043, "percentage": 81.97, "elapsed_time": "0:09:34", "remaining_time": "0:02:06"}
+{"current_steps": 301, "total_steps": 366, "loss": 0.4647, "lr": 1.0465927289205452e-05, "epoch": 2.4498977505112474, "percentage": 82.24, "elapsed_time": "0:09:36", "remaining_time": "0:02:04"}
+{"current_steps": 302, "total_steps": 366, "loss": 0.5614, "lr": 1.017542860110644e-05, "epoch": 2.458077709611452, "percentage": 82.51, "elapsed_time": "0:09:38", "remaining_time": "0:02:02"}
+{"current_steps": 303, "total_steps": 366, "loss": 0.2343, "lr": 9.888561159748993e-06, "epoch": 2.4662576687116564, "percentage": 82.79, "elapsed_time": "0:09:40", "remaining_time": "0:02:00"}
+{"current_steps": 304, "total_steps": 366, "loss": 0.5084, "lr": 9.605351122011309e-06, "epoch": 2.474437627811861, "percentage": 83.06, "elapsed_time": "0:09:42", "remaining_time": "0:01:58"}
+{"current_steps": 305, "total_steps": 366, "loss": 0.4916, "lr": 9.325824311285564e-06, "epoch": 2.4826175869120655, "percentage": 83.33, "elapsed_time": "0:09:44", "remaining_time": "0:01:56"}
+{"current_steps": 306, "total_steps": 366, "loss": 0.5896, "lr": 9.050006215123419e-06, "epoch": 2.4907975460122698, "percentage": 83.61, "elapsed_time": "0:09:45", "remaining_time": "0:01:54"}
+{"current_steps": 307, "total_steps": 366, "loss": 0.3472, "lr": 8.777921982911996e-06, "epoch": 2.4989775051124745, "percentage": 83.88, "elapsed_time": "0:09:47", "remaining_time": "0:01:52"}
+{"current_steps": 308, "total_steps": 366, "loss": 0.6561, "lr": 8.509596423580712e-06, "epoch": 2.507157464212679, "percentage": 84.15, "elapsed_time": "0:09:49", "remaining_time": "0:01:51"}
+{"current_steps": 309, "total_steps": 366, "loss": 0.446, "lr": 8.245054003339247e-06, "epoch": 2.5153374233128836, "percentage": 84.43, "elapsed_time": "0:09:51", "remaining_time": "0:01:49"}
+{"current_steps": 310, "total_steps": 366, "loss": 0.6626, "lr": 7.984318843446593e-06, "epoch": 2.523517382413088, "percentage": 84.7, "elapsed_time": "0:09:53", "remaining_time": "0:01:47"}
+{"current_steps": 311, "total_steps": 366, "loss": 0.6779, "lr": 7.727414718011704e-06, "epoch": 2.5316973415132926, "percentage": 84.97, "elapsed_time": "0:09:55", "remaining_time": "0:01:45"}
+{"current_steps": 312, "total_steps": 366, "loss": 0.4369, "lr": 7.474365051825749e-06, "epoch": 2.539877300613497, "percentage": 85.25, "elapsed_time": "0:09:57", "remaining_time": "0:01:43"}
+{"current_steps": 313, "total_steps": 366, "loss": 0.5339, "lr": 7.225192918226214e-06, "epoch": 2.5480572597137012, "percentage": 85.52, "elapsed_time": "0:09:59", "remaining_time": "0:01:41"}
+{"current_steps": 314, "total_steps": 366, "loss": 0.4714, "lr": 6.979921036993042e-06, "epoch": 2.556237218813906, "percentage": 85.79, "elapsed_time": "0:10:01", "remaining_time": "0:01:39"}
+{"current_steps": 315, "total_steps": 366, "loss": 0.3589, "lr": 6.738571772276997e-06, "epoch": 2.5644171779141103, "percentage": 86.07, "elapsed_time": "0:10:02", "remaining_time": "0:01:37"}
+{"current_steps": 316, "total_steps": 366, "loss": 0.7677, "lr": 6.501167130560515e-06, "epoch": 2.572597137014315, "percentage": 86.34, "elapsed_time": "0:10:04", "remaining_time": "0:01:35"}
+{"current_steps": 317, "total_steps": 366, "loss": 0.6019, "lr": 6.267728758651132e-06, "epoch": 2.5807770961145193, "percentage": 86.61, "elapsed_time": "0:10:06", "remaining_time": "0:01:33"}
+{"current_steps": 318, "total_steps": 366, "loss": 0.3813, "lr": 6.03827794170767e-06, "epoch": 2.588957055214724, "percentage": 86.89, "elapsed_time": "0:10:08", "remaining_time": "0:01:31"}
+{"current_steps": 319, "total_steps": 366, "loss": 0.5397, "lr": 5.8128356012994375e-06, "epoch": 2.5971370143149284, "percentage": 87.16, "elapsed_time": "0:10:10", "remaining_time": "0:01:29"}
+{"current_steps": 320, "total_steps": 366, "loss": 0.5326, "lr": 5.591422293498633e-06, "epoch": 2.6053169734151327, "percentage": 87.43, "elapsed_time": "0:10:12", "remaining_time": "0:01:28"}
+{"current_steps": 321, "total_steps": 366, "loss": 0.6279, "lr": 5.374058207005944e-06, "epoch": 2.6134969325153374, "percentage": 87.7, "elapsed_time": "0:10:14", "remaining_time": "0:01:26"}
+{"current_steps": 322, "total_steps": 366, "loss": 0.7064, "lr": 5.160763161309767e-06, "epoch": 2.621676891615542, "percentage": 87.98, "elapsed_time": "0:10:16", "remaining_time": "0:01:24"}
+{"current_steps": 323, "total_steps": 366, "loss": 0.3176, "lr": 4.951556604879048e-06, "epoch": 2.6298568507157465, "percentage": 88.25, "elapsed_time": "0:10:17", "remaining_time": "0:01:22"}
+{"current_steps": 324, "total_steps": 366, "loss": 0.1989, "lr": 4.746457613389904e-06, "epoch": 2.638036809815951, "percentage": 88.52, "elapsed_time": "0:10:19", "remaining_time": "0:01:20"}
+{"current_steps": 325, "total_steps": 366, "loss": 0.4488, "lr": 4.545484887986368e-06, "epoch": 2.6462167689161555, "percentage": 88.8, "elapsed_time": "0:10:21", "remaining_time": "0:01:18"}
+{"current_steps": 326, "total_steps": 366, "loss": 0.8159, "lr": 4.348656753575092e-06, "epoch": 2.65439672801636, "percentage": 89.07, "elapsed_time": "0:10:23", "remaining_time": "0:01:16"}
+{"current_steps": 327, "total_steps": 366, "loss": 0.5941, "lr": 4.155991157154554e-06, "epoch": 2.662576687116564, "percentage": 89.34, "elapsed_time": "0:10:25", "remaining_time": "0:01:14"}
+{"current_steps": 328, "total_steps": 366, "loss": 0.6167, "lr": 3.967505666178556e-06, "epoch": 2.670756646216769, "percentage": 89.62, "elapsed_time": "0:10:27", "remaining_time": "0:01:12"}
+{"current_steps": 329, "total_steps": 366, "loss": 0.5432, "lr": 3.783217466954503e-06, "epoch": 2.6789366053169736, "percentage": 89.89, "elapsed_time": "0:10:29", "remaining_time": "0:01:10"}
+{"current_steps": 330, "total_steps": 366, "loss": 0.2688, "lr": 3.603143363076217e-06, "epoch": 2.687116564417178, "percentage": 90.16, "elapsed_time": "0:10:31", "remaining_time": "0:01:08"}
+{"current_steps": 331, "total_steps": 366, "loss": 0.3968, "lr": 3.427299773891868e-06, "epoch": 2.6952965235173822, "percentage": 90.44, "elapsed_time": "0:10:33", "remaining_time": "0:01:06"}
+{"current_steps": 332, "total_steps": 366, "loss": 0.3143, "lr": 3.2557027330067658e-06, "epoch": 2.703476482617587, "percentage": 90.71, "elapsed_time": "0:10:35", "remaining_time": "0:01:05"}
+{"current_steps": 333, "total_steps": 366, "loss": 0.4023, "lr": 3.0883678868214806e-06, "epoch": 2.7116564417177913, "percentage": 90.98, "elapsed_time": "0:10:37", "remaining_time": "0:01:03"}
+{"current_steps": 334, "total_steps": 366, "loss": 0.6501, "lr": 2.925310493105099e-06, "epoch": 2.719836400817996, "percentage": 91.26, "elapsed_time": "0:10:39", "remaining_time": "0:01:01"}
+{"current_steps": 335, "total_steps": 366, "loss": 0.3418, "lr": 2.7665454196040664e-06, "epoch": 2.7280163599182004, "percentage": 91.53, "elapsed_time": "0:10:41", "remaining_time": "0:00:59"}
+{"current_steps": 336, "total_steps": 366, "loss": 0.4047, "lr": 2.612087142686487e-06, "epoch": 2.736196319018405, "percentage": 91.8, "elapsed_time": "0:10:43", "remaining_time": "0:00:57"}
+{"current_steps": 337, "total_steps": 366, "loss": 0.3707, "lr": 2.4619497460222184e-06, "epoch": 2.7443762781186094, "percentage": 92.08, "elapsed_time": "0:10:44", "remaining_time": "0:00:55"}
+{"current_steps": 338, "total_steps": 366, "loss": 0.7776, "lr": 2.316146919298623e-06, "epoch": 2.7525562372188137, "percentage": 92.35, "elapsed_time": "0:10:46", "remaining_time": "0:00:53"}
+{"current_steps": 339, "total_steps": 366, "loss": 0.7055, "lr": 2.1746919569723855e-06, "epoch": 2.7607361963190185, "percentage": 92.62, "elapsed_time": "0:10:48", "remaining_time": "0:00:51"}
+{"current_steps": 340, "total_steps": 366, "loss": 0.6423, "lr": 2.0375977570572967e-06, "epoch": 2.7689161554192228, "percentage": 92.9, "elapsed_time": "0:10:50", "remaining_time": "0:00:49"}
+{"current_steps": 341, "total_steps": 366, "loss": 0.5679, "lr": 1.9048768199481982e-06, "epoch": 2.7770961145194275, "percentage": 93.17, "elapsed_time": "0:10:52", "remaining_time": "0:00:47"}
+{"current_steps": 342, "total_steps": 366, "loss": 0.3036, "lr": 1.7765412472811771e-06, "epoch": 2.785276073619632, "percentage": 93.44, "elapsed_time": "0:10:54", "remaining_time": "0:00:45"}
+{"current_steps": 343, "total_steps": 366, "loss": 0.5829, "lr": 1.6526027408301226e-06, "epoch": 2.7934560327198366, "percentage": 93.72, "elapsed_time": "0:10:56", "remaining_time": "0:00:44"}
+{"current_steps": 344, "total_steps": 366, "loss": 0.4617, "lr": 1.5330726014397668e-06, "epoch": 2.801635991820041, "percentage": 93.99, "elapsed_time": "0:10:58", "remaining_time": "0:00:42"}
+{"current_steps": 345, "total_steps": 366, "loss": 0.4604, "lr": 1.417961727995254e-06, "epoch": 2.809815950920245, "percentage": 94.26, "elapsed_time": "0:11:00", "remaining_time": "0:00:40"}
+{"current_steps": 346, "total_steps": 366, "loss": 0.3398, "lr": 1.3072806164283358e-06, "epoch": 2.81799591002045, "percentage": 94.54, "elapsed_time": "0:11:02", "remaining_time": "0:00:38"}
+{"current_steps": 347, "total_steps": 366, "loss": 0.3707, "lr": 1.2010393587603974e-06, "epoch": 2.8261758691206547, "percentage": 94.81, "elapsed_time": "0:11:04", "remaining_time": "0:00:36"}
+{"current_steps": 348, "total_steps": 366, "loss": 0.6991, "lr": 1.099247642182205e-06, "epoch": 2.834355828220859, "percentage": 95.08, "elapsed_time": "0:11:05", "remaining_time": "0:00:34"}
+{"current_steps": 349, "total_steps": 366, "loss": 0.6069, "lr": 1.0019147481706625e-06, "epoch": 2.8425357873210633, "percentage": 95.36, "elapsed_time": "0:11:07", "remaining_time": "0:00:32"}
+{"current_steps": 350, "total_steps": 366, "loss": 0.3841, "lr": 9.090495516424713e-07, "epoch": 2.850715746421268, "percentage": 95.63, "elapsed_time": "0:11:10", "remaining_time": "0:00:30"}
+{"current_steps": 351, "total_steps": 366, "loss": 0.3186, "lr": 8.206605201449447e-07, "epoch": 2.8588957055214723, "percentage": 95.9, "elapsed_time": "0:11:12", "remaining_time": "0:00:28"}
+{"current_steps": 352, "total_steps": 366, "loss": 0.1676, "lr": 7.36755713083892e-07, "epoch": 2.8670756646216766, "percentage": 96.17, "elapsed_time": "0:11:13", "remaining_time": "0:00:26"}
+{"current_steps": 353, "total_steps": 366, "loss": 0.7295, "lr": 6.573427809888067e-07, "epoch": 2.8752556237218814, "percentage": 96.45, "elapsed_time": "0:11:15", "remaining_time": "0:00:24"}
+{"current_steps": 354, "total_steps": 366, "loss": 0.8187, "lr": 5.824289648152126e-07, "epoch": 2.883435582822086, "percentage": 96.72, "elapsed_time": "0:11:17", "remaining_time": "0:00:22"}
+{"current_steps": 355, "total_steps": 366, "loss": 0.5205, "lr": 5.120210952844872e-07, "epoch": 2.8916155419222904, "percentage": 96.99, "elapsed_time": "0:11:19", "remaining_time": "0:00:21"}
+{"current_steps": 356, "total_steps": 366, "loss": 0.4557, "lr": 4.461255922609986e-07, "epoch": 2.8997955010224947, "percentage": 97.27, "elapsed_time": "0:11:21", "remaining_time": "0:00:19"}
+{"current_steps": 357, "total_steps": 366, "loss": 0.3251, "lr": 3.8474846416672874e-07, "epoch": 2.9079754601226995, "percentage": 97.54, "elapsed_time": "0:11:23", "remaining_time": "0:00:17"}
+{"current_steps": 358, "total_steps": 366, "loss": 0.5001, "lr": 3.278953074334512e-07, "epoch": 2.9161554192229038, "percentage": 97.81, "elapsed_time": "0:11:25", "remaining_time": "0:00:15"}
+{"current_steps": 359, "total_steps": 366, "loss": 0.2414, "lr": 2.75571305992417e-07, "epoch": 2.9243353783231085, "percentage": 98.09, "elapsed_time": "0:11:27", "remaining_time": "0:00:13"}
+{"current_steps": 360, "total_steps": 366, "loss": 0.4585, "lr": 2.2778123080167135e-07, "epoch": 2.932515337423313, "percentage": 98.36, "elapsed_time": "0:11:28", "remaining_time": "0:00:11"}
+{"current_steps": 361, "total_steps": 366, "loss": 0.5382, "lr": 1.8452943941106859e-07, "epoch": 2.9406952965235176, "percentage": 98.63, "elapsed_time": "0:11:30", "remaining_time": "0:00:09"}
+{"current_steps": 362, "total_steps": 366, "loss": 0.3326, "lr": 1.4581987556490095e-07, "epoch": 2.948875255623722, "percentage": 98.91, "elapsed_time": "0:11:32", "remaining_time": "0:00:07"}
+{"current_steps": 363, "total_steps": 366, "loss": 0.4901, "lr": 1.1165606884234181e-07, "epoch": 2.957055214723926, "percentage": 99.18, "elapsed_time": "0:11:34", "remaining_time": "0:00:05"}
+{"current_steps": 364, "total_steps": 366, "loss": 0.2756, "lr": 8.204113433559201e-08, "epoch": 2.965235173824131, "percentage": 99.45, "elapsed_time": "0:11:36", "remaining_time": "0:00:03"}
+{"current_steps": 365, "total_steps": 366, "loss": 0.3303, "lr": 5.697777236585711e-08, "epoch": 2.9734151329243352, "percentage": 99.73, "elapsed_time": "0:11:38", "remaining_time": "0:00:01"}
+{"current_steps": 366, "total_steps": 366, "loss": 0.558, "lr": 3.6468268237105366e-08, "epoch": 2.98159509202454, "percentage": 100.0, "elapsed_time": "0:11:40", "remaining_time": "0:00:00"}
+{"current_steps": 366, "total_steps": 366, "epoch": 2.98159509202454, "percentage": 100.0, "elapsed_time": "0:11:41", "remaining_time": "0:00:00"}

trainer_state.json

@@ -0,0 +1,2605 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 2.98159509202454,
+  "eval_steps": 500,
+  "global_step": 366,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.0081799591002045,
+      "grad_norm": NaN,
+      "learning_rate": 0.0,
+      "loss": 1.978,
+      "step": 1
+    },
+    {
+      "epoch": 0.016359918200409,
+      "grad_norm": NaN,
+      "learning_rate": 0.0,
+      "loss": 2.0417,
+      "step": 2
+    },
+    {
+      "epoch": 0.024539877300613498,
+      "grad_norm": 3.960280656814575,
+      "learning_rate": 0.0,
+      "loss": 2.4526,
+      "step": 3
+    },
+    {
+      "epoch": 0.032719836400818,
+      "grad_norm": 4.871670722961426,
+      "learning_rate": 2.702702702702703e-06,
+      "loss": 2.3708,
+      "step": 4
+    },
+    {
+      "epoch": 0.0408997955010225,
+      "grad_norm": NaN,
+      "learning_rate": 5.405405405405406e-06,
+      "loss": 2.7402,
+      "step": 5
+    },
+    {
+      "epoch": 0.049079754601226995,
+      "grad_norm": 2.961730718612671,
+      "learning_rate": 5.405405405405406e-06,
+      "loss": 2.6255,
+      "step": 6
+    },
+    {
+      "epoch": 0.05725971370143149,
+      "grad_norm": 2.4267706871032715,
+      "learning_rate": 8.108108108108109e-06,
+      "loss": 2.1672,
+      "step": 7
+    },
+    {
+      "epoch": 0.065439672801636,
+      "grad_norm": 4.815489768981934,
+      "learning_rate": 1.0810810810810812e-05,
+      "loss": 2.6853,
+      "step": 8
+    },
+    {
+      "epoch": 0.0736196319018405,
+      "grad_norm": 5.672784805297852,
+      "learning_rate": 1.3513513513513515e-05,
+      "loss": 2.5256,
+      "step": 9
+    },
+    {
+      "epoch": 0.081799591002045,
+      "grad_norm": 2.97552490234375,
+      "learning_rate": 1.6216216216216218e-05,
+      "loss": 2.1108,
+      "step": 10
+    },
+    {
+      "epoch": 0.08997955010224949,
+      "grad_norm": 2.305542469024658,
+      "learning_rate": 1.891891891891892e-05,
+      "loss": 2.2621,
+      "step": 11
+    },
+    {
+      "epoch": 0.09815950920245399,
+      "grad_norm": 2.330063581466675,
+      "learning_rate": 2.1621621621621624e-05,
+      "loss": 2.2973,
+      "step": 12
+    },
+    {
+      "epoch": 0.10633946830265849,
+      "grad_norm": 2.395848274230957,
+      "learning_rate": 2.4324324324324327e-05,
+      "loss": 2.2853,
+      "step": 13
+    },
+    {
+      "epoch": 0.11451942740286299,
+      "grad_norm": 3.5902745723724365,
+      "learning_rate": 2.702702702702703e-05,
+      "loss": 2.3628,
+      "step": 14
+    },
+    {
+      "epoch": 0.12269938650306748,
+      "grad_norm": 3.785466194152832,
+      "learning_rate": 2.9729729729729733e-05,
+      "loss": 1.6333,
+      "step": 15
+    },
+    {
+      "epoch": 0.130879345603272,
+      "grad_norm": 2.845073699951172,
+      "learning_rate": 3.2432432432432436e-05,
+      "loss": 1.6592,
+      "step": 16
+    },
+    {
+      "epoch": 0.1390593047034765,
+      "grad_norm": 2.9714362621307373,
+      "learning_rate": 3.513513513513514e-05,
+      "loss": 2.4303,
+      "step": 17
+    },
+    {
+      "epoch": 0.147239263803681,
+      "grad_norm": 3.2374515533447266,
+      "learning_rate": 3.783783783783784e-05,
+      "loss": 1.6276,
+      "step": 18
+    },
+    {
+      "epoch": 0.1554192229038855,
+      "grad_norm": 2.4501020908355713,
+      "learning_rate": 4.0540540540540545e-05,
+      "loss": 2.1786,
+      "step": 19
+    },
+    {
+      "epoch": 0.16359918200409,
+      "grad_norm": 1.689795970916748,
+      "learning_rate": 4.324324324324325e-05,
+      "loss": 1.7335,
+      "step": 20
+    },
+    {
+      "epoch": 0.17177914110429449,
+      "grad_norm": 1.5767645835876465,
+      "learning_rate": 4.594594594594595e-05,
+      "loss": 1.9214,
+      "step": 21
+    },
+    {
+      "epoch": 0.17995910020449898,
+      "grad_norm": 2.6853578090667725,
+      "learning_rate": 4.8648648648648654e-05,
+      "loss": 1.9089,
+      "step": 22
+    },
+    {
+      "epoch": 0.18813905930470348,
+      "grad_norm": 3.5681397914886475,
+      "learning_rate": 5.135135135135135e-05,
+      "loss": 2.092,
+      "step": 23
+    },
+    {
+      "epoch": 0.19631901840490798,
+      "grad_norm": 3.208242416381836,
+      "learning_rate": 5.405405405405406e-05,
+      "loss": 1.9184,
+      "step": 24
+    },
+    {
+      "epoch": 0.20449897750511248,
+      "grad_norm": 3.677274227142334,
+      "learning_rate": 5.6756756756756757e-05,
+      "loss": 1.6762,
+      "step": 25
+    },
+    {
+      "epoch": 0.21267893660531698,
+      "grad_norm": 1.9500216245651245,
+      "learning_rate": 5.9459459459459466e-05,
+      "loss": 1.8447,
+      "step": 26
+    },
+    {
+      "epoch": 0.22085889570552147,
+      "grad_norm": 2.72743821144104,
+      "learning_rate": 6.216216216216216e-05,
+      "loss": 1.659,
+      "step": 27
+    },
+    {
+      "epoch": 0.22903885480572597,
+      "grad_norm": 1.4266787767410278,
+      "learning_rate": 6.486486486486487e-05,
+      "loss": 1.5518,
+      "step": 28
+    },
+    {
+      "epoch": 0.23721881390593047,
+      "grad_norm": 1.8338373899459839,
+      "learning_rate": 6.756756756756757e-05,
+      "loss": 1.7725,
+      "step": 29
+    },
+    {
+      "epoch": 0.24539877300613497,
+      "grad_norm": 2.702836751937866,
+      "learning_rate": 7.027027027027028e-05,
+      "loss": 1.3214,
+      "step": 30
+    },
+    {
+      "epoch": 0.25357873210633947,
+      "grad_norm": 3.3664143085479736,
+      "learning_rate": 7.297297297297297e-05,
+      "loss": 1.5599,
+      "step": 31
+    },
+    {
+      "epoch": 0.261758691206544,
+      "grad_norm": 1.7983371019363403,
+      "learning_rate": 7.567567567567568e-05,
+      "loss": 1.4007,
+      "step": 32
+    },
+    {
+      "epoch": 0.26993865030674846,
+      "grad_norm": 1.4321403503417969,
+      "learning_rate": 7.837837837837838e-05,
+      "loss": 1.4209,
+      "step": 33
+    },
+    {
+      "epoch": 0.278118609406953,
+      "grad_norm": 1.7886905670166016,
+      "learning_rate": 8.108108108108109e-05,
+      "loss": 1.1963,
+      "step": 34
+    },
+    {
+      "epoch": 0.28629856850715746,
+      "grad_norm": 2.0502827167510986,
+      "learning_rate": 8.378378378378379e-05,
+      "loss": 1.6071,
+      "step": 35
+    },
+    {
+      "epoch": 0.294478527607362,
+      "grad_norm": 2.351100206375122,
+      "learning_rate": 8.64864864864865e-05,
+      "loss": 1.1902,
+      "step": 36
+    },
+    {
+      "epoch": 0.30265848670756645,
+      "grad_norm": 3.3446481227874756,
+      "learning_rate": 8.918918918918919e-05,
+      "loss": 1.1838,
+      "step": 37
+    },
+    {
+      "epoch": 0.310838445807771,
+      "grad_norm": 1.3906322717666626,
+      "learning_rate": 9.18918918918919e-05,
+      "loss": 0.9948,
+      "step": 38
+    },
+    {
+      "epoch": 0.31901840490797545,
+      "grad_norm": 1.9603602886199951,
+      "learning_rate": 9.45945945945946e-05,
+      "loss": 1.0997,
+      "step": 39
+    },
+    {
+      "epoch": 0.32719836400818,
+      "grad_norm": 1.3630380630493164,
+      "learning_rate": 9.729729729729731e-05,
+      "loss": 0.6781,
+      "step": 40
+    },
+    {
+      "epoch": 0.33537832310838445,
+      "grad_norm": 2.0062906742095947,
+      "learning_rate": 0.0001,
+      "loss": 1.4761,
+      "step": 41
+    },
+    {
+      "epoch": 0.34355828220858897,
+      "grad_norm": 1.2718311548233032,
+      "learning_rate": 9.99977204734326e-05,
+      "loss": 1.0233,
+      "step": 42
+    },
+    {
+      "epoch": 0.35173824130879344,
+      "grad_norm": 1.5002365112304688,
+      "learning_rate": 9.999088210158001e-05,
+      "loss": 0.8199,
+      "step": 43
+    },
+    {
+      "epoch": 0.35991820040899797,
+      "grad_norm": 1.430582880973816,
+      "learning_rate": 9.997948550797227e-05,
+      "loss": 1.1087,
+      "step": 44
+    },
+    {
+      "epoch": 0.36809815950920244,
+      "grad_norm": 2.0703279972076416,
+      "learning_rate": 9.996353173176289e-05,
+      "loss": 0.8149,
+      "step": 45
+    },
+    {
+      "epoch": 0.37627811860940696,
+      "grad_norm": 1.5086464881896973,
+      "learning_rate": 9.994302222763414e-05,
+      "loss": 0.9549,
+      "step": 46
+    },
+    {
+      "epoch": 0.38445807770961143,
+      "grad_norm": 1.5020016431808472,
+      "learning_rate": 9.991795886566441e-05,
+      "loss": 0.9542,
+      "step": 47
+    },
+    {
+      "epoch": 0.39263803680981596,
+      "grad_norm": 1.6791132688522339,
+      "learning_rate": 9.988834393115767e-05,
+      "loss": 0.8864,
+      "step": 48
+    },
+    {
+      "epoch": 0.40081799591002043,
+      "grad_norm": 2.7517454624176025,
+      "learning_rate": 9.98541801244351e-05,
+      "loss": 1.0826,
+      "step": 49
+    },
+    {
+      "epoch": 0.40899795501022496,
+      "grad_norm": 1.6971582174301147,
+      "learning_rate": 9.981547056058893e-05,
+      "loss": 0.9114,
+      "step": 50
+    },
+    {
+      "epoch": 0.4171779141104294,
+      "grad_norm": 1.4928799867630005,
+      "learning_rate": 9.977221876919833e-05,
+      "loss": 0.9755,
+      "step": 51
+    },
+    {
+      "epoch": 0.42535787321063395,
+      "grad_norm": 1.1075421571731567,
+      "learning_rate": 9.972442869400759e-05,
+      "loss": 0.741,
+      "step": 52
+    },
+    {
+      "epoch": 0.4335378323108384,
+      "grad_norm": 1.3488271236419678,
+      "learning_rate": 9.967210469256656e-05,
+      "loss": 1.0441,
+      "step": 53
+    },
+    {
+      "epoch": 0.44171779141104295,
+      "grad_norm": 3.3010079860687256,
+      "learning_rate": 9.961525153583327e-05,
+      "loss": 1.2885,
+      "step": 54
+    },
+    {
+      "epoch": 0.4498977505112474,
+      "grad_norm": 1.24274742603302,
+      "learning_rate": 9.9553874407739e-05,
+      "loss": 0.8107,
+      "step": 55
+    },
+    {
+      "epoch": 0.45807770961145194,
+      "grad_norm": 1.1659082174301147,
+      "learning_rate": 9.948797890471551e-05,
+      "loss": 0.6129,
+      "step": 56
+    },
+    {
+      "epoch": 0.4662576687116564,
+      "grad_norm": 1.5525306463241577,
+      "learning_rate": 9.941757103518478e-05,
+      "loss": 0.9262,
+      "step": 57
+    },
+    {
+      "epoch": 0.47443762781186094,
+      "grad_norm": 1.3923064470291138,
+      "learning_rate": 9.93426572190112e-05,
+      "loss": 0.7552,
+      "step": 58
+    },
+    {
+      "epoch": 0.48261758691206547,
+      "grad_norm": 1.166669487953186,
+      "learning_rate": 9.926324428691611e-05,
+      "loss": 0.6346,
+      "step": 59
+    },
+    {
+      "epoch": 0.49079754601226994,
+      "grad_norm": 2.103994131088257,
+      "learning_rate": 9.917933947985507e-05,
+      "loss": 0.8199,
+      "step": 60
+    },
+    {
+      "epoch": 0.49897750511247446,
+      "grad_norm": 1.633812665939331,
+      "learning_rate": 9.909095044835754e-05,
+      "loss": 0.7485,
+      "step": 61
+    },
+    {
+      "epoch": 0.5071574642126789,
+      "grad_norm": 1.4533647298812866,
+      "learning_rate": 9.899808525182935e-05,
+      "loss": 1.1649,
+      "step": 62
+    },
+    {
+      "epoch": 0.5153374233128835,
+      "grad_norm": 1.6267237663269043,
+      "learning_rate": 9.890075235781779e-05,
+      "loss": 1.1159,
+      "step": 63
+    },
+    {
+      "epoch": 0.523517382413088,
+      "grad_norm": 1.2796165943145752,
+      "learning_rate": 9.879896064123961e-05,
+      "loss": 0.9613,
+      "step": 64
+    },
+    {
+      "epoch": 0.5316973415132924,
+      "grad_norm": 1.3240866661071777,
+      "learning_rate": 9.869271938357167e-05,
+      "loss": 1.047,
+      "step": 65
+    },
+    {
+      "epoch": 0.5398773006134969,
+      "grad_norm": 1.190612554550171,
+      "learning_rate": 9.858203827200476e-05,
+      "loss": 1.1846,
+      "step": 66
+    },
+    {
+      "epoch": 0.5480572597137015,
+      "grad_norm": 1.1655223369598389,
+      "learning_rate": 9.846692739856024e-05,
+      "loss": 0.9566,
+      "step": 67
+    },
+    {
+      "epoch": 0.556237218813906,
+      "grad_norm": 1.2617253065109253,
+      "learning_rate": 9.834739725916988e-05,
+      "loss": 1.1108,
+      "step": 68
+    },
+    {
+      "epoch": 0.5644171779141104,
+      "grad_norm": 1.3576512336730957,
+      "learning_rate": 9.822345875271883e-05,
+      "loss": 1.1265,
+      "step": 69
+    },
+    {
+      "epoch": 0.5725971370143149,
+      "grad_norm": 1.4342156648635864,
+      "learning_rate": 9.809512318005181e-05,
+      "loss": 0.7757,
+      "step": 70
+    },
+    {
+      "epoch": 0.5807770961145194,
+      "grad_norm": 1.0733706951141357,
+      "learning_rate": 9.796240224294271e-05,
+      "loss": 0.9006,
+      "step": 71
+    },
+    {
+      "epoch": 0.588957055214724,
+      "grad_norm": 1.323440432548523,
+      "learning_rate": 9.782530804302763e-05,
+      "loss": 0.9322,
+      "step": 72
+    },
+    {
+      "epoch": 0.5971370143149284,
+      "grad_norm": 1.2899342775344849,
+      "learning_rate": 9.768385308070138e-05,
+      "loss": 0.8403,
+      "step": 73
+    },
+    {
+      "epoch": 0.6053169734151329,
+      "grad_norm": 1.2755167484283447,
+      "learning_rate": 9.753805025397779e-05,
+      "loss": 0.8397,
+      "step": 74
+    },
+    {
+      "epoch": 0.6134969325153374,
+      "grad_norm": 1.265972375869751,
+      "learning_rate": 9.738791285731352e-05,
+      "loss": 0.8143,
+      "step": 75
+    },
+    {
+      "epoch": 0.621676891615542,
+      "grad_norm": 1.1493557691574097,
+      "learning_rate": 9.723345458039594e-05,
+      "loss": 1.059,
+      "step": 76
+    },
+    {
+      "epoch": 0.6298568507157464,
+      "grad_norm": 1.1361910104751587,
+      "learning_rate": 9.707468950689491e-05,
+      "loss": 0.9112,
+      "step": 77
+    },
+    {
+      "epoch": 0.6380368098159509,
+      "grad_norm": 1.4090393781661987,
+      "learning_rate": 9.691163211317853e-05,
+      "loss": 0.7847,
+      "step": 78
+    },
+    {
+      "epoch": 0.6462167689161554,
+      "grad_norm": 1.300688624382019,
+      "learning_rate": 9.674429726699323e-05,
+      "loss": 0.9806,
+      "step": 79
+    },
+    {
+      "epoch": 0.65439672801636,
+      "grad_norm": 1.0143762826919556,
+      "learning_rate": 9.657270022610813e-05,
+      "loss": 0.6648,
+      "step": 80
+    },
+    {
+      "epoch": 0.6625766871165644,
+      "grad_norm": 1.2058460712432861,
+      "learning_rate": 9.63968566369238e-05,
+      "loss": 0.9702,
+      "step": 81
+    },
+    {
+      "epoch": 0.6707566462167689,
+      "grad_norm": 1.0876555442810059,
+      "learning_rate": 9.62167825330455e-05,
+      "loss": 0.7861,
+      "step": 82
+    },
+    {
+      "epoch": 0.6789366053169734,
+      "grad_norm": 1.502429723739624,
+      "learning_rate": 9.603249433382144e-05,
+      "loss": 1.1252,
+      "step": 83
+    },
+    {
+      "epoch": 0.6871165644171779,
+      "grad_norm": 4.1136860847473145,
+      "learning_rate": 9.584400884284545e-05,
+      "loss": 0.7109,
+      "step": 84
+    },
+    {
+      "epoch": 0.6952965235173824,
+      "grad_norm": 0.9980061650276184,
+      "learning_rate": 9.56513432464249e-05,
+      "loss": 0.6421,
+      "step": 85
+    },
+    {
+      "epoch": 0.7034764826175869,
+      "grad_norm": 1.1087136268615723,
+      "learning_rate": 9.545451511201364e-05,
+      "loss": 0.6337,
+      "step": 86
+    },
+    {
+      "epoch": 0.7116564417177914,
+      "grad_norm": 1.4353466033935547,
+      "learning_rate": 9.525354238661009e-05,
+      "loss": 1.0757,
+      "step": 87
+    },
+    {
+      "epoch": 0.7198364008179959,
+      "grad_norm": 1.1413462162017822,
+      "learning_rate": 9.504844339512095e-05,
+      "loss": 0.7056,
+      "step": 88
+    },
+    {
+      "epoch": 0.7280163599182005,
+      "grad_norm": 1.5157971382141113,
+      "learning_rate": 9.483923683869024e-05,
+      "loss": 0.8767,
+      "step": 89
+    },
+    {
+      "epoch": 0.7361963190184049,
+      "grad_norm": 0.999251663684845,
+      "learning_rate": 9.462594179299406e-05,
+      "loss": 0.922,
+      "step": 90
+    },
+    {
+      "epoch": 0.7443762781186094,
+      "grad_norm": 1.2393922805786133,
+      "learning_rate": 9.440857770650138e-05,
+      "loss": 0.8301,
+      "step": 91
+    },
+    {
+      "epoch": 0.7525562372188139,
+      "grad_norm": 1.1513807773590088,
+      "learning_rate": 9.418716439870057e-05,
+      "loss": 0.5308,
+      "step": 92
+    },
+    {
+      "epoch": 0.7607361963190185,
+      "grad_norm": 1.4229981899261475,
+      "learning_rate": 9.396172205829234e-05,
+      "loss": 1.1116,
+      "step": 93
+    },
+    {
+      "epoch": 0.7689161554192229,
+      "grad_norm": 1.4916852712631226,
+      "learning_rate": 9.373227124134888e-05,
+      "loss": 0.8806,
+      "step": 94
+    },
+    {
+      "epoch": 0.7770961145194274,
+      "grad_norm": 1.693434715270996,
+      "learning_rate": 9.34988328694395e-05,
+      "loss": 0.7924,
+      "step": 95
+    },
+    {
+      "epoch": 0.7852760736196319,
+      "grad_norm": 1.1455564498901367,
+      "learning_rate": 9.326142822772302e-05,
+      "loss": 0.7091,
+      "step": 96
+    },
+    {
+      "epoch": 0.7934560327198364,
+      "grad_norm": 1.031115174293518,
+      "learning_rate": 9.302007896300698e-05,
+      "loss": 0.8828,
+      "step": 97
+    },
+    {
+      "epoch": 0.8016359918200409,
+      "grad_norm": 1.5678527355194092,
+      "learning_rate": 9.27748070817738e-05,
+      "loss": 1.0531,
+      "step": 98
+    },
+    {
+      "epoch": 0.8098159509202454,
+      "grad_norm": 1.2558964490890503,
+      "learning_rate": 9.252563494817425e-05,
+      "loss": 0.9222,
+      "step": 99
+    },
+    {
+      "epoch": 0.8179959100204499,
+      "grad_norm": 1.2573111057281494,
+      "learning_rate": 9.227258528198831e-05,
+      "loss": 0.8988,
+      "step": 100
+    },
+    {
+      "epoch": 0.8261758691206544,
+      "grad_norm": 1.3229575157165527,
+      "learning_rate": 9.201568115655342e-05,
+      "loss": 1.0139,
+      "step": 101
+    },
+    {
+      "epoch": 0.8343558282208589,
+      "grad_norm": 0.913388729095459,
+      "learning_rate": 9.175494599666077e-05,
+      "loss": 0.6802,
+      "step": 102
+    },
+    {
+      "epoch": 0.8425357873210634,
+      "grad_norm": 3.0073134899139404,
+      "learning_rate": 9.149040357641929e-05,
+      "loss": 0.8834,
+      "step": 103
+    },
+    {
+      "epoch": 0.8507157464212679,
+      "grad_norm": 0.9436314105987549,
+      "learning_rate": 9.122207801708802e-05,
+      "loss": 0.6559,
+      "step": 104
+    },
+    {
+      "epoch": 0.8588957055214724,
+      "grad_norm": 1.1913410425186157,
+      "learning_rate": 9.094999378487659e-05,
+      "loss": 0.7427,
+      "step": 105
+    },
+    {
+      "epoch": 0.8670756646216768,
+      "grad_norm": 1.03123140335083,
+      "learning_rate": 9.067417568871445e-05,
+      "loss": 0.6253,
+      "step": 106
+    },
+    {
+      "epoch": 0.8752556237218814,
+      "grad_norm": 0.9424878358840942,
+      "learning_rate": 9.03946488779887e-05,
+      "loss": 0.7601,
+      "step": 107
+    },
+    {
+      "epoch": 0.8834355828220859,
+      "grad_norm": 1.2202138900756836,
+      "learning_rate": 9.011143884025101e-05,
+      "loss": 1.024,
+      "step": 108
+    },
+    {
+      "epoch": 0.8916155419222904,
+      "grad_norm": 1.3849170207977295,
+      "learning_rate": 8.982457139889357e-05,
+      "loss": 0.8283,
+      "step": 109
+    },
+    {
+      "epoch": 0.8997955010224948,
+      "grad_norm": 1.2288554906845093,
+      "learning_rate": 8.953407271079455e-05,
+      "loss": 0.8297,
+      "step": 110
+    },
+    {
+      "epoch": 0.9079754601226994,
+      "grad_norm": 0.9206739664077759,
+      "learning_rate": 8.923996926393305e-05,
+      "loss": 0.5966,
+      "step": 111
+    },
+    {
+      "epoch": 0.9161554192229039,
+      "grad_norm": 0.9282044172286987,
+      "learning_rate": 8.894228787497389e-05,
+      "loss": 0.7775,
+      "step": 112
+    },
+    {
+      "epoch": 0.9243353783231084,
+      "grad_norm": 1.2421815395355225,
+      "learning_rate": 8.864105568682244e-05,
+      "loss": 0.7838,
+      "step": 113
+    },
+    {
+      "epoch": 0.9325153374233128,
+      "grad_norm": 1.1172124147415161,
+      "learning_rate": 8.833630016614976e-05,
+      "loss": 0.5921,
+      "step": 114
+    },
+    {
+      "epoch": 0.9406952965235174,
+      "grad_norm": 1.310473918914795,
+      "learning_rate": 8.802804910088809e-05,
+      "loss": 1.0578,
+      "step": 115
+    },
+    {
+      "epoch": 0.9488752556237219,
+      "grad_norm": 1.1656848192214966,
+      "learning_rate": 8.771633059769711e-05,
+      "loss": 0.8205,
+      "step": 116
+    },
+    {
+      "epoch": 0.9570552147239264,
+      "grad_norm": 1.0159672498703003,
+      "learning_rate": 8.740117307940123e-05,
+      "loss": 0.8237,
+      "step": 117
+    },
+    {
+      "epoch": 0.9652351738241309,
+      "grad_norm": 1.2568827867507935,
+      "learning_rate": 8.708260528239788e-05,
+      "loss": 1.0473,
+      "step": 118
+    },
+    {
+      "epoch": 0.9734151329243353,
+      "grad_norm": 1.2711869478225708,
+      "learning_rate": 8.676065625403733e-05,
+      "loss": 1.0789,
+      "step": 119
+    },
+    {
+      "epoch": 0.9815950920245399,
+      "grad_norm": 1.2562803030014038,
+      "learning_rate": 8.64353553499741e-05,
+      "loss": 0.5315,
+      "step": 120
+    },
+    {
+      "epoch": 0.9897750511247444,
+      "grad_norm": 1.4107680320739746,
+      "learning_rate": 8.610673223149034e-05,
+      "loss": 0.9738,
+      "step": 121
+    },
+    {
+      "epoch": 0.9979550102249489,
+      "grad_norm": 1.0771377086639404,
+      "learning_rate": 8.577481686279123e-05,
+      "loss": 0.6114,
+      "step": 122
+    },
+    {
+      "epoch": 1.0,
+      "grad_norm": 0.6168379187583923,
+      "learning_rate": 8.543963950827279e-05,
+      "loss": 0.1857,
+      "step": 123
+    },
+    {
+      "epoch": 1.0081799591002045,
+      "grad_norm": 1.0099263191223145,
+      "learning_rate": 8.510123072976239e-05,
+      "loss": 0.77,
+      "step": 124
+    },
+    {
+      "epoch": 1.016359918200409,
+      "grad_norm": 1.0608845949172974,
+      "learning_rate": 8.475962138373213e-05,
+      "loss": 0.588,
+      "step": 125
+    },
+    {
+      "epoch": 1.0245398773006136,
+      "grad_norm": 1.2696729898452759,
+      "learning_rate": 8.441484261848514e-05,
+      "loss": 0.8879,
+      "step": 126
+    },
+    {
+      "epoch": 1.032719836400818,
+      "grad_norm": 1.556289792060852,
+      "learning_rate": 8.406692587131568e-05,
+      "loss": 1.1292,
+      "step": 127
+    },
+    {
+      "epoch": 1.0408997955010224,
+      "grad_norm": 1.027779459953308,
+      "learning_rate": 8.371590286564247e-05,
+      "loss": 0.6589,
+      "step": 128
+    },
+    {
+      "epoch": 1.049079754601227,
+      "grad_norm": 1.1132937669754028,
+      "learning_rate": 8.336180560811619e-05,
+      "loss": 0.6267,
+      "step": 129
+    },
+    {
+      "epoch": 1.0572597137014315,
+      "grad_norm": 1.7684708833694458,
+      "learning_rate": 8.30046663857011e-05,
+      "loss": 0.7399,
+      "step": 130
+    },
+    {
+      "epoch": 1.065439672801636,
+      "grad_norm": 1.372917890548706,
+      "learning_rate": 8.264451776273104e-05,
+      "loss": 0.8849,
+      "step": 131
+    },
+    {
+      "epoch": 1.0736196319018405,
+      "grad_norm": 1.1678389310836792,
+      "learning_rate": 8.228139257794012e-05,
+      "loss": 0.901,
+      "step": 132
+    },
+    {
+      "epoch": 1.081799591002045,
+      "grad_norm": 1.0002321004867554,
+      "learning_rate": 8.191532394146865e-05,
+      "loss": 0.3923,
+      "step": 133
+    },
+    {
+      "epoch": 1.0899795501022496,
+      "grad_norm": 1.0493892431259155,
+      "learning_rate": 8.154634523184388e-05,
+      "loss": 0.77,
+      "step": 134
+    },
+    {
+      "epoch": 1.098159509202454,
+      "grad_norm": 1.4020309448242188,
+      "learning_rate": 8.117449009293668e-05,
+      "loss": 0.7849,
+      "step": 135
+    },
+    {
+      "epoch": 1.1063394683026584,
+      "grad_norm": 1.3133962154388428,
+      "learning_rate": 8.07997924308938e-05,
+      "loss": 0.7912,
+      "step": 136
+    },
+    {
+      "epoch": 1.114519427402863,
+      "grad_norm": 1.3940467834472656,
+      "learning_rate": 8.042228641104622e-05,
+      "loss": 0.7142,
+      "step": 137
+    },
+    {
+      "epoch": 1.1226993865030674,
+      "grad_norm": 1.3877304792404175,
+      "learning_rate": 8.004200645479403e-05,
+      "loss": 0.5454,
+      "step": 138
+    },
+    {
+      "epoch": 1.130879345603272,
+      "grad_norm": 1.1132571697235107,
+      "learning_rate": 7.965898723646776e-05,
+      "loss": 0.7387,
+      "step": 139
+    },
+    {
+      "epoch": 1.1390593047034765,
+      "grad_norm": 1.206382155418396,
+      "learning_rate": 7.927326368016677e-05,
+      "loss": 0.7271,
+      "step": 140
+    },
+    {
+      "epoch": 1.147239263803681,
+      "grad_norm": 1.2813421487808228,
+      "learning_rate": 7.888487095657484e-05,
+      "loss": 0.8452,
+      "step": 141
+    },
+    {
+      "epoch": 1.1554192229038855,
+      "grad_norm": 1.1103274822235107,
+      "learning_rate": 7.849384447975321e-05,
+      "loss": 0.5735,
+      "step": 142
+    },
+    {
+      "epoch": 1.16359918200409,
+      "grad_norm": 1.1904572248458862,
+      "learning_rate": 7.810021990391164e-05,
+      "loss": 0.486,
+      "step": 143
+    },
+    {
+      "epoch": 1.1717791411042944,
+      "grad_norm": 1.361222743988037,
+      "learning_rate": 7.770403312015721e-05,
+      "loss": 0.9265,
+      "step": 144
+    },
+    {
+      "epoch": 1.179959100204499,
+      "grad_norm": 1.1453652381896973,
+      "learning_rate": 7.73053202532219e-05,
+      "loss": 0.6186,
+      "step": 145
+    },
+    {
+      "epoch": 1.1881390593047034,
+      "grad_norm": 1.2070741653442383,
+      "learning_rate": 7.690411765816864e-05,
+      "loss": 0.7012,
+      "step": 146
+    },
+    {
+      "epoch": 1.196319018404908,
+      "grad_norm": 1.4246371984481812,
+      "learning_rate": 7.650046191707641e-05,
+      "loss": 0.7644,
+      "step": 147
+    },
+    {
+      "epoch": 1.2044989775051125,
+      "grad_norm": 1.2275187969207764,
+      "learning_rate": 7.60943898357046e-05,
+      "loss": 0.614,
+      "step": 148
+    },
+    {
+      "epoch": 1.212678936605317,
+      "grad_norm": 1.292330265045166,
+      "learning_rate": 7.568593844013718e-05,
+      "loss": 0.6722,
+      "step": 149
+    },
+    {
+      "epoch": 1.2208588957055215,
+      "grad_norm": 1.54197359085083,
+      "learning_rate": 7.527514497340642e-05,
+      "loss": 0.6981,
+      "step": 150
+    },
+    {
+      "epoch": 1.229038854805726,
+      "grad_norm": 1.605914831161499,
+      "learning_rate": 7.48620468920972e-05,
+      "loss": 0.7524,
+      "step": 151
+    },
+    {
+      "epoch": 1.2372188139059306,
+      "grad_norm": 1.2442176342010498,
+      "learning_rate": 7.444668186293153e-05,
+      "loss": 0.6238,
+      "step": 152
+    },
+    {
+      "epoch": 1.2453987730061349,
+      "grad_norm": 1.4838721752166748,
+      "learning_rate": 7.402908775933419e-05,
+      "loss": 0.7599,
+      "step": 153
+    },
+    {
+      "epoch": 1.2535787321063394,
+      "grad_norm": 1.8454984426498413,
+      "learning_rate": 7.360930265797935e-05,
+      "loss": 1.1331,
+      "step": 154
+    },
+    {
+      "epoch": 1.261758691206544,
+      "grad_norm": 1.3571646213531494,
+      "learning_rate": 7.31873648353186e-05,
+      "loss": 0.6468,
+      "step": 155
+    },
+    {
+      "epoch": 1.2699386503067485,
+      "grad_norm": 1.3795866966247559,
+      "learning_rate": 7.276331276409106e-05,
+      "loss": 0.7253,
+      "step": 156
+    },
+    {
+      "epoch": 1.278118609406953,
+      "grad_norm": 1.4821308851242065,
+      "learning_rate": 7.23371851098152e-05,
+      "loss": 0.842,
+      "step": 157
+    },
+    {
+      "epoch": 1.2862985685071575,
+      "grad_norm": 1.0921138525009155,
+      "learning_rate": 7.190902072726335e-05,
+      "loss": 0.5379,
+      "step": 158
+    },
+    {
+      "epoch": 1.294478527607362,
+      "grad_norm": 1.5662935972213745,
+      "learning_rate": 7.147885865691899e-05,
+      "loss": 0.918,
+      "step": 159
+    },
+    {
+      "epoch": 1.3026584867075663,
+      "grad_norm": 1.3333555459976196,
+      "learning_rate": 7.104673812141675e-05,
+      "loss": 0.6727,
+      "step": 160
+    },
+    {
+      "epoch": 1.310838445807771,
+      "grad_norm": 1.1487497091293335,
+      "learning_rate": 7.061269852196632e-05,
+      "loss": 0.4279,
+      "step": 161
+    },
+    {
+      "epoch": 1.3190184049079754,
+      "grad_norm": 1.1033565998077393,
+      "learning_rate": 7.017677943475961e-05,
+      "loss": 0.6372,
+      "step": 162
+    },
+    {
+      "epoch": 1.32719836400818,
+      "grad_norm": 1.2100588083267212,
+      "learning_rate": 6.973902060736226e-05,
+      "loss": 0.7071,
+      "step": 163
+    },
+    {
+      "epoch": 1.3353783231083844,
+      "grad_norm": 1.421066403388977,
+      "learning_rate": 6.929946195508932e-05,
+      "loss": 0.767,
+      "step": 164
+    },
+    {
+      "epoch": 1.343558282208589,
+      "grad_norm": 1.2306902408599854,
+      "learning_rate": 6.885814355736586e-05,
+      "loss": 0.5587,
+      "step": 165
+    },
+    {
+      "epoch": 1.3517382413087935,
+      "grad_norm": 1.5315287113189697,
+      "learning_rate": 6.841510565407235e-05,
+      "loss": 0.7519,
+      "step": 166
+    },
+    {
+      "epoch": 1.359918200408998,
+      "grad_norm": 1.2497670650482178,
+      "learning_rate": 6.797038864187564e-05,
+      "loss": 0.5612,
+      "step": 167
+    },
+    {
+      "epoch": 1.3680981595092025,
+      "grad_norm": 1.6106078624725342,
+      "learning_rate": 6.752403307054549e-05,
+      "loss": 0.7194,
+      "step": 168
+    },
+    {
+      "epoch": 1.3762781186094069,
+      "grad_norm": 1.2407530546188354,
+      "learning_rate": 6.707607963925724e-05,
+      "loss": 0.531,
+      "step": 169
+    },
+    {
+      "epoch": 1.3844580777096114,
+      "grad_norm": 1.663898229598999,
+      "learning_rate": 6.66265691928808e-05,
+      "loss": 0.7906,
+      "step": 170
+    },
+    {
+      "epoch": 1.392638036809816,
+      "grad_norm": 1.3650121688842773,
+      "learning_rate": 6.617554271825636e-05,
+      "loss": 0.7207,
+      "step": 171
+    },
+    {
+      "epoch": 1.4008179959100204,
+      "grad_norm": 1.1001300811767578,
+      "learning_rate": 6.572304134045717e-05,
+      "loss": 0.5145,
+      "step": 172
+    },
+    {
+      "epoch": 1.408997955010225,
+      "grad_norm": 1.0687707662582397,
+      "learning_rate": 6.526910631903973e-05,
+      "loss": 0.3521,
+      "step": 173
+    },
+    {
+      "epoch": 1.4171779141104295,
+      "grad_norm": 1.2442213296890259,
+      "learning_rate": 6.481377904428171e-05,
+      "loss": 0.7026,
+      "step": 174
+    },
+    {
+      "epoch": 1.425357873210634,
+      "grad_norm": 1.31452214717865,
+      "learning_rate": 6.435710103340786e-05,
+      "loss": 0.7313,
+      "step": 175
+    },
+    {
+      "epoch": 1.4335378323108383,
+      "grad_norm": 1.5573769807815552,
+      "learning_rate": 6.389911392680456e-05,
+      "loss": 0.7659,
+      "step": 176
+    },
+    {
+      "epoch": 1.441717791411043,
+      "grad_norm": 1.2089431285858154,
+      "learning_rate": 6.343985948422287e-05,
+      "loss": 0.6916,
+      "step": 177
+    },
+    {
+      "epoch": 1.4498977505112474,
+      "grad_norm": 1.5785194635391235,
+      "learning_rate": 6.297937958097094e-05,
+      "loss": 0.8101,
+      "step": 178
+    },
+    {
+      "epoch": 1.4580777096114519,
+      "grad_norm": 1.4134269952774048,
+      "learning_rate": 6.251771620409563e-05,
+      "loss": 0.7504,
+      "step": 179
+    },
+    {
+      "epoch": 1.4662576687116564,
+      "grad_norm": 1.4751485586166382,
+      "learning_rate": 6.205491144855432e-05,
+      "loss": 0.5948,
+      "step": 180
+    },
+    {
+      "epoch": 1.474437627811861,
+      "grad_norm": 1.31548273563385,
+      "learning_rate": 6.159100751337642e-05,
+      "loss": 0.7057,
+      "step": 181
+    },
+    {
+      "epoch": 1.4826175869120655,
+      "grad_norm": 1.8151648044586182,
+      "learning_rate": 6.112604669781572e-05,
+      "loss": 0.9037,
+      "step": 182
+    },
+    {
+      "epoch": 1.49079754601227,
+      "grad_norm": 1.3681972026824951,
+      "learning_rate": 6.0660071397493514e-05,
+      "loss": 0.7223,
+      "step": 183
+    },
+    {
+      "epoch": 1.4989775051124745,
+      "grad_norm": 1.6292760372161865,
+      "learning_rate": 6.019312410053286e-05,
+      "loss": 0.6083,
+      "step": 184
+    },
+    {
+      "epoch": 1.5071574642126788,
+      "grad_norm": 1.8144514560699463,
+      "learning_rate": 5.972524738368452e-05,
+      "loss": 0.7662,
+      "step": 185
+    },
+    {
+      "epoch": 1.5153374233128836,
+      "grad_norm": 1.650654911994934,
+      "learning_rate": 5.925648390844476e-05,
+      "loss": 0.902,
+      "step": 186
+    },
+    {
+      "epoch": 1.5235173824130879,
+      "grad_norm": 1.4780257940292358,
+      "learning_rate": 5.878687641716538e-05,
+      "loss": 0.6566,
+      "step": 187
+    },
+    {
+      "epoch": 1.5316973415132924,
+      "grad_norm": 1.1706862449645996,
+      "learning_rate": 5.831646772915651e-05,
+      "loss": 0.4189,
+      "step": 188
+    },
+    {
+      "epoch": 1.539877300613497,
+      "grad_norm": 1.287718653678894,
+      "learning_rate": 5.7845300736782204e-05,
+      "loss": 0.5549,
+      "step": 189
+    },
+    {
+      "epoch": 1.5480572597137015,
+      "grad_norm": 1.3776918649673462,
+      "learning_rate": 5.737341840154956e-05,
+      "loss": 0.5456,
+      "step": 190
+    },
+    {
+      "epoch": 1.556237218813906,
+      "grad_norm": 1.2569301128387451,
+      "learning_rate": 5.6900863750191347e-05,
+      "loss": 0.6808,
+      "step": 191
+    },
+    {
+      "epoch": 1.5644171779141103,
+      "grad_norm": 1.7013508081436157,
+      "learning_rate": 5.642767987074288e-05,
+      "loss": 0.7974,
+      "step": 192
+    },
+    {
+      "epoch": 1.572597137014315,
+      "grad_norm": 1.5190353393554688,
+      "learning_rate": 5.5953909908613114e-05,
+      "loss": 0.5416,
+      "step": 193
+    },
+    {
+      "epoch": 1.5807770961145193,
+      "grad_norm": 1.4736334085464478,
+      "learning_rate": 5.547959706265068e-05,
+      "loss": 0.6788,
+      "step": 194
+    },
+    {
+      "epoch": 1.588957055214724,
+      "grad_norm": 2.006303548812866,
+      "learning_rate": 5.5004784581204927e-05,
+      "loss": 0.9634,
+      "step": 195
+    },
+    {
+      "epoch": 1.5971370143149284,
+      "grad_norm": 1.3578423261642456,
+      "learning_rate": 5.4529515758182506e-05,
+      "loss": 0.6563,
+      "step": 196
+    },
+    {
+      "epoch": 1.605316973415133,
+      "grad_norm": 1.4116990566253662,
+      "learning_rate": 5.405383392909973e-05,
+      "loss": 0.6062,
+      "step": 197
+    },
+    {
+      "epoch": 1.6134969325153374,
+      "grad_norm": 1.2362536191940308,
+      "learning_rate": 5.357778246713131e-05,
+      "loss": 0.4829,
+      "step": 198
+    },
+    {
+      "epoch": 1.621676891615542,
+      "grad_norm": 1.1780372858047485,
+      "learning_rate": 5.310140477915544e-05,
+      "loss": 0.465,
+      "step": 199
+    },
+    {
+      "epoch": 1.6298568507157465,
+      "grad_norm": 1.3919291496276855,
+      "learning_rate": 5.262474430179597e-05,
+      "loss": 0.6967,
+      "step": 200
+    },
+    {
+      "epoch": 1.6380368098159508,
+      "grad_norm": 1.7452629804611206,
+      "learning_rate": 5.214784449746174e-05,
+      "loss": 0.9096,
+      "step": 201
+    },
+    {
+      "epoch": 1.6462167689161555,
+      "grad_norm": 1.4730846881866455,
+      "learning_rate": 5.167074885038373e-05,
+      "loss": 0.7473,
+      "step": 202
+    },
+    {
+      "epoch": 1.6543967280163598,
+      "grad_norm": 1.5404870510101318,
+      "learning_rate": 5.119350086265004e-05,
+      "loss": 0.6233,
+      "step": 203
+    },
+    {
+      "epoch": 1.6625766871165644,
+      "grad_norm": 1.4780898094177246,
+      "learning_rate": 5.0716144050239375e-05,
+      "loss": 0.7599,
+      "step": 204
+    },
+    {
+      "epoch": 1.670756646216769,
+      "grad_norm": 1.194542407989502,
+      "learning_rate": 5.023872193905316e-05,
+      "loss": 0.5638,
+      "step": 205
+    },
+    {
+      "epoch": 1.6789366053169734,
+      "grad_norm": 1.3504347801208496,
+      "learning_rate": 4.976127806094684e-05,
+      "loss": 0.4701,
+      "step": 206
+    },
+    {
+      "epoch": 1.687116564417178,
+      "grad_norm": 2.1446099281311035,
+      "learning_rate": 4.928385594976063e-05,
+      "loss": 0.9383,
+      "step": 207
+    },
+    {
+      "epoch": 1.6952965235173822,
+      "grad_norm": 1.4745142459869385,
+      "learning_rate": 4.880649913734996e-05,
+      "loss": 0.5817,
+      "step": 208
+    },
+    {
+      "epoch": 1.703476482617587,
+      "grad_norm": 1.3029444217681885,
+      "learning_rate": 4.832925114961629e-05,
+      "loss": 0.3481,
+      "step": 209
+    },
+    {
+      "epoch": 1.7116564417177913,
+      "grad_norm": 1.1414580345153809,
+      "learning_rate": 4.785215550253826e-05,
+      "loss": 0.4348,
+      "step": 210
+    },
+    {
+      "epoch": 1.719836400817996,
+      "grad_norm": 1.4996669292449951,
+      "learning_rate": 4.7375255698204045e-05,
+      "loss": 0.653,
+      "step": 211
+    },
+    {
+      "epoch": 1.7280163599182004,
+      "grad_norm": 1.66719388961792,
+      "learning_rate": 4.6898595220844574e-05,
+      "loss": 0.7181,
+      "step": 212
+    },
+    {
+      "epoch": 1.7361963190184049,
+      "grad_norm": 1.476560354232788,
+      "learning_rate": 4.64222175328687e-05,
+      "loss": 0.6183,
+      "step": 213
+    },
+    {
+      "epoch": 1.7443762781186094,
+      "grad_norm": 1.7405219078063965,
+      "learning_rate": 4.594616607090028e-05,
+      "loss": 0.689,
+      "step": 214
+    },
+    {
+      "epoch": 1.752556237218814,
+      "grad_norm": 1.1866732835769653,
+      "learning_rate": 4.547048424181751e-05,
+      "loss": 0.4616,
+      "step": 215
+    },
+    {
+      "epoch": 1.7607361963190185,
+      "grad_norm": 1.7068077325820923,
+      "learning_rate": 4.4995215418795085e-05,
+      "loss": 0.6318,
+      "step": 216
+    },
+    {
+      "epoch": 1.7689161554192228,
+      "grad_norm": 1.4736443758010864,
+      "learning_rate": 4.452040293734934e-05,
+      "loss": 0.4611,
+      "step": 217
+    },
+    {
+      "epoch": 1.7770961145194275,
+      "grad_norm": 0.8084559440612793,
+      "learning_rate": 4.404609009138689e-05,
+      "loss": 0.1962,
+      "step": 218
+    },
+    {
+      "epoch": 1.7852760736196318,
+      "grad_norm": 1.1126220226287842,
+      "learning_rate": 4.357232012925714e-05,
+      "loss": 0.3804,
+      "step": 219
+    },
+    {
+      "epoch": 1.7934560327198366,
+      "grad_norm": 1.4977810382843018,
+      "learning_rate": 4.3099136249808665e-05,
+      "loss": 0.5431,
+      "step": 220
+    },
+    {
+      "epoch": 1.8016359918200409,
+      "grad_norm": 1.47788405418396,
+      "learning_rate": 4.262658159845046e-05,
+      "loss": 0.6498,
+      "step": 221
+    },
+    {
+      "epoch": 1.8098159509202454,
+      "grad_norm": 1.2339309453964233,
+      "learning_rate": 4.215469926321779e-05,
+      "loss": 0.4812,
+      "step": 222
+    },
+    {
+      "epoch": 1.81799591002045,
+      "grad_norm": 1.4342414140701294,
+      "learning_rate": 4.1683532270843504e-05,
+      "loss": 0.5703,
+      "step": 223
+    },
+    {
+      "epoch": 1.8261758691206544,
+      "grad_norm": 1.795954942703247,
+      "learning_rate": 4.121312358283463e-05,
+      "loss": 0.992,
+      "step": 224
+    },
+    {
+      "epoch": 1.834355828220859,
+      "grad_norm": 1.3253310918807983,
+      "learning_rate": 4.074351609155527e-05,
+      "loss": 0.5907,
+      "step": 225
+    },
+    {
+      "epoch": 1.8425357873210633,
+      "grad_norm": 1.4935518503189087,
+      "learning_rate": 4.027475261631548e-05,
+      "loss": 0.7448,
+      "step": 226
+    },
+    {
+      "epoch": 1.850715746421268,
+      "grad_norm": 1.8565064668655396,
+      "learning_rate": 3.980687589946715e-05,
+      "loss": 0.8506,
+      "step": 227
+    },
+    {
+      "epoch": 1.8588957055214723,
+      "grad_norm": 2.0860605239868164,
+      "learning_rate": 3.9339928602506505e-05,
+      "loss": 0.4935,
+      "step": 228
+    },
+    {
+      "epoch": 1.8670756646216768,
+      "grad_norm": 1.6324408054351807,
+      "learning_rate": 3.887395330218429e-05,
+      "loss": 0.586,
+      "step": 229
+    },
+    {
+      "epoch": 1.8752556237218814,
+      "grad_norm": 1.8466233015060425,
+      "learning_rate": 3.840899248662358e-05,
+      "loss": 0.7596,
+      "step": 230
+    },
+    {
+      "epoch": 1.883435582822086,
+      "grad_norm": 1.867876648902893,
+      "learning_rate": 3.7945088551445693e-05,
+      "loss": 0.7946,
+      "step": 231
+    },
+    {
+      "epoch": 1.8916155419222904,
+      "grad_norm": 1.3713316917419434,
+      "learning_rate": 3.748228379590438e-05,
+      "loss": 0.5414,
+      "step": 232
+    },
+    {
+      "epoch": 1.8997955010224947,
+      "grad_norm": 1.6689939498901367,
+      "learning_rate": 3.7020620419029094e-05,
+      "loss": 0.6574,
+      "step": 233
+    },
+    {
+      "epoch": 1.9079754601226995,
+      "grad_norm": 1.4076114892959595,
+      "learning_rate": 3.656014051577713e-05,
+      "loss": 0.5052,
+      "step": 234
+    },
+    {
+      "epoch": 1.9161554192229038,
+      "grad_norm": 1.6957688331604004,
+      "learning_rate": 3.610088607319544e-05,
+      "loss": 0.4209,
+      "step": 235
+    },
+    {
+      "epoch": 1.9243353783231085,
+      "grad_norm": 1.465134620666504,
+      "learning_rate": 3.564289896659214e-05,
+      "loss": 0.562,
+      "step": 236
+    },
+    {
+      "epoch": 1.9325153374233128,
+      "grad_norm": 1.626769781112671,
+      "learning_rate": 3.5186220955718306e-05,
+      "loss": 0.6494,
+      "step": 237
+    },
+    {
+      "epoch": 1.9406952965235174,
+      "grad_norm": 1.2987111806869507,
+      "learning_rate": 3.473089368096026e-05,
+      "loss": 0.5365,
+      "step": 238
+    },
+    {
+      "epoch": 1.9488752556237219,
+      "grad_norm": 1.7133764028549194,
+      "learning_rate": 3.427695865954284e-05,
+      "loss": 0.7972,
+      "step": 239
+    },
+    {
+      "epoch": 1.9570552147239264,
+      "grad_norm": 1.067958116531372,
+      "learning_rate": 3.3824457281743646e-05,
+      "loss": 0.2413,
+      "step": 240
+    },
+    {
+      "epoch": 1.965235173824131,
+      "grad_norm": 1.5035715103149414,
+      "learning_rate": 3.337343080711921e-05,
+      "loss": 0.655,
+      "step": 241
+    },
+    {
+      "epoch": 1.9734151329243352,
+      "grad_norm": 1.9790688753128052,
+      "learning_rate": 3.2923920360742774e-05,
+      "loss": 0.7517,
+      "step": 242
+    },
+    {
+      "epoch": 1.98159509202454,
+      "grad_norm": 1.79633367061615,
+      "learning_rate": 3.2475966929454504e-05,
+      "loss": 0.527,
+      "step": 243
+    },
+    {
+      "epoch": 1.9897750511247443,
+      "grad_norm": 1.59013032913208,
+      "learning_rate": 3.202961135812437e-05,
+      "loss": 0.5922,
+      "step": 244
+    },
+    {
+      "epoch": 1.997955010224949,
+      "grad_norm": 1.6466726064682007,
+      "learning_rate": 3.158489434592766e-05,
+      "loss": 0.6738,
+      "step": 245
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 0.8072463274002075,
+      "learning_rate": 3.114185644263415e-05,
+      "loss": 0.1228,
+      "step": 246
+    },
+    {
+      "epoch": 2.0081799591002043,
+      "grad_norm": 1.412455439567566,
+      "learning_rate": 3.070053804491068e-05,
+      "loss": 0.5372,
+      "step": 247
+    },
+    {
+      "epoch": 2.016359918200409,
+      "grad_norm": 1.15187406539917,
+      "learning_rate": 3.026097939263775e-05,
+      "loss": 0.3056,
+      "step": 248
+    },
+    {
+      "epoch": 2.0245398773006134,
+      "grad_norm": 1.3967753648757935,
+      "learning_rate": 2.9823220565240394e-05,
+      "loss": 0.5469,
+      "step": 249
+    },
+    {
+      "epoch": 2.032719836400818,
+      "grad_norm": 1.5053709745407104,
+      "learning_rate": 2.938730147803369e-05,
+      "loss": 0.5333,
+      "step": 250
+    },
+    {
+      "epoch": 2.0408997955010224,
+      "grad_norm": 1.2719367742538452,
+      "learning_rate": 2.895326187858326e-05,
+      "loss": 0.4873,
+      "step": 251
+    },
+    {
+      "epoch": 2.049079754601227,
+      "grad_norm": 1.3638321161270142,
+      "learning_rate": 2.852114134308104e-05,
+      "loss": 0.4676,
+      "step": 252
+    },
+    {
+      "epoch": 2.0572597137014315,
+      "grad_norm": 1.4079426527023315,
+      "learning_rate": 2.8090979272736662e-05,
+      "loss": 0.5474,
+      "step": 253
+    },
+    {
+      "epoch": 2.065439672801636,
+      "grad_norm": 1.3648539781570435,
+      "learning_rate": 2.7662814890184818e-05,
+      "loss": 0.3774,
+      "step": 254
+    },
+    {
+      "epoch": 2.0736196319018405,
+      "grad_norm": 1.411365032196045,
+      "learning_rate": 2.7236687235908953e-05,
+      "loss": 0.5021,
+      "step": 255
+    },
+    {
+      "epoch": 2.081799591002045,
+      "grad_norm": 0.8350080251693726,
+      "learning_rate": 2.6812635164681386e-05,
+      "loss": 0.295,
+      "step": 256
+    },
+    {
+      "epoch": 2.0899795501022496,
+      "grad_norm": 1.4837121963500977,
+      "learning_rate": 2.6390697342020665e-05,
+      "loss": 0.4359,
+      "step": 257
+    },
+    {
+      "epoch": 2.098159509202454,
+      "grad_norm": 1.447041392326355,
+      "learning_rate": 2.5970912240665813e-05,
+      "loss": 0.4699,
+      "step": 258
+    },
+    {
+      "epoch": 2.1063394683026586,
+      "grad_norm": 1.5087660551071167,
+      "learning_rate": 2.555331813706847e-05,
+      "loss": 0.5016,
+      "step": 259
+    },
+    {
+      "epoch": 2.114519427402863,
+      "grad_norm": 1.4970585107803345,
+      "learning_rate": 2.5137953107902813e-05,
+      "loss": 0.4827,
+      "step": 260
+    },
+    {
+      "epoch": 2.1226993865030677,
+      "grad_norm": 1.5823018550872803,
+      "learning_rate": 2.472485502659358e-05,
+      "loss": 0.3951,
+      "step": 261
+    },
+    {
+      "epoch": 2.130879345603272,
+      "grad_norm": 1.208630919456482,
+      "learning_rate": 2.4314061559862833e-05,
+      "loss": 0.3384,
+      "step": 262
+    },
+    {
+      "epoch": 2.1390593047034763,
+      "grad_norm": 1.6956486701965332,
+      "learning_rate": 2.3905610164295394e-05,
+      "loss": 0.4982,
+      "step": 263
+    },
+    {
+      "epoch": 2.147239263803681,
+      "grad_norm": 1.4397342205047607,
+      "learning_rate": 2.3499538082923606e-05,
+      "loss": 0.4574,
+      "step": 264
+    },
+    {
+      "epoch": 2.1554192229038853,
+      "grad_norm": 1.3102678060531616,
+      "learning_rate": 2.3095882341831372e-05,
+      "loss": 0.3559,
+      "step": 265
+    },
+    {
+      "epoch": 2.16359918200409,
+      "grad_norm": 1.2937331199645996,
+      "learning_rate": 2.2694679746778115e-05,
+      "loss": 0.3721,
+      "step": 266
+    },
+    {
+      "epoch": 2.1717791411042944,
+      "grad_norm": 1.5506526231765747,
+      "learning_rate": 2.22959668798428e-05,
+      "loss": 0.4909,
+      "step": 267
+    },
+    {
+      "epoch": 2.179959100204499,
+      "grad_norm": 1.8627556562423706,
+      "learning_rate": 2.1899780096088375e-05,
+      "loss": 0.7858,
+      "step": 268
+    },
+    {
+      "epoch": 2.1881390593047034,
+      "grad_norm": 1.7848111391067505,
+      "learning_rate": 2.1506155520246797e-05,
+      "loss": 0.6337,
+      "step": 269
+    },
+    {
+      "epoch": 2.196319018404908,
+      "grad_norm": 1.2659337520599365,
+      "learning_rate": 2.1115129043425187e-05,
+      "loss": 0.2693,
+      "step": 270
+    },
+    {
+      "epoch": 2.2044989775051125,
+      "grad_norm": 1.6412032842636108,
+      "learning_rate": 2.0726736319833228e-05,
+      "loss": 0.5306,
+      "step": 271
+    },
+    {
+      "epoch": 2.212678936605317,
+      "grad_norm": 1.611624002456665,
+      "learning_rate": 2.0341012763532243e-05,
+      "loss": 0.3388,
+      "step": 272
+    },
+    {
+      "epoch": 2.2208588957055215,
+      "grad_norm": 1.1925326585769653,
+      "learning_rate": 1.995799354520598e-05,
+      "loss": 0.3615,
+      "step": 273
+    },
+    {
+      "epoch": 2.229038854805726,
+      "grad_norm": 1.7512476444244385,
+      "learning_rate": 1.9577713588953795e-05,
+      "loss": 0.5129,
+      "step": 274
+    },
+    {
+      "epoch": 2.2372188139059306,
+      "grad_norm": 1.5006930828094482,
+      "learning_rate": 1.9200207569106216e-05,
+      "loss": 0.4129,
+      "step": 275
+    },
+    {
+      "epoch": 2.245398773006135,
+      "grad_norm": 1.8680585622787476,
+      "learning_rate": 1.8825509907063327e-05,
+      "loss": 0.5374,
+      "step": 276
+    },
+    {
+      "epoch": 2.2535787321063396,
+      "grad_norm": 1.8856024742126465,
+      "learning_rate": 1.8453654768156138e-05,
+      "loss": 0.562,
+      "step": 277
+    },
+    {
+      "epoch": 2.261758691206544,
+      "grad_norm": 1.9243358373641968,
+      "learning_rate": 1.8084676058531373e-05,
+      "loss": 0.6637,
+      "step": 278
+    },
+    {
+      "epoch": 2.2699386503067487,
+      "grad_norm": 2.3150854110717773,
+      "learning_rate": 1.771860742205988e-05,
+      "loss": 0.5932,
+      "step": 279
+    },
+    {
+      "epoch": 2.278118609406953,
+      "grad_norm": 1.2950345277786255,
+      "learning_rate": 1.7355482237268983e-05,
+      "loss": 0.341,
+      "step": 280
+    },
+    {
+      "epoch": 2.2862985685071573,
+      "grad_norm": 1.5685244798660278,
+      "learning_rate": 1.699533361429891e-05,
+      "loss": 0.4248,
+      "step": 281
+    },
+    {
+      "epoch": 2.294478527607362,
+      "grad_norm": 1.7234948873519897,
+      "learning_rate": 1.663819439188382e-05,
+      "loss": 0.7139,
+      "step": 282
+    },
+    {
+      "epoch": 2.3026584867075663,
+      "grad_norm": 1.5493229627609253,
+      "learning_rate": 1.6284097134357536e-05,
+      "loss": 0.4609,
+      "step": 283
+    },
+    {
+      "epoch": 2.310838445807771,
+      "grad_norm": 1.262978196144104,
+      "learning_rate": 1.5933074128684332e-05,
+      "loss": 0.3572,
+      "step": 284
+    },
+    {
+      "epoch": 2.3190184049079754,
+      "grad_norm": 1.7874940633773804,
+      "learning_rate": 1.5585157381514875e-05,
+      "loss": 0.5078,
+      "step": 285
+    },
+    {
+      "epoch": 2.32719836400818,
+      "grad_norm": 1.7057137489318848,
+      "learning_rate": 1.5240378616267886e-05,
+      "loss": 0.5262,
+      "step": 286
+    },
+    {
+      "epoch": 2.3353783231083844,
+      "grad_norm": 1.5174486637115479,
+      "learning_rate": 1.489876927023761e-05,
+      "loss": 0.4075,
+      "step": 287
+    },
+    {
+      "epoch": 2.3435582822085887,
+      "grad_norm": 1.473712682723999,
+      "learning_rate": 1.4560360491727231e-05,
+      "loss": 0.4237,
+      "step": 288
+    },
+    {
+      "epoch": 2.3517382413087935,
+      "grad_norm": 2.0275111198425293,
+      "learning_rate": 1.4225183137208776e-05,
+      "loss": 0.7344,
+      "step": 289
+    },
+    {
+      "epoch": 2.359918200408998,
+      "grad_norm": 1.5504990816116333,
+      "learning_rate": 1.389326776850966e-05,
+      "loss": 0.5226,
+      "step": 290
+    },
+    {
+      "epoch": 2.3680981595092025,
+      "grad_norm": 0.9763877987861633,
+      "learning_rate": 1.3564644650025893e-05,
+      "loss": 0.2004,
+      "step": 291
+    },
+    {
+      "epoch": 2.376278118609407,
+      "grad_norm": 1.6431723833084106,
+      "learning_rate": 1.3239343745962679e-05,
+      "loss": 0.5426,
+      "step": 292
+    },
+    {
+      "epoch": 2.3844580777096116,
+      "grad_norm": 1.4661204814910889,
+      "learning_rate": 1.2917394717602121e-05,
+      "loss": 0.3689,
+      "step": 293
+    },
+    {
+      "epoch": 2.392638036809816,
+      "grad_norm": 1.3995070457458496,
+      "learning_rate": 1.2598826920598772e-05,
+      "loss": 0.3994,
+      "step": 294
+    },
+    {
+      "epoch": 2.40081799591002,
+      "grad_norm": 1.6375926733016968,
+      "learning_rate": 1.2283669402302878e-05,
+      "loss": 0.4635,
+      "step": 295
+    },
+    {
+      "epoch": 2.408997955010225,
+      "grad_norm": 1.6579980850219727,
+      "learning_rate": 1.197195089911191e-05,
+      "loss": 0.44,
+      "step": 296
+    },
+    {
+      "epoch": 2.4171779141104293,
+      "grad_norm": 2.057859420776367,
+      "learning_rate": 1.1663699833850238e-05,
+      "loss": 0.809,
+      "step": 297
+    },
+    {
+      "epoch": 2.425357873210634,
+      "grad_norm": 1.9846243858337402,
+      "learning_rate": 1.1358944313177567e-05,
+      "loss": 0.526,
+      "step": 298
+    },
+    {
+      "epoch": 2.4335378323108383,
+      "grad_norm": 1.8454967737197876,
+      "learning_rate": 1.1057712125026116e-05,
+      "loss": 0.4943,
+      "step": 299
+    },
+    {
+      "epoch": 2.441717791411043,
+      "grad_norm": 1.3751471042633057,
+      "learning_rate": 1.0760030736066951e-05,
+      "loss": 0.2973,
+      "step": 300
+    },
+    {
+      "epoch": 2.4498977505112474,
+      "grad_norm": 1.7352081537246704,
+      "learning_rate": 1.0465927289205452e-05,
+      "loss": 0.4647,
+      "step": 301
+    },
+    {
+      "epoch": 2.458077709611452,
+      "grad_norm": 1.6583192348480225,
+      "learning_rate": 1.017542860110644e-05,
+      "loss": 0.5614,
+      "step": 302
+    },
+    {
+      "epoch": 2.4662576687116564,
+      "grad_norm": 1.1086567640304565,
+      "learning_rate": 9.888561159748993e-06,
+      "loss": 0.2343,
+      "step": 303
+    },
+    {
+      "epoch": 2.474437627811861,
+      "grad_norm": 1.2182183265686035,
+      "learning_rate": 9.605351122011309e-06,
+      "loss": 0.5084,
+      "step": 304
+    },
+    {
+      "epoch": 2.4826175869120655,
+      "grad_norm": 1.5897687673568726,
+      "learning_rate": 9.325824311285564e-06,
+      "loss": 0.4916,
+      "step": 305
+    },
+    {
+      "epoch": 2.4907975460122698,
+      "grad_norm": 1.7576637268066406,
+      "learning_rate": 9.050006215123419e-06,
+      "loss": 0.5896,
+      "step": 306
+    },
+    {
+      "epoch": 2.4989775051124745,
+      "grad_norm": 1.3375118970870972,
+      "learning_rate": 8.777921982911996e-06,
+      "loss": 0.3472,
+      "step": 307
+    },
+    {
+      "epoch": 2.507157464212679,
+      "grad_norm": 1.643762230873108,
+      "learning_rate": 8.509596423580712e-06,
+      "loss": 0.6561,
+      "step": 308
+    },
+    {
+      "epoch": 2.5153374233128836,
+      "grad_norm": 1.8207759857177734,
+      "learning_rate": 8.245054003339247e-06,
+      "loss": 0.446,
+      "step": 309
+    },
+    {
+      "epoch": 2.523517382413088,
+      "grad_norm": 1.7931218147277832,
+      "learning_rate": 7.984318843446593e-06,
+      "loss": 0.6626,
+      "step": 310
+    },
+    {
+      "epoch": 2.5316973415132926,
+      "grad_norm": 1.5871256589889526,
+      "learning_rate": 7.727414718011704e-06,
+      "loss": 0.6779,
+      "step": 311
+    },
+    {
+      "epoch": 2.539877300613497,
+      "grad_norm": 1.6045511960983276,
+      "learning_rate": 7.474365051825749e-06,
+      "loss": 0.4369,
+      "step": 312
+    },
+    {
+      "epoch": 2.5480572597137012,
+      "grad_norm": 1.9614039659500122,
+      "learning_rate": 7.225192918226214e-06,
+      "loss": 0.5339,
+      "step": 313
+    },
+    {
+      "epoch": 2.556237218813906,
+      "grad_norm": 1.6761356592178345,
+      "learning_rate": 6.979921036993042e-06,
+      "loss": 0.4714,
+      "step": 314
+    },
+    {
+      "epoch": 2.5644171779141103,
+      "grad_norm": 1.268598198890686,
+      "learning_rate": 6.738571772276997e-06,
+      "loss": 0.3589,
+      "step": 315
+    },
+    {
+      "epoch": 2.572597137014315,
+      "grad_norm": 1.9515974521636963,
+      "learning_rate": 6.501167130560515e-06,
+      "loss": 0.7677,
+      "step": 316
+    },
+    {
+      "epoch": 2.5807770961145193,
+      "grad_norm": 1.752503514289856,
+      "learning_rate": 6.267728758651132e-06,
+      "loss": 0.6019,
+      "step": 317
+    },
+    {
+      "epoch": 2.588957055214724,
+      "grad_norm": 1.6404023170471191,
+      "learning_rate": 6.03827794170767e-06,
+      "loss": 0.3813,
+      "step": 318
+    },
+    {
+      "epoch": 2.5971370143149284,
+      "grad_norm": 1.6431866884231567,
+      "learning_rate": 5.8128356012994375e-06,
+      "loss": 0.5397,
+      "step": 319
+    },
+    {
+      "epoch": 2.6053169734151327,
+      "grad_norm": 1.604200005531311,
+      "learning_rate": 5.591422293498633e-06,
+      "loss": 0.5326,
+      "step": 320
+    },
+    {
+      "epoch": 2.6134969325153374,
+      "grad_norm": 1.955712080001831,
+      "learning_rate": 5.374058207005944e-06,
+      "loss": 0.6279,
+      "step": 321
+    },
+    {
+      "epoch": 2.621676891615542,
+      "grad_norm": 1.9583613872528076,
+      "learning_rate": 5.160763161309767e-06,
+      "loss": 0.7064,
+      "step": 322
+    },
+    {
+      "epoch": 2.6298568507157465,
+      "grad_norm": 1.465756893157959,
+      "learning_rate": 4.951556604879048e-06,
+      "loss": 0.3176,
+      "step": 323
+    },
+    {
+      "epoch": 2.638036809815951,
+      "grad_norm": 1.0220084190368652,
+      "learning_rate": 4.746457613389904e-06,
+      "loss": 0.1989,
+      "step": 324
+    },
+    {
+      "epoch": 2.6462167689161555,
+      "grad_norm": 1.9900139570236206,
+      "learning_rate": 4.545484887986368e-06,
+      "loss": 0.4488,
+      "step": 325
+    },
+    {
+      "epoch": 2.65439672801636,
+      "grad_norm": 1.8389681577682495,
+      "learning_rate": 4.348656753575092e-06,
+      "loss": 0.8159,
+      "step": 326
+    },
+    {
+      "epoch": 2.662576687116564,
+      "grad_norm": 1.8046656847000122,
+      "learning_rate": 4.155991157154554e-06,
+      "loss": 0.5941,
+      "step": 327
+    },
+    {
+      "epoch": 2.670756646216769,
+      "grad_norm": 1.5946298837661743,
+      "learning_rate": 3.967505666178556e-06,
+      "loss": 0.6167,
+      "step": 328
+    },
+    {
+      "epoch": 2.6789366053169736,
+      "grad_norm": 1.6215424537658691,
+      "learning_rate": 3.783217466954503e-06,
+      "loss": 0.5432,
+      "step": 329
+    },
+    {
+      "epoch": 2.687116564417178,
+      "grad_norm": 1.5136370658874512,
+      "learning_rate": 3.603143363076217e-06,
+      "loss": 0.2688,
+      "step": 330
+    },
+    {
+      "epoch": 2.6952965235173822,
+      "grad_norm": 2.0225648880004883,
+      "learning_rate": 3.427299773891868e-06,
+      "loss": 0.3968,
+      "step": 331
+    },
+    {
+      "epoch": 2.703476482617587,
+      "grad_norm": 1.170069694519043,
+      "learning_rate": 3.2557027330067658e-06,
+      "loss": 0.3143,
+      "step": 332
+    },
+    {
+      "epoch": 2.7116564417177913,
+      "grad_norm": 1.2336766719818115,
+      "learning_rate": 3.0883678868214806e-06,
+      "loss": 0.4023,
+      "step": 333
+    },
+    {
+      "epoch": 2.719836400817996,
+      "grad_norm": 1.8785996437072754,
+      "learning_rate": 2.925310493105099e-06,
+      "loss": 0.6501,
+      "step": 334
+    },
+    {
+      "epoch": 2.7280163599182004,
+      "grad_norm": 1.7136589288711548,
+      "learning_rate": 2.7665454196040664e-06,
+      "loss": 0.3418,
+      "step": 335
+    },
+    {
+      "epoch": 2.736196319018405,
+      "grad_norm": 1.5453672409057617,
+      "learning_rate": 2.612087142686487e-06,
+      "loss": 0.4047,
+      "step": 336
+    },
+    {
+      "epoch": 2.7443762781186094,
+      "grad_norm": 1.5091831684112549,
+      "learning_rate": 2.4619497460222184e-06,
+      "loss": 0.3707,
+      "step": 337
+    },
+    {
+      "epoch": 2.7525562372188137,
+      "grad_norm": 1.996533751487732,
+      "learning_rate": 2.316146919298623e-06,
+      "loss": 0.7776,
+      "step": 338
+    },
+    {
+      "epoch": 2.7607361963190185,
+      "grad_norm": 2.2293291091918945,
+      "learning_rate": 2.1746919569723855e-06,
+      "loss": 0.7055,
+      "step": 339
+    },
+    {
+      "epoch": 2.7689161554192228,
+      "grad_norm": 1.906553864479065,
+      "learning_rate": 2.0375977570572967e-06,
+      "loss": 0.6423,
+      "step": 340
+    },
+    {
+      "epoch": 2.7770961145194275,
+      "grad_norm": 1.684910535812378,
+      "learning_rate": 1.9048768199481982e-06,
+      "loss": 0.5679,
+      "step": 341
+    },
+    {
+      "epoch": 2.785276073619632,
+      "grad_norm": 1.3118062019348145,
+      "learning_rate": 1.7765412472811771e-06,
+      "loss": 0.3036,
+      "step": 342
+    },
+    {
+      "epoch": 2.7934560327198366,
+      "grad_norm": 1.9178974628448486,
+      "learning_rate": 1.6526027408301226e-06,
+      "loss": 0.5829,
+      "step": 343
+    },
+    {
+      "epoch": 2.801635991820041,
+      "grad_norm": 1.860939860343933,
+      "learning_rate": 1.5330726014397668e-06,
+      "loss": 0.4617,
+      "step": 344
+    },
+    {
+      "epoch": 2.809815950920245,
+      "grad_norm": 1.7959818840026855,
+      "learning_rate": 1.417961727995254e-06,
+      "loss": 0.4604,
+      "step": 345
+    },
+    {
+      "epoch": 2.81799591002045,
+      "grad_norm": 1.414788842201233,
+      "learning_rate": 1.3072806164283358e-06,
+      "loss": 0.3398,
+      "step": 346
+    },
+    {
+      "epoch": 2.8261758691206547,
+      "grad_norm": 1.316179633140564,
+      "learning_rate": 1.2010393587603974e-06,
+      "loss": 0.3707,
+      "step": 347
+    },
+    {
+      "epoch": 2.834355828220859,
+      "grad_norm": 2.140214443206787,
+      "learning_rate": 1.099247642182205e-06,
+      "loss": 0.6991,
+      "step": 348
+    },
+    {
+      "epoch": 2.8425357873210633,
+      "grad_norm": 1.6871670484542847,
+      "learning_rate": 1.0019147481706625e-06,
+      "loss": 0.6069,
+      "step": 349
+    },
+    {
+      "epoch": 2.850715746421268,
+      "grad_norm": 1.7937754392623901,
+      "learning_rate": 9.090495516424713e-07,
+      "loss": 0.3841,
+      "step": 350
+    },
+    {
+      "epoch": 2.8588957055214723,
+      "grad_norm": 1.3567599058151245,
+      "learning_rate": 8.206605201449447e-07,
+      "loss": 0.3186,
+      "step": 351
+    },
+    {
+      "epoch": 2.8670756646216766,
+      "grad_norm": 1.0344539880752563,
+      "learning_rate": 7.36755713083892e-07,
+      "loss": 0.1676,
+      "step": 352
+    },
+    {
+      "epoch": 2.8752556237218814,
+      "grad_norm": 1.908477783203125,
+      "learning_rate": 6.573427809888067e-07,
+      "loss": 0.7295,
+      "step": 353
+    },
+    {
+      "epoch": 2.883435582822086,
+      "grad_norm": 2.14554500579834,
+      "learning_rate": 5.824289648152126e-07,
+      "loss": 0.8187,
+      "step": 354
+    },
+    {
+      "epoch": 2.8916155419222904,
+      "grad_norm": 1.6814268827438354,
+      "learning_rate": 5.120210952844872e-07,
+      "loss": 0.5205,
+      "step": 355
+    },
+    {
+      "epoch": 2.8997955010224947,
+      "grad_norm": 1.6498082876205444,
+      "learning_rate": 4.461255922609986e-07,
+      "loss": 0.4557,
+      "step": 356
+    },
+    {
+      "epoch": 2.9079754601226995,
+      "grad_norm": 1.4337708950042725,
+      "learning_rate": 3.8474846416672874e-07,
+      "loss": 0.3251,
+      "step": 357
+    },
+    {
+      "epoch": 2.9161554192229038,
+      "grad_norm": 1.875313401222229,
+      "learning_rate": 3.278953074334512e-07,
+      "loss": 0.5001,
+      "step": 358
+    },
+    {
+      "epoch": 2.9243353783231085,
+      "grad_norm": 1.350846529006958,
+      "learning_rate": 2.75571305992417e-07,
+      "loss": 0.2414,
+      "step": 359
+    },
+    {
+      "epoch": 2.932515337423313,
+      "grad_norm": 1.5978336334228516,
+      "learning_rate": 2.2778123080167135e-07,
+      "loss": 0.4585,
+      "step": 360
+    },
+    {
+      "epoch": 2.9406952965235176,
+      "grad_norm": 1.672541856765747,
+      "learning_rate": 1.8452943941106859e-07,
+      "loss": 0.5382,
+      "step": 361
+    },
+    {
+      "epoch": 2.948875255623722,
+      "grad_norm": 1.3987324237823486,
+      "learning_rate": 1.4581987556490095e-07,
+      "loss": 0.3326,
+      "step": 362
+    },
+    {
+      "epoch": 2.957055214723926,
+      "grad_norm": 1.4565430879592896,
+      "learning_rate": 1.1165606884234181e-07,
+      "loss": 0.4901,
+      "step": 363
+    },
+    {
+      "epoch": 2.965235173824131,
+      "grad_norm": 1.3861486911773682,
+      "learning_rate": 8.204113433559201e-08,
+      "loss": 0.2756,
+      "step": 364
+    },
+    {
+      "epoch": 2.9734151329243352,
+      "grad_norm": 1.4839295148849487,
+      "learning_rate": 5.697777236585711e-08,
+      "loss": 0.3303,
+      "step": 365
+    },
+    {
+      "epoch": 2.98159509202454,
+      "grad_norm": 1.6138904094696045,
+      "learning_rate": 3.6468268237105366e-08,
+      "loss": 0.558,
+      "step": 366
+    },
+    {
+      "epoch": 2.98159509202454,
+      "step": 366,
+      "total_flos": 3.142935032247091e+16,
+      "train_loss": 0.7803121163341843,
+      "train_runtime": 701.3998,
+      "train_samples_per_second": 4.183,
+      "train_steps_per_second": 0.522
+    }
+  ],
+  "logging_steps": 1,
+  "max_steps": 366,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 3,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 3.142935032247091e+16,
+  "train_batch_size": 2,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2d8fa4ccfbbafc0e16806b535f96bbe57bf08f0c48d862e66bc54bb36f8a41a4
+size 5688