Upload llava-v1.6-34b-chatml_direct-anyres model
Browse files- README.md +202 -0
- adapter_config.json +34 -0
- adapter_model.safetensors +3 -0
- config.json +77 -0
- non_lora_trainables.bin +3 -0
- optimizer.pt +3 -0
- trainer_state.json +1842 -0
README.md
ADDED
|
@@ -0,0 +1,202 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
base_model: liuhaotian/llava-v1.6-34b
|
| 3 |
+
library_name: peft
|
| 4 |
+
---
|
| 5 |
+
|
| 6 |
+
# Model Card for Model ID
|
| 7 |
+
|
| 8 |
+
<!-- Provide a quick summary of what the model is/does. -->
|
| 9 |
+
|
| 10 |
+
|
| 11 |
+
|
| 12 |
+
## Model Details
|
| 13 |
+
|
| 14 |
+
### Model Description
|
| 15 |
+
|
| 16 |
+
<!-- Provide a longer summary of what this model is. -->
|
| 17 |
+
|
| 18 |
+
|
| 19 |
+
|
| 20 |
+
- **Developed by:** [More Information Needed]
|
| 21 |
+
- **Funded by [optional]:** [More Information Needed]
|
| 22 |
+
- **Shared by [optional]:** [More Information Needed]
|
| 23 |
+
- **Model type:** [More Information Needed]
|
| 24 |
+
- **Language(s) (NLP):** [More Information Needed]
|
| 25 |
+
- **License:** [More Information Needed]
|
| 26 |
+
- **Finetuned from model [optional]:** [More Information Needed]
|
| 27 |
+
|
| 28 |
+
### Model Sources [optional]
|
| 29 |
+
|
| 30 |
+
<!-- Provide the basic links for the model. -->
|
| 31 |
+
|
| 32 |
+
- **Repository:** [More Information Needed]
|
| 33 |
+
- **Paper [optional]:** [More Information Needed]
|
| 34 |
+
- **Demo [optional]:** [More Information Needed]
|
| 35 |
+
|
| 36 |
+
## Uses
|
| 37 |
+
|
| 38 |
+
<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
|
| 39 |
+
|
| 40 |
+
### Direct Use
|
| 41 |
+
|
| 42 |
+
<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
|
| 43 |
+
|
| 44 |
+
[More Information Needed]
|
| 45 |
+
|
| 46 |
+
### Downstream Use [optional]
|
| 47 |
+
|
| 48 |
+
<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
|
| 49 |
+
|
| 50 |
+
[More Information Needed]
|
| 51 |
+
|
| 52 |
+
### Out-of-Scope Use
|
| 53 |
+
|
| 54 |
+
<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
|
| 55 |
+
|
| 56 |
+
[More Information Needed]
|
| 57 |
+
|
| 58 |
+
## Bias, Risks, and Limitations
|
| 59 |
+
|
| 60 |
+
<!-- This section is meant to convey both technical and sociotechnical limitations. -->
|
| 61 |
+
|
| 62 |
+
[More Information Needed]
|
| 63 |
+
|
| 64 |
+
### Recommendations
|
| 65 |
+
|
| 66 |
+
<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
|
| 67 |
+
|
| 68 |
+
Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
|
| 69 |
+
|
| 70 |
+
## How to Get Started with the Model
|
| 71 |
+
|
| 72 |
+
Use the code below to get started with the model.
|
| 73 |
+
|
| 74 |
+
[More Information Needed]
|
| 75 |
+
|
| 76 |
+
## Training Details
|
| 77 |
+
|
| 78 |
+
### Training Data
|
| 79 |
+
|
| 80 |
+
<!-- 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. -->
|
| 81 |
+
|
| 82 |
+
[More Information Needed]
|
| 83 |
+
|
| 84 |
+
### Training Procedure
|
| 85 |
+
|
| 86 |
+
<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
|
| 87 |
+
|
| 88 |
+
#### Preprocessing [optional]
|
| 89 |
+
|
| 90 |
+
[More Information Needed]
|
| 91 |
+
|
| 92 |
+
|
| 93 |
+
#### Training Hyperparameters
|
| 94 |
+
|
| 95 |
+
- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
|
| 96 |
+
|
| 97 |
+
#### Speeds, Sizes, Times [optional]
|
| 98 |
+
|
| 99 |
+
<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
|
| 100 |
+
|
| 101 |
+
[More Information Needed]
|
| 102 |
+
|
| 103 |
+
## Evaluation
|
| 104 |
+
|
| 105 |
+
<!-- This section describes the evaluation protocols and provides the results. -->
|
| 106 |
+
|
| 107 |
+
### Testing Data, Factors & Metrics
|
| 108 |
+
|
| 109 |
+
#### Testing Data
|
| 110 |
+
|
| 111 |
+
<!-- This should link to a Dataset Card if possible. -->
|
| 112 |
+
|
| 113 |
+
[More Information Needed]
|
| 114 |
+
|
| 115 |
+
#### Factors
|
| 116 |
+
|
| 117 |
+
<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
|
| 118 |
+
|
| 119 |
+
[More Information Needed]
|
| 120 |
+
|
| 121 |
+
#### Metrics
|
| 122 |
+
|
| 123 |
+
<!-- These are the evaluation metrics being used, ideally with a description of why. -->
|
| 124 |
+
|
| 125 |
+
[More Information Needed]
|
| 126 |
+
|
| 127 |
+
### Results
|
| 128 |
+
|
| 129 |
+
[More Information Needed]
|
| 130 |
+
|
| 131 |
+
#### Summary
|
| 132 |
+
|
| 133 |
+
|
| 134 |
+
|
| 135 |
+
## Model Examination [optional]
|
| 136 |
+
|
| 137 |
+
<!-- Relevant interpretability work for the model goes here -->
|
| 138 |
+
|
| 139 |
+
[More Information Needed]
|
| 140 |
+
|
| 141 |
+
## Environmental Impact
|
| 142 |
+
|
| 143 |
+
<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
|
| 144 |
+
|
| 145 |
+
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).
|
| 146 |
+
|
| 147 |
+
- **Hardware Type:** [More Information Needed]
|
| 148 |
+
- **Hours used:** [More Information Needed]
|
| 149 |
+
- **Cloud Provider:** [More Information Needed]
|
| 150 |
+
- **Compute Region:** [More Information Needed]
|
| 151 |
+
- **Carbon Emitted:** [More Information Needed]
|
| 152 |
+
|
| 153 |
+
## Technical Specifications [optional]
|
| 154 |
+
|
| 155 |
+
### Model Architecture and Objective
|
| 156 |
+
|
| 157 |
+
[More Information Needed]
|
| 158 |
+
|
| 159 |
+
### Compute Infrastructure
|
| 160 |
+
|
| 161 |
+
[More Information Needed]
|
| 162 |
+
|
| 163 |
+
#### Hardware
|
| 164 |
+
|
| 165 |
+
[More Information Needed]
|
| 166 |
+
|
| 167 |
+
#### Software
|
| 168 |
+
|
| 169 |
+
[More Information Needed]
|
| 170 |
+
|
| 171 |
+
## Citation [optional]
|
| 172 |
+
|
| 173 |
+
<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
|
| 174 |
+
|
| 175 |
+
**BibTeX:**
|
| 176 |
+
|
| 177 |
+
[More Information Needed]
|
| 178 |
+
|
| 179 |
+
**APA:**
|
| 180 |
+
|
| 181 |
+
[More Information Needed]
|
| 182 |
+
|
| 183 |
+
## Glossary [optional]
|
| 184 |
+
|
| 185 |
+
<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
|
| 186 |
+
|
| 187 |
+
[More Information Needed]
|
| 188 |
+
|
| 189 |
+
## More Information [optional]
|
| 190 |
+
|
| 191 |
+
[More Information Needed]
|
| 192 |
+
|
| 193 |
+
## Model Card Authors [optional]
|
| 194 |
+
|
| 195 |
+
[More Information Needed]
|
| 196 |
+
|
| 197 |
+
## Model Card Contact
|
| 198 |
+
|
| 199 |
+
[More Information Needed]
|
| 200 |
+
### Framework versions
|
| 201 |
+
|
| 202 |
+
- PEFT 0.13.2
|
adapter_config.json
ADDED
|
@@ -0,0 +1,34 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"alpha_pattern": {},
|
| 3 |
+
"auto_mapping": null,
|
| 4 |
+
"base_model_name_or_path": "liuhaotian/llava-v1.6-34b",
|
| 5 |
+
"bias": "none",
|
| 6 |
+
"fan_in_fan_out": false,
|
| 7 |
+
"inference_mode": true,
|
| 8 |
+
"init_lora_weights": true,
|
| 9 |
+
"layer_replication": null,
|
| 10 |
+
"layers_pattern": null,
|
| 11 |
+
"layers_to_transform": null,
|
| 12 |
+
"loftq_config": {},
|
| 13 |
+
"lora_alpha": 16,
|
| 14 |
+
"lora_dropout": 0.05,
|
| 15 |
+
"megatron_config": null,
|
| 16 |
+
"megatron_core": "megatron.core",
|
| 17 |
+
"modules_to_save": null,
|
| 18 |
+
"peft_type": "LORA",
|
| 19 |
+
"r": 8,
|
| 20 |
+
"rank_pattern": {},
|
| 21 |
+
"revision": null,
|
| 22 |
+
"target_modules": [
|
| 23 |
+
"v_proj",
|
| 24 |
+
"down_proj",
|
| 25 |
+
"o_proj",
|
| 26 |
+
"up_proj",
|
| 27 |
+
"q_proj",
|
| 28 |
+
"k_proj",
|
| 29 |
+
"gate_proj"
|
| 30 |
+
],
|
| 31 |
+
"task_type": "CAUSAL_LM",
|
| 32 |
+
"use_dora": false,
|
| 33 |
+
"use_rslora": false
|
| 34 |
+
}
|
adapter_model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5803648ac0903277e08008b9b5a060dbf8b81b16509b7e57e42777915edd12d6
|
| 3 |
+
size 125377072
|
config.json
ADDED
|
@@ -0,0 +1,77 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_attn_implementation_autoset": true,
|
| 3 |
+
"_name_or_path": "liuhaotian/llava-v1.6-34b",
|
| 4 |
+
"architectures": [
|
| 5 |
+
"LlavaLlamaForCausalLM"
|
| 6 |
+
],
|
| 7 |
+
"attention_bias": false,
|
| 8 |
+
"attention_dropout": 0.0,
|
| 9 |
+
"bos_token_id": 1,
|
| 10 |
+
"eos_token_id": 7,
|
| 11 |
+
"freeze_mm_mlp_adapter": false,
|
| 12 |
+
"freeze_mm_vision_resampler": false,
|
| 13 |
+
"head_dim": 128,
|
| 14 |
+
"hidden_act": "silu",
|
| 15 |
+
"hidden_size": 7168,
|
| 16 |
+
"image_aspect_ratio": "anyres",
|
| 17 |
+
"image_crop_resolution": 224,
|
| 18 |
+
"image_grid_pinpoints": [
|
| 19 |
+
[
|
| 20 |
+
336,
|
| 21 |
+
672
|
| 22 |
+
],
|
| 23 |
+
[
|
| 24 |
+
672,
|
| 25 |
+
336
|
| 26 |
+
],
|
| 27 |
+
[
|
| 28 |
+
672,
|
| 29 |
+
672
|
| 30 |
+
],
|
| 31 |
+
[
|
| 32 |
+
1008,
|
| 33 |
+
336
|
| 34 |
+
],
|
| 35 |
+
[
|
| 36 |
+
336,
|
| 37 |
+
1008
|
| 38 |
+
]
|
| 39 |
+
],
|
| 40 |
+
"image_split_resolution": 224,
|
| 41 |
+
"image_token_index": 64002,
|
| 42 |
+
"initializer_range": 0.02,
|
| 43 |
+
"intermediate_size": 20480,
|
| 44 |
+
"max_position_embeddings": 4096,
|
| 45 |
+
"mlp_bias": false,
|
| 46 |
+
"mm_hidden_size": 1024,
|
| 47 |
+
"mm_patch_merge_type": "spatial_unpad",
|
| 48 |
+
"mm_projector_lr": 2e-05,
|
| 49 |
+
"mm_projector_type": "mlp2x_gelu",
|
| 50 |
+
"mm_resampler_type": null,
|
| 51 |
+
"mm_use_im_patch_token": false,
|
| 52 |
+
"mm_use_im_start_end": false,
|
| 53 |
+
"mm_vision_select_feature": "patch",
|
| 54 |
+
"mm_vision_select_layer": -2,
|
| 55 |
+
"mm_vision_tower": "openai/clip-vit-large-patch14-336",
|
| 56 |
+
"mm_vision_tower_lr": 2e-06,
|
| 57 |
+
"model_type": "llava_llama",
|
| 58 |
+
"num_attention_heads": 56,
|
| 59 |
+
"num_hidden_layers": 60,
|
| 60 |
+
"num_key_value_heads": 8,
|
| 61 |
+
"pad_token_id": 0,
|
| 62 |
+
"pretraining_tp": 1,
|
| 63 |
+
"rms_norm_eps": 1e-05,
|
| 64 |
+
"rope_scaling": null,
|
| 65 |
+
"rope_theta": 5000000.0,
|
| 66 |
+
"tie_word_embeddings": false,
|
| 67 |
+
"tokenizer_model_max_length": 2048,
|
| 68 |
+
"tokenizer_padding_side": "right",
|
| 69 |
+
"torch_dtype": "bfloat16",
|
| 70 |
+
"transformers_version": "4.46.0",
|
| 71 |
+
"tune_mm_mlp_adapter": false,
|
| 72 |
+
"tune_mm_vision_resampler": false,
|
| 73 |
+
"unfreeze_mm_vision_tower": true,
|
| 74 |
+
"use_cache": true,
|
| 75 |
+
"use_mm_proj": true,
|
| 76 |
+
"vocab_size": 64000
|
| 77 |
+
}
|
non_lora_trainables.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f1406c16f0a45d0a292e968ad8272373f978ff3eb3ce1ab7ce9cbe4f730dd767
|
| 3 |
+
size 117471408
|
optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3fbf568e25cb77a88312274afb8ba233d015b9e3e5400bc2f1ad3396a3266a8a
|
| 3 |
+
size 364065314
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1842 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.7502281069755554,
|
| 3 |
+
"best_model_checkpoint": "./checkpoints/llava-v1.6-34b-chatml_direct-anyres/checkpoint-115",
|
| 4 |
+
"epoch": 3.75,
|
| 5 |
+
"eval_steps": 1.0,
|
| 6 |
+
"global_step": 120,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.03125,
|
| 13 |
+
"grad_norm": 0.3345325833952841,
|
| 14 |
+
"learning_rate": 0.0,
|
| 15 |
+
"loss": 1.2677,
|
| 16 |
+
"step": 1
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.03125,
|
| 20 |
+
"eval_loss": 1.3042412996292114,
|
| 21 |
+
"eval_runtime": 195.8656,
|
| 22 |
+
"eval_samples_per_second": 1.021,
|
| 23 |
+
"eval_steps_per_second": 0.128,
|
| 24 |
+
"step": 1
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.0625,
|
| 28 |
+
"grad_norm": 0.35890614982474045,
|
| 29 |
+
"learning_rate": 8.613531161467863e-06,
|
| 30 |
+
"loss": 1.3505,
|
| 31 |
+
"step": 2
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.0625,
|
| 35 |
+
"eval_loss": 1.3042412996292114,
|
| 36 |
+
"eval_runtime": 190.4145,
|
| 37 |
+
"eval_samples_per_second": 1.05,
|
| 38 |
+
"eval_steps_per_second": 0.131,
|
| 39 |
+
"step": 2
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 0.09375,
|
| 43 |
+
"grad_norm": 0.3316701046946783,
|
| 44 |
+
"learning_rate": 1.3652123889719709e-05,
|
| 45 |
+
"loss": 1.2569,
|
| 46 |
+
"step": 3
|
| 47 |
+
},
|
| 48 |
+
{
|
| 49 |
+
"epoch": 0.09375,
|
| 50 |
+
"eval_loss": 1.2942205667495728,
|
| 51 |
+
"eval_runtime": 190.6475,
|
| 52 |
+
"eval_samples_per_second": 1.049,
|
| 53 |
+
"eval_steps_per_second": 0.131,
|
| 54 |
+
"step": 3
|
| 55 |
+
},
|
| 56 |
+
{
|
| 57 |
+
"epoch": 0.125,
|
| 58 |
+
"grad_norm": 0.32188096642751235,
|
| 59 |
+
"learning_rate": 1.7227062322935725e-05,
|
| 60 |
+
"loss": 1.2323,
|
| 61 |
+
"step": 4
|
| 62 |
+
},
|
| 63 |
+
{
|
| 64 |
+
"epoch": 0.125,
|
| 65 |
+
"eval_loss": 1.2789555788040161,
|
| 66 |
+
"eval_runtime": 189.7666,
|
| 67 |
+
"eval_samples_per_second": 1.054,
|
| 68 |
+
"eval_steps_per_second": 0.132,
|
| 69 |
+
"step": 4
|
| 70 |
+
},
|
| 71 |
+
{
|
| 72 |
+
"epoch": 0.15625,
|
| 73 |
+
"grad_norm": 0.3767527705004001,
|
| 74 |
+
"learning_rate": 2e-05,
|
| 75 |
+
"loss": 1.2785,
|
| 76 |
+
"step": 5
|
| 77 |
+
},
|
| 78 |
+
{
|
| 79 |
+
"epoch": 0.15625,
|
| 80 |
+
"eval_loss": 1.258152723312378,
|
| 81 |
+
"eval_runtime": 189.5935,
|
| 82 |
+
"eval_samples_per_second": 1.055,
|
| 83 |
+
"eval_steps_per_second": 0.132,
|
| 84 |
+
"step": 5
|
| 85 |
+
},
|
| 86 |
+
{
|
| 87 |
+
"epoch": 0.1875,
|
| 88 |
+
"grad_norm": 0.3287126070774628,
|
| 89 |
+
"learning_rate": 2e-05,
|
| 90 |
+
"loss": 1.2151,
|
| 91 |
+
"step": 6
|
| 92 |
+
},
|
| 93 |
+
{
|
| 94 |
+
"epoch": 0.1875,
|
| 95 |
+
"eval_loss": 1.2347216606140137,
|
| 96 |
+
"eval_runtime": 190.4111,
|
| 97 |
+
"eval_samples_per_second": 1.05,
|
| 98 |
+
"eval_steps_per_second": 0.131,
|
| 99 |
+
"step": 6
|
| 100 |
+
},
|
| 101 |
+
{
|
| 102 |
+
"epoch": 0.21875,
|
| 103 |
+
"grad_norm": 0.34451127286331007,
|
| 104 |
+
"learning_rate": 2e-05,
|
| 105 |
+
"loss": 1.2968,
|
| 106 |
+
"step": 7
|
| 107 |
+
},
|
| 108 |
+
{
|
| 109 |
+
"epoch": 0.21875,
|
| 110 |
+
"eval_loss": 1.210167646408081,
|
| 111 |
+
"eval_runtime": 190.9799,
|
| 112 |
+
"eval_samples_per_second": 1.047,
|
| 113 |
+
"eval_steps_per_second": 0.131,
|
| 114 |
+
"step": 7
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.25,
|
| 118 |
+
"grad_norm": 0.36105870692958336,
|
| 119 |
+
"learning_rate": 2e-05,
|
| 120 |
+
"loss": 1.2277,
|
| 121 |
+
"step": 8
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.25,
|
| 125 |
+
"eval_loss": 1.1862907409667969,
|
| 126 |
+
"eval_runtime": 190.685,
|
| 127 |
+
"eval_samples_per_second": 1.049,
|
| 128 |
+
"eval_steps_per_second": 0.131,
|
| 129 |
+
"step": 8
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.28125,
|
| 133 |
+
"grad_norm": 0.35460549637546845,
|
| 134 |
+
"learning_rate": 2e-05,
|
| 135 |
+
"loss": 1.2101,
|
| 136 |
+
"step": 9
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.28125,
|
| 140 |
+
"eval_loss": 1.1649302244186401,
|
| 141 |
+
"eval_runtime": 190.1569,
|
| 142 |
+
"eval_samples_per_second": 1.052,
|
| 143 |
+
"eval_steps_per_second": 0.131,
|
| 144 |
+
"step": 9
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 0.3125,
|
| 148 |
+
"grad_norm": 0.3134923556618721,
|
| 149 |
+
"learning_rate": 2e-05,
|
| 150 |
+
"loss": 1.1163,
|
| 151 |
+
"step": 10
|
| 152 |
+
},
|
| 153 |
+
{
|
| 154 |
+
"epoch": 0.3125,
|
| 155 |
+
"eval_loss": 1.144965410232544,
|
| 156 |
+
"eval_runtime": 190.0982,
|
| 157 |
+
"eval_samples_per_second": 1.052,
|
| 158 |
+
"eval_steps_per_second": 0.132,
|
| 159 |
+
"step": 10
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 0.34375,
|
| 163 |
+
"grad_norm": 0.3069481492118633,
|
| 164 |
+
"learning_rate": 2e-05,
|
| 165 |
+
"loss": 1.1483,
|
| 166 |
+
"step": 11
|
| 167 |
+
},
|
| 168 |
+
{
|
| 169 |
+
"epoch": 0.34375,
|
| 170 |
+
"eval_loss": 1.124668002128601,
|
| 171 |
+
"eval_runtime": 192.0572,
|
| 172 |
+
"eval_samples_per_second": 1.041,
|
| 173 |
+
"eval_steps_per_second": 0.13,
|
| 174 |
+
"step": 11
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"epoch": 0.375,
|
| 178 |
+
"grad_norm": 0.2801324709168811,
|
| 179 |
+
"learning_rate": 2e-05,
|
| 180 |
+
"loss": 1.1172,
|
| 181 |
+
"step": 12
|
| 182 |
+
},
|
| 183 |
+
{
|
| 184 |
+
"epoch": 0.375,
|
| 185 |
+
"eval_loss": 1.1061824560165405,
|
| 186 |
+
"eval_runtime": 192.6406,
|
| 187 |
+
"eval_samples_per_second": 1.038,
|
| 188 |
+
"eval_steps_per_second": 0.13,
|
| 189 |
+
"step": 12
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 0.40625,
|
| 193 |
+
"grad_norm": 0.33156251919932406,
|
| 194 |
+
"learning_rate": 2e-05,
|
| 195 |
+
"loss": 1.1902,
|
| 196 |
+
"step": 13
|
| 197 |
+
},
|
| 198 |
+
{
|
| 199 |
+
"epoch": 0.40625,
|
| 200 |
+
"eval_loss": 1.0897018909454346,
|
| 201 |
+
"eval_runtime": 192.8064,
|
| 202 |
+
"eval_samples_per_second": 1.037,
|
| 203 |
+
"eval_steps_per_second": 0.13,
|
| 204 |
+
"step": 13
|
| 205 |
+
},
|
| 206 |
+
{
|
| 207 |
+
"epoch": 0.4375,
|
| 208 |
+
"grad_norm": 0.3307149375898363,
|
| 209 |
+
"learning_rate": 2e-05,
|
| 210 |
+
"loss": 1.1014,
|
| 211 |
+
"step": 14
|
| 212 |
+
},
|
| 213 |
+
{
|
| 214 |
+
"epoch": 0.4375,
|
| 215 |
+
"eval_loss": 1.075058937072754,
|
| 216 |
+
"eval_runtime": 192.1353,
|
| 217 |
+
"eval_samples_per_second": 1.041,
|
| 218 |
+
"eval_steps_per_second": 0.13,
|
| 219 |
+
"step": 14
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.46875,
|
| 223 |
+
"grad_norm": 0.31999611930431227,
|
| 224 |
+
"learning_rate": 2e-05,
|
| 225 |
+
"loss": 1.0847,
|
| 226 |
+
"step": 15
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.46875,
|
| 230 |
+
"eval_loss": 1.0613062381744385,
|
| 231 |
+
"eval_runtime": 192.0586,
|
| 232 |
+
"eval_samples_per_second": 1.041,
|
| 233 |
+
"eval_steps_per_second": 0.13,
|
| 234 |
+
"step": 15
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.5,
|
| 238 |
+
"grad_norm": 0.2494159223446848,
|
| 239 |
+
"learning_rate": 2e-05,
|
| 240 |
+
"loss": 1.0428,
|
| 241 |
+
"step": 16
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.5,
|
| 245 |
+
"eval_loss": 1.0484405755996704,
|
| 246 |
+
"eval_runtime": 192.52,
|
| 247 |
+
"eval_samples_per_second": 1.039,
|
| 248 |
+
"eval_steps_per_second": 0.13,
|
| 249 |
+
"step": 16
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 0.53125,
|
| 253 |
+
"grad_norm": 0.2899303168196212,
|
| 254 |
+
"learning_rate": 2e-05,
|
| 255 |
+
"loss": 1.122,
|
| 256 |
+
"step": 17
|
| 257 |
+
},
|
| 258 |
+
{
|
| 259 |
+
"epoch": 0.53125,
|
| 260 |
+
"eval_loss": 1.036120891571045,
|
| 261 |
+
"eval_runtime": 192.5716,
|
| 262 |
+
"eval_samples_per_second": 1.039,
|
| 263 |
+
"eval_steps_per_second": 0.13,
|
| 264 |
+
"step": 17
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 0.5625,
|
| 268 |
+
"grad_norm": 0.2995776829874209,
|
| 269 |
+
"learning_rate": 2e-05,
|
| 270 |
+
"loss": 1.0425,
|
| 271 |
+
"step": 18
|
| 272 |
+
},
|
| 273 |
+
{
|
| 274 |
+
"epoch": 0.5625,
|
| 275 |
+
"eval_loss": 1.0226774215698242,
|
| 276 |
+
"eval_runtime": 192.5256,
|
| 277 |
+
"eval_samples_per_second": 1.039,
|
| 278 |
+
"eval_steps_per_second": 0.13,
|
| 279 |
+
"step": 18
|
| 280 |
+
},
|
| 281 |
+
{
|
| 282 |
+
"epoch": 0.59375,
|
| 283 |
+
"grad_norm": 0.28709859243892955,
|
| 284 |
+
"learning_rate": 2e-05,
|
| 285 |
+
"loss": 1.0098,
|
| 286 |
+
"step": 19
|
| 287 |
+
},
|
| 288 |
+
{
|
| 289 |
+
"epoch": 0.59375,
|
| 290 |
+
"eval_loss": 1.0081837177276611,
|
| 291 |
+
"eval_runtime": 192.3486,
|
| 292 |
+
"eval_samples_per_second": 1.04,
|
| 293 |
+
"eval_steps_per_second": 0.13,
|
| 294 |
+
"step": 19
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"epoch": 0.625,
|
| 298 |
+
"grad_norm": 0.27612474678791227,
|
| 299 |
+
"learning_rate": 2e-05,
|
| 300 |
+
"loss": 1.0563,
|
| 301 |
+
"step": 20
|
| 302 |
+
},
|
| 303 |
+
{
|
| 304 |
+
"epoch": 0.625,
|
| 305 |
+
"eval_loss": 0.994163990020752,
|
| 306 |
+
"eval_runtime": 191.9782,
|
| 307 |
+
"eval_samples_per_second": 1.042,
|
| 308 |
+
"eval_steps_per_second": 0.13,
|
| 309 |
+
"step": 20
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 0.65625,
|
| 313 |
+
"grad_norm": 0.24260720679126926,
|
| 314 |
+
"learning_rate": 2e-05,
|
| 315 |
+
"loss": 1.0355,
|
| 316 |
+
"step": 21
|
| 317 |
+
},
|
| 318 |
+
{
|
| 319 |
+
"epoch": 0.65625,
|
| 320 |
+
"eval_loss": 0.9819543361663818,
|
| 321 |
+
"eval_runtime": 191.9306,
|
| 322 |
+
"eval_samples_per_second": 1.042,
|
| 323 |
+
"eval_steps_per_second": 0.13,
|
| 324 |
+
"step": 21
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.6875,
|
| 328 |
+
"grad_norm": 0.25336536603884946,
|
| 329 |
+
"learning_rate": 2e-05,
|
| 330 |
+
"loss": 1.0525,
|
| 331 |
+
"step": 22
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.6875,
|
| 335 |
+
"eval_loss": 0.9709838032722473,
|
| 336 |
+
"eval_runtime": 192.9913,
|
| 337 |
+
"eval_samples_per_second": 1.036,
|
| 338 |
+
"eval_steps_per_second": 0.13,
|
| 339 |
+
"step": 22
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.71875,
|
| 343 |
+
"grad_norm": 0.24820839136364292,
|
| 344 |
+
"learning_rate": 2e-05,
|
| 345 |
+
"loss": 1.1392,
|
| 346 |
+
"step": 23
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.71875,
|
| 350 |
+
"eval_loss": 0.9616628885269165,
|
| 351 |
+
"eval_runtime": 192.6673,
|
| 352 |
+
"eval_samples_per_second": 1.038,
|
| 353 |
+
"eval_steps_per_second": 0.13,
|
| 354 |
+
"step": 23
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 0.75,
|
| 358 |
+
"grad_norm": 0.24589291203527217,
|
| 359 |
+
"learning_rate": 2e-05,
|
| 360 |
+
"loss": 1.058,
|
| 361 |
+
"step": 24
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"epoch": 0.75,
|
| 365 |
+
"eval_loss": 0.9531083703041077,
|
| 366 |
+
"eval_runtime": 193.0994,
|
| 367 |
+
"eval_samples_per_second": 1.036,
|
| 368 |
+
"eval_steps_per_second": 0.129,
|
| 369 |
+
"step": 24
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 0.78125,
|
| 373 |
+
"grad_norm": 0.249532642718915,
|
| 374 |
+
"learning_rate": 2e-05,
|
| 375 |
+
"loss": 0.938,
|
| 376 |
+
"step": 25
|
| 377 |
+
},
|
| 378 |
+
{
|
| 379 |
+
"epoch": 0.78125,
|
| 380 |
+
"eval_loss": 0.9455437660217285,
|
| 381 |
+
"eval_runtime": 191.9941,
|
| 382 |
+
"eval_samples_per_second": 1.042,
|
| 383 |
+
"eval_steps_per_second": 0.13,
|
| 384 |
+
"step": 25
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 0.8125,
|
| 388 |
+
"grad_norm": 0.28034242585086017,
|
| 389 |
+
"learning_rate": 2e-05,
|
| 390 |
+
"loss": 0.9387,
|
| 391 |
+
"step": 26
|
| 392 |
+
},
|
| 393 |
+
{
|
| 394 |
+
"epoch": 0.8125,
|
| 395 |
+
"eval_loss": 0.93752121925354,
|
| 396 |
+
"eval_runtime": 195.4083,
|
| 397 |
+
"eval_samples_per_second": 1.023,
|
| 398 |
+
"eval_steps_per_second": 0.128,
|
| 399 |
+
"step": 26
|
| 400 |
+
},
|
| 401 |
+
{
|
| 402 |
+
"epoch": 0.84375,
|
| 403 |
+
"grad_norm": 0.2692565070546352,
|
| 404 |
+
"learning_rate": 2e-05,
|
| 405 |
+
"loss": 1.0474,
|
| 406 |
+
"step": 27
|
| 407 |
+
},
|
| 408 |
+
{
|
| 409 |
+
"epoch": 0.84375,
|
| 410 |
+
"eval_loss": 0.9300512075424194,
|
| 411 |
+
"eval_runtime": 195.3651,
|
| 412 |
+
"eval_samples_per_second": 1.024,
|
| 413 |
+
"eval_steps_per_second": 0.128,
|
| 414 |
+
"step": 27
|
| 415 |
+
},
|
| 416 |
+
{
|
| 417 |
+
"epoch": 0.875,
|
| 418 |
+
"grad_norm": 0.24705041646949316,
|
| 419 |
+
"learning_rate": 2e-05,
|
| 420 |
+
"loss": 0.9596,
|
| 421 |
+
"step": 28
|
| 422 |
+
},
|
| 423 |
+
{
|
| 424 |
+
"epoch": 0.875,
|
| 425 |
+
"eval_loss": 0.9226720929145813,
|
| 426 |
+
"eval_runtime": 195.4848,
|
| 427 |
+
"eval_samples_per_second": 1.023,
|
| 428 |
+
"eval_steps_per_second": 0.128,
|
| 429 |
+
"step": 28
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.90625,
|
| 433 |
+
"grad_norm": 0.24799871352606165,
|
| 434 |
+
"learning_rate": 2e-05,
|
| 435 |
+
"loss": 1.0172,
|
| 436 |
+
"step": 29
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.90625,
|
| 440 |
+
"eval_loss": 0.9159422516822815,
|
| 441 |
+
"eval_runtime": 196.157,
|
| 442 |
+
"eval_samples_per_second": 1.02,
|
| 443 |
+
"eval_steps_per_second": 0.127,
|
| 444 |
+
"step": 29
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 0.9375,
|
| 448 |
+
"grad_norm": 0.29755264040904106,
|
| 449 |
+
"learning_rate": 2e-05,
|
| 450 |
+
"loss": 0.9324,
|
| 451 |
+
"step": 30
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 0.9375,
|
| 455 |
+
"eval_loss": 0.9090733528137207,
|
| 456 |
+
"eval_runtime": 196.5295,
|
| 457 |
+
"eval_samples_per_second": 1.018,
|
| 458 |
+
"eval_steps_per_second": 0.127,
|
| 459 |
+
"step": 30
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 0.96875,
|
| 463 |
+
"grad_norm": 0.2629221961008751,
|
| 464 |
+
"learning_rate": 2e-05,
|
| 465 |
+
"loss": 0.9265,
|
| 466 |
+
"step": 31
|
| 467 |
+
},
|
| 468 |
+
{
|
| 469 |
+
"epoch": 0.96875,
|
| 470 |
+
"eval_loss": 0.9027940630912781,
|
| 471 |
+
"eval_runtime": 196.281,
|
| 472 |
+
"eval_samples_per_second": 1.019,
|
| 473 |
+
"eval_steps_per_second": 0.127,
|
| 474 |
+
"step": 31
|
| 475 |
+
},
|
| 476 |
+
{
|
| 477 |
+
"epoch": 1.0,
|
| 478 |
+
"grad_norm": 0.2901110704218056,
|
| 479 |
+
"learning_rate": 2e-05,
|
| 480 |
+
"loss": 0.9933,
|
| 481 |
+
"step": 32
|
| 482 |
+
},
|
| 483 |
+
{
|
| 484 |
+
"epoch": 1.0,
|
| 485 |
+
"eval_loss": 0.8970211148262024,
|
| 486 |
+
"eval_runtime": 190.2702,
|
| 487 |
+
"eval_samples_per_second": 1.051,
|
| 488 |
+
"eval_steps_per_second": 0.131,
|
| 489 |
+
"step": 32
|
| 490 |
+
},
|
| 491 |
+
{
|
| 492 |
+
"epoch": 1.03125,
|
| 493 |
+
"grad_norm": 0.27746608883483487,
|
| 494 |
+
"learning_rate": 2e-05,
|
| 495 |
+
"loss": 0.9339,
|
| 496 |
+
"step": 33
|
| 497 |
+
},
|
| 498 |
+
{
|
| 499 |
+
"epoch": 1.03125,
|
| 500 |
+
"eval_loss": 0.8916085958480835,
|
| 501 |
+
"eval_runtime": 189.4543,
|
| 502 |
+
"eval_samples_per_second": 1.056,
|
| 503 |
+
"eval_steps_per_second": 0.132,
|
| 504 |
+
"step": 33
|
| 505 |
+
},
|
| 506 |
+
{
|
| 507 |
+
"epoch": 1.0625,
|
| 508 |
+
"grad_norm": 0.26134437145600353,
|
| 509 |
+
"learning_rate": 2e-05,
|
| 510 |
+
"loss": 0.9438,
|
| 511 |
+
"step": 34
|
| 512 |
+
},
|
| 513 |
+
{
|
| 514 |
+
"epoch": 1.0625,
|
| 515 |
+
"eval_loss": 0.8867039680480957,
|
| 516 |
+
"eval_runtime": 189.6926,
|
| 517 |
+
"eval_samples_per_second": 1.054,
|
| 518 |
+
"eval_steps_per_second": 0.132,
|
| 519 |
+
"step": 34
|
| 520 |
+
},
|
| 521 |
+
{
|
| 522 |
+
"epoch": 1.09375,
|
| 523 |
+
"grad_norm": 0.252882507519195,
|
| 524 |
+
"learning_rate": 2e-05,
|
| 525 |
+
"loss": 0.8979,
|
| 526 |
+
"step": 35
|
| 527 |
+
},
|
| 528 |
+
{
|
| 529 |
+
"epoch": 1.09375,
|
| 530 |
+
"eval_loss": 0.8824067711830139,
|
| 531 |
+
"eval_runtime": 189.9217,
|
| 532 |
+
"eval_samples_per_second": 1.053,
|
| 533 |
+
"eval_steps_per_second": 0.132,
|
| 534 |
+
"step": 35
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 1.125,
|
| 538 |
+
"grad_norm": 0.25443025949474585,
|
| 539 |
+
"learning_rate": 2e-05,
|
| 540 |
+
"loss": 0.9411,
|
| 541 |
+
"step": 36
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 1.125,
|
| 545 |
+
"eval_loss": 0.8788293600082397,
|
| 546 |
+
"eval_runtime": 191.4083,
|
| 547 |
+
"eval_samples_per_second": 1.045,
|
| 548 |
+
"eval_steps_per_second": 0.131,
|
| 549 |
+
"step": 36
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 1.15625,
|
| 553 |
+
"grad_norm": 0.2559343621244427,
|
| 554 |
+
"learning_rate": 2e-05,
|
| 555 |
+
"loss": 0.9827,
|
| 556 |
+
"step": 37
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 1.15625,
|
| 560 |
+
"eval_loss": 0.8760793805122375,
|
| 561 |
+
"eval_runtime": 191.2732,
|
| 562 |
+
"eval_samples_per_second": 1.046,
|
| 563 |
+
"eval_steps_per_second": 0.131,
|
| 564 |
+
"step": 37
|
| 565 |
+
},
|
| 566 |
+
{
|
| 567 |
+
"epoch": 1.1875,
|
| 568 |
+
"grad_norm": 0.25403189851366254,
|
| 569 |
+
"learning_rate": 2e-05,
|
| 570 |
+
"loss": 0.8658,
|
| 571 |
+
"step": 38
|
| 572 |
+
},
|
| 573 |
+
{
|
| 574 |
+
"epoch": 1.1875,
|
| 575 |
+
"eval_loss": 0.8727380633354187,
|
| 576 |
+
"eval_runtime": 190.4281,
|
| 577 |
+
"eval_samples_per_second": 1.05,
|
| 578 |
+
"eval_steps_per_second": 0.131,
|
| 579 |
+
"step": 38
|
| 580 |
+
},
|
| 581 |
+
{
|
| 582 |
+
"epoch": 1.21875,
|
| 583 |
+
"grad_norm": 0.2493777578005398,
|
| 584 |
+
"learning_rate": 2e-05,
|
| 585 |
+
"loss": 1.0053,
|
| 586 |
+
"step": 39
|
| 587 |
+
},
|
| 588 |
+
{
|
| 589 |
+
"epoch": 1.21875,
|
| 590 |
+
"eval_loss": 0.869698703289032,
|
| 591 |
+
"eval_runtime": 190.3431,
|
| 592 |
+
"eval_samples_per_second": 1.051,
|
| 593 |
+
"eval_steps_per_second": 0.131,
|
| 594 |
+
"step": 39
|
| 595 |
+
},
|
| 596 |
+
{
|
| 597 |
+
"epoch": 1.25,
|
| 598 |
+
"grad_norm": 0.24823573574563138,
|
| 599 |
+
"learning_rate": 2e-05,
|
| 600 |
+
"loss": 0.8967,
|
| 601 |
+
"step": 40
|
| 602 |
+
},
|
| 603 |
+
{
|
| 604 |
+
"epoch": 1.25,
|
| 605 |
+
"eval_loss": 0.8664910793304443,
|
| 606 |
+
"eval_runtime": 189.9802,
|
| 607 |
+
"eval_samples_per_second": 1.053,
|
| 608 |
+
"eval_steps_per_second": 0.132,
|
| 609 |
+
"step": 40
|
| 610 |
+
},
|
| 611 |
+
{
|
| 612 |
+
"epoch": 1.28125,
|
| 613 |
+
"grad_norm": 0.25462243237743476,
|
| 614 |
+
"learning_rate": 2e-05,
|
| 615 |
+
"loss": 1.0064,
|
| 616 |
+
"step": 41
|
| 617 |
+
},
|
| 618 |
+
{
|
| 619 |
+
"epoch": 1.28125,
|
| 620 |
+
"eval_loss": 0.8638657927513123,
|
| 621 |
+
"eval_runtime": 195.3373,
|
| 622 |
+
"eval_samples_per_second": 1.024,
|
| 623 |
+
"eval_steps_per_second": 0.128,
|
| 624 |
+
"step": 41
|
| 625 |
+
},
|
| 626 |
+
{
|
| 627 |
+
"epoch": 1.3125,
|
| 628 |
+
"grad_norm": 0.2604089386111215,
|
| 629 |
+
"learning_rate": 2e-05,
|
| 630 |
+
"loss": 0.9898,
|
| 631 |
+
"step": 42
|
| 632 |
+
},
|
| 633 |
+
{
|
| 634 |
+
"epoch": 1.3125,
|
| 635 |
+
"eval_loss": 0.8607734441757202,
|
| 636 |
+
"eval_runtime": 195.219,
|
| 637 |
+
"eval_samples_per_second": 1.024,
|
| 638 |
+
"eval_steps_per_second": 0.128,
|
| 639 |
+
"step": 42
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 1.34375,
|
| 643 |
+
"grad_norm": 0.27139202440805793,
|
| 644 |
+
"learning_rate": 2e-05,
|
| 645 |
+
"loss": 1.0539,
|
| 646 |
+
"step": 43
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 1.34375,
|
| 650 |
+
"eval_loss": 0.8573687672615051,
|
| 651 |
+
"eval_runtime": 195.8828,
|
| 652 |
+
"eval_samples_per_second": 1.021,
|
| 653 |
+
"eval_steps_per_second": 0.128,
|
| 654 |
+
"step": 43
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 1.375,
|
| 658 |
+
"grad_norm": 0.27474433057157854,
|
| 659 |
+
"learning_rate": 2e-05,
|
| 660 |
+
"loss": 0.86,
|
| 661 |
+
"step": 44
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 1.375,
|
| 665 |
+
"eval_loss": 0.8537396192550659,
|
| 666 |
+
"eval_runtime": 194.9741,
|
| 667 |
+
"eval_samples_per_second": 1.026,
|
| 668 |
+
"eval_steps_per_second": 0.128,
|
| 669 |
+
"step": 44
|
| 670 |
+
},
|
| 671 |
+
{
|
| 672 |
+
"epoch": 1.40625,
|
| 673 |
+
"grad_norm": 0.2537208760747199,
|
| 674 |
+
"learning_rate": 2e-05,
|
| 675 |
+
"loss": 0.9562,
|
| 676 |
+
"step": 45
|
| 677 |
+
},
|
| 678 |
+
{
|
| 679 |
+
"epoch": 1.40625,
|
| 680 |
+
"eval_loss": 0.8497809767723083,
|
| 681 |
+
"eval_runtime": 194.9162,
|
| 682 |
+
"eval_samples_per_second": 1.026,
|
| 683 |
+
"eval_steps_per_second": 0.128,
|
| 684 |
+
"step": 45
|
| 685 |
+
},
|
| 686 |
+
{
|
| 687 |
+
"epoch": 1.4375,
|
| 688 |
+
"grad_norm": 0.27560461131090846,
|
| 689 |
+
"learning_rate": 2e-05,
|
| 690 |
+
"loss": 0.8767,
|
| 691 |
+
"step": 46
|
| 692 |
+
},
|
| 693 |
+
{
|
| 694 |
+
"epoch": 1.4375,
|
| 695 |
+
"eval_loss": 0.8458660244941711,
|
| 696 |
+
"eval_runtime": 195.1157,
|
| 697 |
+
"eval_samples_per_second": 1.025,
|
| 698 |
+
"eval_steps_per_second": 0.128,
|
| 699 |
+
"step": 46
|
| 700 |
+
},
|
| 701 |
+
{
|
| 702 |
+
"epoch": 1.46875,
|
| 703 |
+
"grad_norm": 0.2594536794112662,
|
| 704 |
+
"learning_rate": 2e-05,
|
| 705 |
+
"loss": 0.9256,
|
| 706 |
+
"step": 47
|
| 707 |
+
},
|
| 708 |
+
{
|
| 709 |
+
"epoch": 1.46875,
|
| 710 |
+
"eval_loss": 0.8429936766624451,
|
| 711 |
+
"eval_runtime": 195.048,
|
| 712 |
+
"eval_samples_per_second": 1.025,
|
| 713 |
+
"eval_steps_per_second": 0.128,
|
| 714 |
+
"step": 47
|
| 715 |
+
},
|
| 716 |
+
{
|
| 717 |
+
"epoch": 1.5,
|
| 718 |
+
"grad_norm": 0.28583207453838866,
|
| 719 |
+
"learning_rate": 2e-05,
|
| 720 |
+
"loss": 0.9858,
|
| 721 |
+
"step": 48
|
| 722 |
+
},
|
| 723 |
+
{
|
| 724 |
+
"epoch": 1.5,
|
| 725 |
+
"eval_loss": 0.84013831615448,
|
| 726 |
+
"eval_runtime": 194.3046,
|
| 727 |
+
"eval_samples_per_second": 1.029,
|
| 728 |
+
"eval_steps_per_second": 0.129,
|
| 729 |
+
"step": 48
|
| 730 |
+
},
|
| 731 |
+
{
|
| 732 |
+
"epoch": 1.53125,
|
| 733 |
+
"grad_norm": 0.28118976636788506,
|
| 734 |
+
"learning_rate": 2e-05,
|
| 735 |
+
"loss": 0.9158,
|
| 736 |
+
"step": 49
|
| 737 |
+
},
|
| 738 |
+
{
|
| 739 |
+
"epoch": 1.53125,
|
| 740 |
+
"eval_loss": 0.8369531035423279,
|
| 741 |
+
"eval_runtime": 194.521,
|
| 742 |
+
"eval_samples_per_second": 1.028,
|
| 743 |
+
"eval_steps_per_second": 0.129,
|
| 744 |
+
"step": 49
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 1.5625,
|
| 748 |
+
"grad_norm": 0.29276573776696546,
|
| 749 |
+
"learning_rate": 2e-05,
|
| 750 |
+
"loss": 0.8745,
|
| 751 |
+
"step": 50
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 1.5625,
|
| 755 |
+
"eval_loss": 0.8341982960700989,
|
| 756 |
+
"eval_runtime": 194.1114,
|
| 757 |
+
"eval_samples_per_second": 1.03,
|
| 758 |
+
"eval_steps_per_second": 0.129,
|
| 759 |
+
"step": 50
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 1.59375,
|
| 763 |
+
"grad_norm": 0.2860638141439372,
|
| 764 |
+
"learning_rate": 2e-05,
|
| 765 |
+
"loss": 0.854,
|
| 766 |
+
"step": 51
|
| 767 |
+
},
|
| 768 |
+
{
|
| 769 |
+
"epoch": 1.59375,
|
| 770 |
+
"eval_loss": 0.8317239284515381,
|
| 771 |
+
"eval_runtime": 198.1029,
|
| 772 |
+
"eval_samples_per_second": 1.01,
|
| 773 |
+
"eval_steps_per_second": 0.126,
|
| 774 |
+
"step": 51
|
| 775 |
+
},
|
| 776 |
+
{
|
| 777 |
+
"epoch": 1.625,
|
| 778 |
+
"grad_norm": 0.29960349722496704,
|
| 779 |
+
"learning_rate": 2e-05,
|
| 780 |
+
"loss": 0.8399,
|
| 781 |
+
"step": 52
|
| 782 |
+
},
|
| 783 |
+
{
|
| 784 |
+
"epoch": 1.625,
|
| 785 |
+
"eval_loss": 0.8290513753890991,
|
| 786 |
+
"eval_runtime": 198.0764,
|
| 787 |
+
"eval_samples_per_second": 1.01,
|
| 788 |
+
"eval_steps_per_second": 0.126,
|
| 789 |
+
"step": 52
|
| 790 |
+
},
|
| 791 |
+
{
|
| 792 |
+
"epoch": 1.65625,
|
| 793 |
+
"grad_norm": 0.2964234305808419,
|
| 794 |
+
"learning_rate": 2e-05,
|
| 795 |
+
"loss": 0.9694,
|
| 796 |
+
"step": 53
|
| 797 |
+
},
|
| 798 |
+
{
|
| 799 |
+
"epoch": 1.65625,
|
| 800 |
+
"eval_loss": 0.8267760276794434,
|
| 801 |
+
"eval_runtime": 197.8284,
|
| 802 |
+
"eval_samples_per_second": 1.011,
|
| 803 |
+
"eval_steps_per_second": 0.126,
|
| 804 |
+
"step": 53
|
| 805 |
+
},
|
| 806 |
+
{
|
| 807 |
+
"epoch": 1.6875,
|
| 808 |
+
"grad_norm": 0.26183932644077784,
|
| 809 |
+
"learning_rate": 2e-05,
|
| 810 |
+
"loss": 0.8153,
|
| 811 |
+
"step": 54
|
| 812 |
+
},
|
| 813 |
+
{
|
| 814 |
+
"epoch": 1.6875,
|
| 815 |
+
"eval_loss": 0.824044942855835,
|
| 816 |
+
"eval_runtime": 198.1694,
|
| 817 |
+
"eval_samples_per_second": 1.009,
|
| 818 |
+
"eval_steps_per_second": 0.126,
|
| 819 |
+
"step": 54
|
| 820 |
+
},
|
| 821 |
+
{
|
| 822 |
+
"epoch": 1.71875,
|
| 823 |
+
"grad_norm": 0.3067024314453144,
|
| 824 |
+
"learning_rate": 2e-05,
|
| 825 |
+
"loss": 0.883,
|
| 826 |
+
"step": 55
|
| 827 |
+
},
|
| 828 |
+
{
|
| 829 |
+
"epoch": 1.71875,
|
| 830 |
+
"eval_loss": 0.8216392397880554,
|
| 831 |
+
"eval_runtime": 198.0249,
|
| 832 |
+
"eval_samples_per_second": 1.01,
|
| 833 |
+
"eval_steps_per_second": 0.126,
|
| 834 |
+
"step": 55
|
| 835 |
+
},
|
| 836 |
+
{
|
| 837 |
+
"epoch": 1.75,
|
| 838 |
+
"grad_norm": 0.27888658705355013,
|
| 839 |
+
"learning_rate": 2e-05,
|
| 840 |
+
"loss": 0.8771,
|
| 841 |
+
"step": 56
|
| 842 |
+
},
|
| 843 |
+
{
|
| 844 |
+
"epoch": 1.75,
|
| 845 |
+
"eval_loss": 0.8194215297698975,
|
| 846 |
+
"eval_runtime": 195.4688,
|
| 847 |
+
"eval_samples_per_second": 1.023,
|
| 848 |
+
"eval_steps_per_second": 0.128,
|
| 849 |
+
"step": 56
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 1.78125,
|
| 853 |
+
"grad_norm": 0.32571765544245934,
|
| 854 |
+
"learning_rate": 2e-05,
|
| 855 |
+
"loss": 0.897,
|
| 856 |
+
"step": 57
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 1.78125,
|
| 860 |
+
"eval_loss": 0.8167170882225037,
|
| 861 |
+
"eval_runtime": 189.6243,
|
| 862 |
+
"eval_samples_per_second": 1.055,
|
| 863 |
+
"eval_steps_per_second": 0.132,
|
| 864 |
+
"step": 57
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 1.8125,
|
| 868 |
+
"grad_norm": 0.292216058855145,
|
| 869 |
+
"learning_rate": 2e-05,
|
| 870 |
+
"loss": 0.9277,
|
| 871 |
+
"step": 58
|
| 872 |
+
},
|
| 873 |
+
{
|
| 874 |
+
"epoch": 1.8125,
|
| 875 |
+
"eval_loss": 0.8145509958267212,
|
| 876 |
+
"eval_runtime": 190.2429,
|
| 877 |
+
"eval_samples_per_second": 1.051,
|
| 878 |
+
"eval_steps_per_second": 0.131,
|
| 879 |
+
"step": 58
|
| 880 |
+
},
|
| 881 |
+
{
|
| 882 |
+
"epoch": 1.84375,
|
| 883 |
+
"grad_norm": 0.29002612820437024,
|
| 884 |
+
"learning_rate": 2e-05,
|
| 885 |
+
"loss": 0.8971,
|
| 886 |
+
"step": 59
|
| 887 |
+
},
|
| 888 |
+
{
|
| 889 |
+
"epoch": 1.84375,
|
| 890 |
+
"eval_loss": 0.8122230768203735,
|
| 891 |
+
"eval_runtime": 189.9403,
|
| 892 |
+
"eval_samples_per_second": 1.053,
|
| 893 |
+
"eval_steps_per_second": 0.132,
|
| 894 |
+
"step": 59
|
| 895 |
+
},
|
| 896 |
+
{
|
| 897 |
+
"epoch": 1.875,
|
| 898 |
+
"grad_norm": 0.2926088029288858,
|
| 899 |
+
"learning_rate": 2e-05,
|
| 900 |
+
"loss": 0.9225,
|
| 901 |
+
"step": 60
|
| 902 |
+
},
|
| 903 |
+
{
|
| 904 |
+
"epoch": 1.875,
|
| 905 |
+
"eval_loss": 0.8100479245185852,
|
| 906 |
+
"eval_runtime": 190.2569,
|
| 907 |
+
"eval_samples_per_second": 1.051,
|
| 908 |
+
"eval_steps_per_second": 0.131,
|
| 909 |
+
"step": 60
|
| 910 |
+
},
|
| 911 |
+
{
|
| 912 |
+
"epoch": 1.90625,
|
| 913 |
+
"grad_norm": 0.30068993111077397,
|
| 914 |
+
"learning_rate": 2e-05,
|
| 915 |
+
"loss": 0.9134,
|
| 916 |
+
"step": 61
|
| 917 |
+
},
|
| 918 |
+
{
|
| 919 |
+
"epoch": 1.90625,
|
| 920 |
+
"eval_loss": 0.808087944984436,
|
| 921 |
+
"eval_runtime": 192.4896,
|
| 922 |
+
"eval_samples_per_second": 1.039,
|
| 923 |
+
"eval_steps_per_second": 0.13,
|
| 924 |
+
"step": 61
|
| 925 |
+
},
|
| 926 |
+
{
|
| 927 |
+
"epoch": 1.9375,
|
| 928 |
+
"grad_norm": 0.3157573686768343,
|
| 929 |
+
"learning_rate": 2e-05,
|
| 930 |
+
"loss": 0.8965,
|
| 931 |
+
"step": 62
|
| 932 |
+
},
|
| 933 |
+
{
|
| 934 |
+
"epoch": 1.9375,
|
| 935 |
+
"eval_loss": 0.8057371377944946,
|
| 936 |
+
"eval_runtime": 190.0158,
|
| 937 |
+
"eval_samples_per_second": 1.053,
|
| 938 |
+
"eval_steps_per_second": 0.132,
|
| 939 |
+
"step": 62
|
| 940 |
+
},
|
| 941 |
+
{
|
| 942 |
+
"epoch": 1.96875,
|
| 943 |
+
"grad_norm": 0.31215592754506605,
|
| 944 |
+
"learning_rate": 2e-05,
|
| 945 |
+
"loss": 0.7828,
|
| 946 |
+
"step": 63
|
| 947 |
+
},
|
| 948 |
+
{
|
| 949 |
+
"epoch": 1.96875,
|
| 950 |
+
"eval_loss": 0.8031384944915771,
|
| 951 |
+
"eval_runtime": 189.5204,
|
| 952 |
+
"eval_samples_per_second": 1.055,
|
| 953 |
+
"eval_steps_per_second": 0.132,
|
| 954 |
+
"step": 63
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 2.0,
|
| 958 |
+
"grad_norm": 0.29422828766227993,
|
| 959 |
+
"learning_rate": 2e-05,
|
| 960 |
+
"loss": 0.8196,
|
| 961 |
+
"step": 64
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 2.0,
|
| 965 |
+
"eval_loss": 0.8012601733207703,
|
| 966 |
+
"eval_runtime": 189.7041,
|
| 967 |
+
"eval_samples_per_second": 1.054,
|
| 968 |
+
"eval_steps_per_second": 0.132,
|
| 969 |
+
"step": 64
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 2.03125,
|
| 973 |
+
"grad_norm": 0.2885449518895793,
|
| 974 |
+
"learning_rate": 2e-05,
|
| 975 |
+
"loss": 0.9715,
|
| 976 |
+
"step": 65
|
| 977 |
+
},
|
| 978 |
+
{
|
| 979 |
+
"epoch": 2.03125,
|
| 980 |
+
"eval_loss": 0.8001161813735962,
|
| 981 |
+
"eval_runtime": 189.57,
|
| 982 |
+
"eval_samples_per_second": 1.055,
|
| 983 |
+
"eval_steps_per_second": 0.132,
|
| 984 |
+
"step": 65
|
| 985 |
+
},
|
| 986 |
+
{
|
| 987 |
+
"epoch": 2.0625,
|
| 988 |
+
"grad_norm": 0.30260184063348483,
|
| 989 |
+
"learning_rate": 2e-05,
|
| 990 |
+
"loss": 0.7912,
|
| 991 |
+
"step": 66
|
| 992 |
+
},
|
| 993 |
+
{
|
| 994 |
+
"epoch": 2.0625,
|
| 995 |
+
"eval_loss": 0.7989436388015747,
|
| 996 |
+
"eval_runtime": 193.0193,
|
| 997 |
+
"eval_samples_per_second": 1.036,
|
| 998 |
+
"eval_steps_per_second": 0.13,
|
| 999 |
+
"step": 66
|
| 1000 |
+
},
|
| 1001 |
+
{
|
| 1002 |
+
"epoch": 2.09375,
|
| 1003 |
+
"grad_norm": 0.32650294605024255,
|
| 1004 |
+
"learning_rate": 2e-05,
|
| 1005 |
+
"loss": 0.8176,
|
| 1006 |
+
"step": 67
|
| 1007 |
+
},
|
| 1008 |
+
{
|
| 1009 |
+
"epoch": 2.09375,
|
| 1010 |
+
"eval_loss": 0.7972333431243896,
|
| 1011 |
+
"eval_runtime": 193.2225,
|
| 1012 |
+
"eval_samples_per_second": 1.035,
|
| 1013 |
+
"eval_steps_per_second": 0.129,
|
| 1014 |
+
"step": 67
|
| 1015 |
+
},
|
| 1016 |
+
{
|
| 1017 |
+
"epoch": 2.125,
|
| 1018 |
+
"grad_norm": 0.3382679480741134,
|
| 1019 |
+
"learning_rate": 2e-05,
|
| 1020 |
+
"loss": 0.8141,
|
| 1021 |
+
"step": 68
|
| 1022 |
+
},
|
| 1023 |
+
{
|
| 1024 |
+
"epoch": 2.125,
|
| 1025 |
+
"eval_loss": 0.7950598001480103,
|
| 1026 |
+
"eval_runtime": 193.2781,
|
| 1027 |
+
"eval_samples_per_second": 1.035,
|
| 1028 |
+
"eval_steps_per_second": 0.129,
|
| 1029 |
+
"step": 68
|
| 1030 |
+
},
|
| 1031 |
+
{
|
| 1032 |
+
"epoch": 2.15625,
|
| 1033 |
+
"grad_norm": 0.3094090784935889,
|
| 1034 |
+
"learning_rate": 2e-05,
|
| 1035 |
+
"loss": 0.796,
|
| 1036 |
+
"step": 69
|
| 1037 |
+
},
|
| 1038 |
+
{
|
| 1039 |
+
"epoch": 2.15625,
|
| 1040 |
+
"eval_loss": 0.7932476997375488,
|
| 1041 |
+
"eval_runtime": 193.0282,
|
| 1042 |
+
"eval_samples_per_second": 1.036,
|
| 1043 |
+
"eval_steps_per_second": 0.13,
|
| 1044 |
+
"step": 69
|
| 1045 |
+
},
|
| 1046 |
+
{
|
| 1047 |
+
"epoch": 2.1875,
|
| 1048 |
+
"grad_norm": 0.30209558834780514,
|
| 1049 |
+
"learning_rate": 2e-05,
|
| 1050 |
+
"loss": 0.927,
|
| 1051 |
+
"step": 70
|
| 1052 |
+
},
|
| 1053 |
+
{
|
| 1054 |
+
"epoch": 2.1875,
|
| 1055 |
+
"eval_loss": 0.7922118902206421,
|
| 1056 |
+
"eval_runtime": 192.8284,
|
| 1057 |
+
"eval_samples_per_second": 1.037,
|
| 1058 |
+
"eval_steps_per_second": 0.13,
|
| 1059 |
+
"step": 70
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 2.21875,
|
| 1063 |
+
"grad_norm": 0.35958652905266686,
|
| 1064 |
+
"learning_rate": 2e-05,
|
| 1065 |
+
"loss": 0.8224,
|
| 1066 |
+
"step": 71
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 2.21875,
|
| 1070 |
+
"eval_loss": 0.7909810543060303,
|
| 1071 |
+
"eval_runtime": 202.3128,
|
| 1072 |
+
"eval_samples_per_second": 0.989,
|
| 1073 |
+
"eval_steps_per_second": 0.124,
|
| 1074 |
+
"step": 71
|
| 1075 |
+
},
|
| 1076 |
+
{
|
| 1077 |
+
"epoch": 2.25,
|
| 1078 |
+
"grad_norm": 0.356338004067507,
|
| 1079 |
+
"learning_rate": 2e-05,
|
| 1080 |
+
"loss": 0.8376,
|
| 1081 |
+
"step": 72
|
| 1082 |
+
},
|
| 1083 |
+
{
|
| 1084 |
+
"epoch": 2.25,
|
| 1085 |
+
"eval_loss": 0.7894656658172607,
|
| 1086 |
+
"eval_runtime": 195.1481,
|
| 1087 |
+
"eval_samples_per_second": 1.025,
|
| 1088 |
+
"eval_steps_per_second": 0.128,
|
| 1089 |
+
"step": 72
|
| 1090 |
+
},
|
| 1091 |
+
{
|
| 1092 |
+
"epoch": 2.28125,
|
| 1093 |
+
"grad_norm": 0.31886905989727465,
|
| 1094 |
+
"learning_rate": 2e-05,
|
| 1095 |
+
"loss": 0.8688,
|
| 1096 |
+
"step": 73
|
| 1097 |
+
},
|
| 1098 |
+
{
|
| 1099 |
+
"epoch": 2.28125,
|
| 1100 |
+
"eval_loss": 0.7889463901519775,
|
| 1101 |
+
"eval_runtime": 194.8418,
|
| 1102 |
+
"eval_samples_per_second": 1.026,
|
| 1103 |
+
"eval_steps_per_second": 0.128,
|
| 1104 |
+
"step": 73
|
| 1105 |
+
},
|
| 1106 |
+
{
|
| 1107 |
+
"epoch": 2.3125,
|
| 1108 |
+
"grad_norm": 0.35606342918056466,
|
| 1109 |
+
"learning_rate": 2e-05,
|
| 1110 |
+
"loss": 0.835,
|
| 1111 |
+
"step": 74
|
| 1112 |
+
},
|
| 1113 |
+
{
|
| 1114 |
+
"epoch": 2.3125,
|
| 1115 |
+
"eval_loss": 0.7875587344169617,
|
| 1116 |
+
"eval_runtime": 194.7701,
|
| 1117 |
+
"eval_samples_per_second": 1.027,
|
| 1118 |
+
"eval_steps_per_second": 0.128,
|
| 1119 |
+
"step": 74
|
| 1120 |
+
},
|
| 1121 |
+
{
|
| 1122 |
+
"epoch": 2.34375,
|
| 1123 |
+
"grad_norm": 0.3161858862696026,
|
| 1124 |
+
"learning_rate": 2e-05,
|
| 1125 |
+
"loss": 0.873,
|
| 1126 |
+
"step": 75
|
| 1127 |
+
},
|
| 1128 |
+
{
|
| 1129 |
+
"epoch": 2.34375,
|
| 1130 |
+
"eval_loss": 0.7863460779190063,
|
| 1131 |
+
"eval_runtime": 195.4811,
|
| 1132 |
+
"eval_samples_per_second": 1.023,
|
| 1133 |
+
"eval_steps_per_second": 0.128,
|
| 1134 |
+
"step": 75
|
| 1135 |
+
},
|
| 1136 |
+
{
|
| 1137 |
+
"epoch": 2.375,
|
| 1138 |
+
"grad_norm": 0.35771781884741477,
|
| 1139 |
+
"learning_rate": 2e-05,
|
| 1140 |
+
"loss": 0.9021,
|
| 1141 |
+
"step": 76
|
| 1142 |
+
},
|
| 1143 |
+
{
|
| 1144 |
+
"epoch": 2.375,
|
| 1145 |
+
"eval_loss": 0.7847577929496765,
|
| 1146 |
+
"eval_runtime": 195.3724,
|
| 1147 |
+
"eval_samples_per_second": 1.024,
|
| 1148 |
+
"eval_steps_per_second": 0.128,
|
| 1149 |
+
"step": 76
|
| 1150 |
+
},
|
| 1151 |
+
{
|
| 1152 |
+
"epoch": 2.40625,
|
| 1153 |
+
"grad_norm": 0.3549789155823785,
|
| 1154 |
+
"learning_rate": 2e-05,
|
| 1155 |
+
"loss": 0.9195,
|
| 1156 |
+
"step": 77
|
| 1157 |
+
},
|
| 1158 |
+
{
|
| 1159 |
+
"epoch": 2.40625,
|
| 1160 |
+
"eval_loss": 0.783415675163269,
|
| 1161 |
+
"eval_runtime": 190.226,
|
| 1162 |
+
"eval_samples_per_second": 1.051,
|
| 1163 |
+
"eval_steps_per_second": 0.131,
|
| 1164 |
+
"step": 77
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 2.4375,
|
| 1168 |
+
"grad_norm": 0.34734314309709374,
|
| 1169 |
+
"learning_rate": 2e-05,
|
| 1170 |
+
"loss": 0.8386,
|
| 1171 |
+
"step": 78
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 2.4375,
|
| 1175 |
+
"eval_loss": 0.7814657688140869,
|
| 1176 |
+
"eval_runtime": 190.3177,
|
| 1177 |
+
"eval_samples_per_second": 1.051,
|
| 1178 |
+
"eval_steps_per_second": 0.131,
|
| 1179 |
+
"step": 78
|
| 1180 |
+
},
|
| 1181 |
+
{
|
| 1182 |
+
"epoch": 2.46875,
|
| 1183 |
+
"grad_norm": 0.35540762574897183,
|
| 1184 |
+
"learning_rate": 2e-05,
|
| 1185 |
+
"loss": 0.851,
|
| 1186 |
+
"step": 79
|
| 1187 |
+
},
|
| 1188 |
+
{
|
| 1189 |
+
"epoch": 2.46875,
|
| 1190 |
+
"eval_loss": 0.7798058390617371,
|
| 1191 |
+
"eval_runtime": 190.7447,
|
| 1192 |
+
"eval_samples_per_second": 1.049,
|
| 1193 |
+
"eval_steps_per_second": 0.131,
|
| 1194 |
+
"step": 79
|
| 1195 |
+
},
|
| 1196 |
+
{
|
| 1197 |
+
"epoch": 2.5,
|
| 1198 |
+
"grad_norm": 0.3844458514717174,
|
| 1199 |
+
"learning_rate": 2e-05,
|
| 1200 |
+
"loss": 0.767,
|
| 1201 |
+
"step": 80
|
| 1202 |
+
},
|
| 1203 |
+
{
|
| 1204 |
+
"epoch": 2.5,
|
| 1205 |
+
"eval_loss": 0.7777827978134155,
|
| 1206 |
+
"eval_runtime": 190.8548,
|
| 1207 |
+
"eval_samples_per_second": 1.048,
|
| 1208 |
+
"eval_steps_per_second": 0.131,
|
| 1209 |
+
"step": 80
|
| 1210 |
+
},
|
| 1211 |
+
{
|
| 1212 |
+
"epoch": 2.53125,
|
| 1213 |
+
"grad_norm": 0.36232344175264375,
|
| 1214 |
+
"learning_rate": 2e-05,
|
| 1215 |
+
"loss": 0.8508,
|
| 1216 |
+
"step": 81
|
| 1217 |
+
},
|
| 1218 |
+
{
|
| 1219 |
+
"epoch": 2.53125,
|
| 1220 |
+
"eval_loss": 0.7763205170631409,
|
| 1221 |
+
"eval_runtime": 190.335,
|
| 1222 |
+
"eval_samples_per_second": 1.051,
|
| 1223 |
+
"eval_steps_per_second": 0.131,
|
| 1224 |
+
"step": 81
|
| 1225 |
+
},
|
| 1226 |
+
{
|
| 1227 |
+
"epoch": 2.5625,
|
| 1228 |
+
"grad_norm": 0.36279843147857743,
|
| 1229 |
+
"learning_rate": 2e-05,
|
| 1230 |
+
"loss": 0.8331,
|
| 1231 |
+
"step": 82
|
| 1232 |
+
},
|
| 1233 |
+
{
|
| 1234 |
+
"epoch": 2.5625,
|
| 1235 |
+
"eval_loss": 0.7757676839828491,
|
| 1236 |
+
"eval_runtime": 190.9559,
|
| 1237 |
+
"eval_samples_per_second": 1.047,
|
| 1238 |
+
"eval_steps_per_second": 0.131,
|
| 1239 |
+
"step": 82
|
| 1240 |
+
},
|
| 1241 |
+
{
|
| 1242 |
+
"epoch": 2.59375,
|
| 1243 |
+
"grad_norm": 0.395360566032837,
|
| 1244 |
+
"learning_rate": 2e-05,
|
| 1245 |
+
"loss": 0.847,
|
| 1246 |
+
"step": 83
|
| 1247 |
+
},
|
| 1248 |
+
{
|
| 1249 |
+
"epoch": 2.59375,
|
| 1250 |
+
"eval_loss": 0.7743326425552368,
|
| 1251 |
+
"eval_runtime": 190.6372,
|
| 1252 |
+
"eval_samples_per_second": 1.049,
|
| 1253 |
+
"eval_steps_per_second": 0.131,
|
| 1254 |
+
"step": 83
|
| 1255 |
+
},
|
| 1256 |
+
{
|
| 1257 |
+
"epoch": 2.625,
|
| 1258 |
+
"grad_norm": 0.4268568783791123,
|
| 1259 |
+
"learning_rate": 2e-05,
|
| 1260 |
+
"loss": 0.869,
|
| 1261 |
+
"step": 84
|
| 1262 |
+
},
|
| 1263 |
+
{
|
| 1264 |
+
"epoch": 2.625,
|
| 1265 |
+
"eval_loss": 0.772053062915802,
|
| 1266 |
+
"eval_runtime": 190.2072,
|
| 1267 |
+
"eval_samples_per_second": 1.051,
|
| 1268 |
+
"eval_steps_per_second": 0.131,
|
| 1269 |
+
"step": 84
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 2.65625,
|
| 1273 |
+
"grad_norm": 0.3581495538253167,
|
| 1274 |
+
"learning_rate": 2e-05,
|
| 1275 |
+
"loss": 0.8591,
|
| 1276 |
+
"step": 85
|
| 1277 |
+
},
|
| 1278 |
+
{
|
| 1279 |
+
"epoch": 2.65625,
|
| 1280 |
+
"eval_loss": 0.7711917757987976,
|
| 1281 |
+
"eval_runtime": 190.4392,
|
| 1282 |
+
"eval_samples_per_second": 1.05,
|
| 1283 |
+
"eval_steps_per_second": 0.131,
|
| 1284 |
+
"step": 85
|
| 1285 |
+
},
|
| 1286 |
+
{
|
| 1287 |
+
"epoch": 2.6875,
|
| 1288 |
+
"grad_norm": 0.3952841797586726,
|
| 1289 |
+
"learning_rate": 2e-05,
|
| 1290 |
+
"loss": 0.8167,
|
| 1291 |
+
"step": 86
|
| 1292 |
+
},
|
| 1293 |
+
{
|
| 1294 |
+
"epoch": 2.6875,
|
| 1295 |
+
"eval_loss": 0.7714033722877502,
|
| 1296 |
+
"eval_runtime": 193.8038,
|
| 1297 |
+
"eval_samples_per_second": 1.032,
|
| 1298 |
+
"eval_steps_per_second": 0.129,
|
| 1299 |
+
"step": 86
|
| 1300 |
+
},
|
| 1301 |
+
{
|
| 1302 |
+
"epoch": 2.71875,
|
| 1303 |
+
"grad_norm": 0.41820009905687616,
|
| 1304 |
+
"learning_rate": 2e-05,
|
| 1305 |
+
"loss": 0.8165,
|
| 1306 |
+
"step": 87
|
| 1307 |
+
},
|
| 1308 |
+
{
|
| 1309 |
+
"epoch": 2.71875,
|
| 1310 |
+
"eval_loss": 0.771486759185791,
|
| 1311 |
+
"eval_runtime": 194.4791,
|
| 1312 |
+
"eval_samples_per_second": 1.028,
|
| 1313 |
+
"eval_steps_per_second": 0.129,
|
| 1314 |
+
"step": 87
|
| 1315 |
+
},
|
| 1316 |
+
{
|
| 1317 |
+
"epoch": 2.75,
|
| 1318 |
+
"grad_norm": 0.3852566717747202,
|
| 1319 |
+
"learning_rate": 2e-05,
|
| 1320 |
+
"loss": 0.8459,
|
| 1321 |
+
"step": 88
|
| 1322 |
+
},
|
| 1323 |
+
{
|
| 1324 |
+
"epoch": 2.75,
|
| 1325 |
+
"eval_loss": 0.7710732817649841,
|
| 1326 |
+
"eval_runtime": 194.3404,
|
| 1327 |
+
"eval_samples_per_second": 1.029,
|
| 1328 |
+
"eval_steps_per_second": 0.129,
|
| 1329 |
+
"step": 88
|
| 1330 |
+
},
|
| 1331 |
+
{
|
| 1332 |
+
"epoch": 2.78125,
|
| 1333 |
+
"grad_norm": 0.39909292055831935,
|
| 1334 |
+
"learning_rate": 2e-05,
|
| 1335 |
+
"loss": 0.8945,
|
| 1336 |
+
"step": 89
|
| 1337 |
+
},
|
| 1338 |
+
{
|
| 1339 |
+
"epoch": 2.78125,
|
| 1340 |
+
"eval_loss": 0.7708308696746826,
|
| 1341 |
+
"eval_runtime": 194.4483,
|
| 1342 |
+
"eval_samples_per_second": 1.029,
|
| 1343 |
+
"eval_steps_per_second": 0.129,
|
| 1344 |
+
"step": 89
|
| 1345 |
+
},
|
| 1346 |
+
{
|
| 1347 |
+
"epoch": 2.8125,
|
| 1348 |
+
"grad_norm": 0.3916487629667217,
|
| 1349 |
+
"learning_rate": 2e-05,
|
| 1350 |
+
"loss": 0.8029,
|
| 1351 |
+
"step": 90
|
| 1352 |
+
},
|
| 1353 |
+
{
|
| 1354 |
+
"epoch": 2.8125,
|
| 1355 |
+
"eval_loss": 0.7713395953178406,
|
| 1356 |
+
"eval_runtime": 194.6045,
|
| 1357 |
+
"eval_samples_per_second": 1.028,
|
| 1358 |
+
"eval_steps_per_second": 0.128,
|
| 1359 |
+
"step": 90
|
| 1360 |
+
},
|
| 1361 |
+
{
|
| 1362 |
+
"epoch": 2.84375,
|
| 1363 |
+
"grad_norm": 0.36969072235715195,
|
| 1364 |
+
"learning_rate": 2e-05,
|
| 1365 |
+
"loss": 0.7704,
|
| 1366 |
+
"step": 91
|
| 1367 |
+
},
|
| 1368 |
+
{
|
| 1369 |
+
"epoch": 2.84375,
|
| 1370 |
+
"eval_loss": 0.7713618278503418,
|
| 1371 |
+
"eval_runtime": 194.3895,
|
| 1372 |
+
"eval_samples_per_second": 1.029,
|
| 1373 |
+
"eval_steps_per_second": 0.129,
|
| 1374 |
+
"step": 91
|
| 1375 |
+
},
|
| 1376 |
+
{
|
| 1377 |
+
"epoch": 2.875,
|
| 1378 |
+
"grad_norm": 0.3853248559868725,
|
| 1379 |
+
"learning_rate": 2e-05,
|
| 1380 |
+
"loss": 0.8247,
|
| 1381 |
+
"step": 92
|
| 1382 |
+
},
|
| 1383 |
+
{
|
| 1384 |
+
"epoch": 2.875,
|
| 1385 |
+
"eval_loss": 0.7703633308410645,
|
| 1386 |
+
"eval_runtime": 194.0457,
|
| 1387 |
+
"eval_samples_per_second": 1.031,
|
| 1388 |
+
"eval_steps_per_second": 0.129,
|
| 1389 |
+
"step": 92
|
| 1390 |
+
},
|
| 1391 |
+
{
|
| 1392 |
+
"epoch": 2.90625,
|
| 1393 |
+
"grad_norm": 0.38111471762069055,
|
| 1394 |
+
"learning_rate": 2e-05,
|
| 1395 |
+
"loss": 0.855,
|
| 1396 |
+
"step": 93
|
| 1397 |
+
},
|
| 1398 |
+
{
|
| 1399 |
+
"epoch": 2.90625,
|
| 1400 |
+
"eval_loss": 0.7690189480781555,
|
| 1401 |
+
"eval_runtime": 194.3506,
|
| 1402 |
+
"eval_samples_per_second": 1.029,
|
| 1403 |
+
"eval_steps_per_second": 0.129,
|
| 1404 |
+
"step": 93
|
| 1405 |
+
},
|
| 1406 |
+
{
|
| 1407 |
+
"epoch": 2.9375,
|
| 1408 |
+
"grad_norm": 0.3701270310997752,
|
| 1409 |
+
"learning_rate": 2e-05,
|
| 1410 |
+
"loss": 0.7518,
|
| 1411 |
+
"step": 94
|
| 1412 |
+
},
|
| 1413 |
+
{
|
| 1414 |
+
"epoch": 2.9375,
|
| 1415 |
+
"eval_loss": 0.7675644159317017,
|
| 1416 |
+
"eval_runtime": 194.3756,
|
| 1417 |
+
"eval_samples_per_second": 1.029,
|
| 1418 |
+
"eval_steps_per_second": 0.129,
|
| 1419 |
+
"step": 94
|
| 1420 |
+
},
|
| 1421 |
+
{
|
| 1422 |
+
"epoch": 2.96875,
|
| 1423 |
+
"grad_norm": 0.40489524752286055,
|
| 1424 |
+
"learning_rate": 2e-05,
|
| 1425 |
+
"loss": 0.8559,
|
| 1426 |
+
"step": 95
|
| 1427 |
+
},
|
| 1428 |
+
{
|
| 1429 |
+
"epoch": 2.96875,
|
| 1430 |
+
"eval_loss": 0.766002357006073,
|
| 1431 |
+
"eval_runtime": 193.9472,
|
| 1432 |
+
"eval_samples_per_second": 1.031,
|
| 1433 |
+
"eval_steps_per_second": 0.129,
|
| 1434 |
+
"step": 95
|
| 1435 |
+
},
|
| 1436 |
+
{
|
| 1437 |
+
"epoch": 3.0,
|
| 1438 |
+
"grad_norm": 0.39220887464051457,
|
| 1439 |
+
"learning_rate": 2e-05,
|
| 1440 |
+
"loss": 0.8629,
|
| 1441 |
+
"step": 96
|
| 1442 |
+
},
|
| 1443 |
+
{
|
| 1444 |
+
"epoch": 3.0,
|
| 1445 |
+
"eval_loss": 0.7644355893135071,
|
| 1446 |
+
"eval_runtime": 196.1686,
|
| 1447 |
+
"eval_samples_per_second": 1.02,
|
| 1448 |
+
"eval_steps_per_second": 0.127,
|
| 1449 |
+
"step": 96
|
| 1450 |
+
},
|
| 1451 |
+
{
|
| 1452 |
+
"epoch": 3.03125,
|
| 1453 |
+
"grad_norm": 0.3644925708419195,
|
| 1454 |
+
"learning_rate": 2e-05,
|
| 1455 |
+
"loss": 0.7434,
|
| 1456 |
+
"step": 97
|
| 1457 |
+
},
|
| 1458 |
+
{
|
| 1459 |
+
"epoch": 3.03125,
|
| 1460 |
+
"eval_loss": 0.7628399133682251,
|
| 1461 |
+
"eval_runtime": 196.1515,
|
| 1462 |
+
"eval_samples_per_second": 1.02,
|
| 1463 |
+
"eval_steps_per_second": 0.127,
|
| 1464 |
+
"step": 97
|
| 1465 |
+
},
|
| 1466 |
+
{
|
| 1467 |
+
"epoch": 3.0625,
|
| 1468 |
+
"grad_norm": 0.407089942317534,
|
| 1469 |
+
"learning_rate": 2e-05,
|
| 1470 |
+
"loss": 0.8038,
|
| 1471 |
+
"step": 98
|
| 1472 |
+
},
|
| 1473 |
+
{
|
| 1474 |
+
"epoch": 3.0625,
|
| 1475 |
+
"eval_loss": 0.7609645128250122,
|
| 1476 |
+
"eval_runtime": 196.8662,
|
| 1477 |
+
"eval_samples_per_second": 1.016,
|
| 1478 |
+
"eval_steps_per_second": 0.127,
|
| 1479 |
+
"step": 98
|
| 1480 |
+
},
|
| 1481 |
+
{
|
| 1482 |
+
"epoch": 3.09375,
|
| 1483 |
+
"grad_norm": 0.38849177572880716,
|
| 1484 |
+
"learning_rate": 2e-05,
|
| 1485 |
+
"loss": 0.8106,
|
| 1486 |
+
"step": 99
|
| 1487 |
+
},
|
| 1488 |
+
{
|
| 1489 |
+
"epoch": 3.09375,
|
| 1490 |
+
"eval_loss": 0.7598288059234619,
|
| 1491 |
+
"eval_runtime": 196.1846,
|
| 1492 |
+
"eval_samples_per_second": 1.019,
|
| 1493 |
+
"eval_steps_per_second": 0.127,
|
| 1494 |
+
"step": 99
|
| 1495 |
+
},
|
| 1496 |
+
{
|
| 1497 |
+
"epoch": 3.125,
|
| 1498 |
+
"grad_norm": 0.41885563528617265,
|
| 1499 |
+
"learning_rate": 2e-05,
|
| 1500 |
+
"loss": 0.808,
|
| 1501 |
+
"step": 100
|
| 1502 |
+
},
|
| 1503 |
+
{
|
| 1504 |
+
"epoch": 3.125,
|
| 1505 |
+
"eval_loss": 0.7587143778800964,
|
| 1506 |
+
"eval_runtime": 195.7296,
|
| 1507 |
+
"eval_samples_per_second": 1.022,
|
| 1508 |
+
"eval_steps_per_second": 0.128,
|
| 1509 |
+
"step": 100
|
| 1510 |
+
},
|
| 1511 |
+
{
|
| 1512 |
+
"epoch": 3.15625,
|
| 1513 |
+
"grad_norm": 0.4003909227323588,
|
| 1514 |
+
"learning_rate": 2e-05,
|
| 1515 |
+
"loss": 0.791,
|
| 1516 |
+
"step": 101
|
| 1517 |
+
},
|
| 1518 |
+
{
|
| 1519 |
+
"epoch": 3.15625,
|
| 1520 |
+
"eval_loss": 0.7578326463699341,
|
| 1521 |
+
"eval_runtime": 195.2831,
|
| 1522 |
+
"eval_samples_per_second": 1.024,
|
| 1523 |
+
"eval_steps_per_second": 0.128,
|
| 1524 |
+
"step": 101
|
| 1525 |
+
},
|
| 1526 |
+
{
|
| 1527 |
+
"epoch": 3.1875,
|
| 1528 |
+
"grad_norm": 0.4014550365826672,
|
| 1529 |
+
"learning_rate": 2e-05,
|
| 1530 |
+
"loss": 0.7402,
|
| 1531 |
+
"step": 102
|
| 1532 |
+
},
|
| 1533 |
+
{
|
| 1534 |
+
"epoch": 3.1875,
|
| 1535 |
+
"eval_loss": 0.7573958039283752,
|
| 1536 |
+
"eval_runtime": 189.5234,
|
| 1537 |
+
"eval_samples_per_second": 1.055,
|
| 1538 |
+
"eval_steps_per_second": 0.132,
|
| 1539 |
+
"step": 102
|
| 1540 |
+
},
|
| 1541 |
+
{
|
| 1542 |
+
"epoch": 3.21875,
|
| 1543 |
+
"grad_norm": 0.4018554316691014,
|
| 1544 |
+
"learning_rate": 2e-05,
|
| 1545 |
+
"loss": 0.8165,
|
| 1546 |
+
"step": 103
|
| 1547 |
+
},
|
| 1548 |
+
{
|
| 1549 |
+
"epoch": 3.21875,
|
| 1550 |
+
"eval_loss": 0.7571737766265869,
|
| 1551 |
+
"eval_runtime": 190.0146,
|
| 1552 |
+
"eval_samples_per_second": 1.053,
|
| 1553 |
+
"eval_steps_per_second": 0.132,
|
| 1554 |
+
"step": 103
|
| 1555 |
+
},
|
| 1556 |
+
{
|
| 1557 |
+
"epoch": 3.25,
|
| 1558 |
+
"grad_norm": 0.39691385018938347,
|
| 1559 |
+
"learning_rate": 2e-05,
|
| 1560 |
+
"loss": 0.7806,
|
| 1561 |
+
"step": 104
|
| 1562 |
+
},
|
| 1563 |
+
{
|
| 1564 |
+
"epoch": 3.25,
|
| 1565 |
+
"eval_loss": 0.7581367492675781,
|
| 1566 |
+
"eval_runtime": 190.1851,
|
| 1567 |
+
"eval_samples_per_second": 1.052,
|
| 1568 |
+
"eval_steps_per_second": 0.131,
|
| 1569 |
+
"step": 104
|
| 1570 |
+
},
|
| 1571 |
+
{
|
| 1572 |
+
"epoch": 3.28125,
|
| 1573 |
+
"grad_norm": 0.390373263306042,
|
| 1574 |
+
"learning_rate": 2e-05,
|
| 1575 |
+
"loss": 0.7454,
|
| 1576 |
+
"step": 105
|
| 1577 |
+
},
|
| 1578 |
+
{
|
| 1579 |
+
"epoch": 3.28125,
|
| 1580 |
+
"eval_loss": 0.7590533494949341,
|
| 1581 |
+
"eval_runtime": 190.1255,
|
| 1582 |
+
"eval_samples_per_second": 1.052,
|
| 1583 |
+
"eval_steps_per_second": 0.131,
|
| 1584 |
+
"step": 105
|
| 1585 |
+
},
|
| 1586 |
+
{
|
| 1587 |
+
"epoch": 3.3125,
|
| 1588 |
+
"grad_norm": 0.45093404603350434,
|
| 1589 |
+
"learning_rate": 2e-05,
|
| 1590 |
+
"loss": 0.8598,
|
| 1591 |
+
"step": 106
|
| 1592 |
+
},
|
| 1593 |
+
{
|
| 1594 |
+
"epoch": 3.3125,
|
| 1595 |
+
"eval_loss": 0.7584137916564941,
|
| 1596 |
+
"eval_runtime": 193.5956,
|
| 1597 |
+
"eval_samples_per_second": 1.033,
|
| 1598 |
+
"eval_steps_per_second": 0.129,
|
| 1599 |
+
"step": 106
|
| 1600 |
+
},
|
| 1601 |
+
{
|
| 1602 |
+
"epoch": 3.34375,
|
| 1603 |
+
"grad_norm": 0.4112664411035318,
|
| 1604 |
+
"learning_rate": 2e-05,
|
| 1605 |
+
"loss": 0.8612,
|
| 1606 |
+
"step": 107
|
| 1607 |
+
},
|
| 1608 |
+
{
|
| 1609 |
+
"epoch": 3.34375,
|
| 1610 |
+
"eval_loss": 0.757759690284729,
|
| 1611 |
+
"eval_runtime": 191.7864,
|
| 1612 |
+
"eval_samples_per_second": 1.043,
|
| 1613 |
+
"eval_steps_per_second": 0.13,
|
| 1614 |
+
"step": 107
|
| 1615 |
+
},
|
| 1616 |
+
{
|
| 1617 |
+
"epoch": 3.375,
|
| 1618 |
+
"grad_norm": 0.4158875890717671,
|
| 1619 |
+
"learning_rate": 2e-05,
|
| 1620 |
+
"loss": 0.7916,
|
| 1621 |
+
"step": 108
|
| 1622 |
+
},
|
| 1623 |
+
{
|
| 1624 |
+
"epoch": 3.375,
|
| 1625 |
+
"eval_loss": 0.756908655166626,
|
| 1626 |
+
"eval_runtime": 190.1833,
|
| 1627 |
+
"eval_samples_per_second": 1.052,
|
| 1628 |
+
"eval_steps_per_second": 0.131,
|
| 1629 |
+
"step": 108
|
| 1630 |
+
},
|
| 1631 |
+
{
|
| 1632 |
+
"epoch": 3.40625,
|
| 1633 |
+
"grad_norm": 0.4234644828447959,
|
| 1634 |
+
"learning_rate": 2e-05,
|
| 1635 |
+
"loss": 0.798,
|
| 1636 |
+
"step": 109
|
| 1637 |
+
},
|
| 1638 |
+
{
|
| 1639 |
+
"epoch": 3.40625,
|
| 1640 |
+
"eval_loss": 0.7559736371040344,
|
| 1641 |
+
"eval_runtime": 190.0515,
|
| 1642 |
+
"eval_samples_per_second": 1.052,
|
| 1643 |
+
"eval_steps_per_second": 0.132,
|
| 1644 |
+
"step": 109
|
| 1645 |
+
},
|
| 1646 |
+
{
|
| 1647 |
+
"epoch": 3.4375,
|
| 1648 |
+
"grad_norm": 0.480506693884699,
|
| 1649 |
+
"learning_rate": 2e-05,
|
| 1650 |
+
"loss": 0.7964,
|
| 1651 |
+
"step": 110
|
| 1652 |
+
},
|
| 1653 |
+
{
|
| 1654 |
+
"epoch": 3.4375,
|
| 1655 |
+
"eval_loss": 0.7547717094421387,
|
| 1656 |
+
"eval_runtime": 189.9579,
|
| 1657 |
+
"eval_samples_per_second": 1.053,
|
| 1658 |
+
"eval_steps_per_second": 0.132,
|
| 1659 |
+
"step": 110
|
| 1660 |
+
},
|
| 1661 |
+
{
|
| 1662 |
+
"epoch": 3.46875,
|
| 1663 |
+
"grad_norm": 0.4540166631930203,
|
| 1664 |
+
"learning_rate": 2e-05,
|
| 1665 |
+
"loss": 0.8172,
|
| 1666 |
+
"step": 111
|
| 1667 |
+
},
|
| 1668 |
+
{
|
| 1669 |
+
"epoch": 3.46875,
|
| 1670 |
+
"eval_loss": 0.7538467645645142,
|
| 1671 |
+
"eval_runtime": 194.2136,
|
| 1672 |
+
"eval_samples_per_second": 1.03,
|
| 1673 |
+
"eval_steps_per_second": 0.129,
|
| 1674 |
+
"step": 111
|
| 1675 |
+
},
|
| 1676 |
+
{
|
| 1677 |
+
"epoch": 3.5,
|
| 1678 |
+
"grad_norm": 0.46782505650509537,
|
| 1679 |
+
"learning_rate": 2e-05,
|
| 1680 |
+
"loss": 0.7826,
|
| 1681 |
+
"step": 112
|
| 1682 |
+
},
|
| 1683 |
+
{
|
| 1684 |
+
"epoch": 3.5,
|
| 1685 |
+
"eval_loss": 0.7537503838539124,
|
| 1686 |
+
"eval_runtime": 194.396,
|
| 1687 |
+
"eval_samples_per_second": 1.029,
|
| 1688 |
+
"eval_steps_per_second": 0.129,
|
| 1689 |
+
"step": 112
|
| 1690 |
+
},
|
| 1691 |
+
{
|
| 1692 |
+
"epoch": 3.53125,
|
| 1693 |
+
"grad_norm": 0.5212465522615498,
|
| 1694 |
+
"learning_rate": 2e-05,
|
| 1695 |
+
"loss": 0.7715,
|
| 1696 |
+
"step": 113
|
| 1697 |
+
},
|
| 1698 |
+
{
|
| 1699 |
+
"epoch": 3.53125,
|
| 1700 |
+
"eval_loss": 0.7527998089790344,
|
| 1701 |
+
"eval_runtime": 193.1123,
|
| 1702 |
+
"eval_samples_per_second": 1.036,
|
| 1703 |
+
"eval_steps_per_second": 0.129,
|
| 1704 |
+
"step": 113
|
| 1705 |
+
},
|
| 1706 |
+
{
|
| 1707 |
+
"epoch": 3.5625,
|
| 1708 |
+
"grad_norm": 0.4869709286188453,
|
| 1709 |
+
"learning_rate": 2e-05,
|
| 1710 |
+
"loss": 0.8007,
|
| 1711 |
+
"step": 114
|
| 1712 |
+
},
|
| 1713 |
+
{
|
| 1714 |
+
"epoch": 3.5625,
|
| 1715 |
+
"eval_loss": 0.7510444521903992,
|
| 1716 |
+
"eval_runtime": 193.3613,
|
| 1717 |
+
"eval_samples_per_second": 1.034,
|
| 1718 |
+
"eval_steps_per_second": 0.129,
|
| 1719 |
+
"step": 114
|
| 1720 |
+
},
|
| 1721 |
+
{
|
| 1722 |
+
"epoch": 3.59375,
|
| 1723 |
+
"grad_norm": 0.45218808224844204,
|
| 1724 |
+
"learning_rate": 2e-05,
|
| 1725 |
+
"loss": 0.7183,
|
| 1726 |
+
"step": 115
|
| 1727 |
+
},
|
| 1728 |
+
{
|
| 1729 |
+
"epoch": 3.59375,
|
| 1730 |
+
"eval_loss": 0.7502281069755554,
|
| 1731 |
+
"eval_runtime": 192.9756,
|
| 1732 |
+
"eval_samples_per_second": 1.036,
|
| 1733 |
+
"eval_steps_per_second": 0.13,
|
| 1734 |
+
"step": 115
|
| 1735 |
+
},
|
| 1736 |
+
{
|
| 1737 |
+
"epoch": 3.625,
|
| 1738 |
+
"grad_norm": 0.4563394511077237,
|
| 1739 |
+
"learning_rate": 2e-05,
|
| 1740 |
+
"loss": 0.7457,
|
| 1741 |
+
"step": 116
|
| 1742 |
+
},
|
| 1743 |
+
{
|
| 1744 |
+
"epoch": 3.625,
|
| 1745 |
+
"eval_loss": 0.75026535987854,
|
| 1746 |
+
"eval_runtime": 196.5666,
|
| 1747 |
+
"eval_samples_per_second": 1.017,
|
| 1748 |
+
"eval_steps_per_second": 0.127,
|
| 1749 |
+
"step": 116
|
| 1750 |
+
},
|
| 1751 |
+
{
|
| 1752 |
+
"epoch": 3.65625,
|
| 1753 |
+
"grad_norm": 0.4439962576867383,
|
| 1754 |
+
"learning_rate": 2e-05,
|
| 1755 |
+
"loss": 0.8091,
|
| 1756 |
+
"step": 117
|
| 1757 |
+
},
|
| 1758 |
+
{
|
| 1759 |
+
"epoch": 3.65625,
|
| 1760 |
+
"eval_loss": 0.7515027523040771,
|
| 1761 |
+
"eval_runtime": 196.6311,
|
| 1762 |
+
"eval_samples_per_second": 1.017,
|
| 1763 |
+
"eval_steps_per_second": 0.127,
|
| 1764 |
+
"step": 117
|
| 1765 |
+
},
|
| 1766 |
+
{
|
| 1767 |
+
"epoch": 3.6875,
|
| 1768 |
+
"grad_norm": 0.44782841352683267,
|
| 1769 |
+
"learning_rate": 2e-05,
|
| 1770 |
+
"loss": 0.7636,
|
| 1771 |
+
"step": 118
|
| 1772 |
+
},
|
| 1773 |
+
{
|
| 1774 |
+
"epoch": 3.6875,
|
| 1775 |
+
"eval_loss": 0.7535241842269897,
|
| 1776 |
+
"eval_runtime": 196.4625,
|
| 1777 |
+
"eval_samples_per_second": 1.018,
|
| 1778 |
+
"eval_steps_per_second": 0.127,
|
| 1779 |
+
"step": 118
|
| 1780 |
+
},
|
| 1781 |
+
{
|
| 1782 |
+
"epoch": 3.71875,
|
| 1783 |
+
"grad_norm": 0.5651906616770451,
|
| 1784 |
+
"learning_rate": 2e-05,
|
| 1785 |
+
"loss": 0.8106,
|
| 1786 |
+
"step": 119
|
| 1787 |
+
},
|
| 1788 |
+
{
|
| 1789 |
+
"epoch": 3.71875,
|
| 1790 |
+
"eval_loss": 0.7535383701324463,
|
| 1791 |
+
"eval_runtime": 196.0073,
|
| 1792 |
+
"eval_samples_per_second": 1.02,
|
| 1793 |
+
"eval_steps_per_second": 0.128,
|
| 1794 |
+
"step": 119
|
| 1795 |
+
},
|
| 1796 |
+
{
|
| 1797 |
+
"epoch": 3.75,
|
| 1798 |
+
"grad_norm": 0.44715700121507296,
|
| 1799 |
+
"learning_rate": 2e-05,
|
| 1800 |
+
"loss": 0.7991,
|
| 1801 |
+
"step": 120
|
| 1802 |
+
},
|
| 1803 |
+
{
|
| 1804 |
+
"epoch": 3.75,
|
| 1805 |
+
"eval_loss": 0.75335294008255,
|
| 1806 |
+
"eval_runtime": 195.9837,
|
| 1807 |
+
"eval_samples_per_second": 1.02,
|
| 1808 |
+
"eval_steps_per_second": 0.128,
|
| 1809 |
+
"step": 120
|
| 1810 |
+
},
|
| 1811 |
+
{
|
| 1812 |
+
"epoch": 3.75,
|
| 1813 |
+
"step": 120,
|
| 1814 |
+
"total_flos": 266165567225856.0,
|
| 1815 |
+
"train_loss": 0.0,
|
| 1816 |
+
"train_runtime": 1.4333,
|
| 1817 |
+
"train_samples_per_second": 2790.753,
|
| 1818 |
+
"train_steps_per_second": 44.652
|
| 1819 |
+
}
|
| 1820 |
+
],
|
| 1821 |
+
"logging_steps": 1.0,
|
| 1822 |
+
"max_steps": 64,
|
| 1823 |
+
"num_input_tokens_seen": 0,
|
| 1824 |
+
"num_train_epochs": 4,
|
| 1825 |
+
"save_steps": 5,
|
| 1826 |
+
"stateful_callbacks": {
|
| 1827 |
+
"TrainerControl": {
|
| 1828 |
+
"args": {
|
| 1829 |
+
"should_epoch_stop": false,
|
| 1830 |
+
"should_evaluate": false,
|
| 1831 |
+
"should_log": false,
|
| 1832 |
+
"should_save": true,
|
| 1833 |
+
"should_training_stop": false
|
| 1834 |
+
},
|
| 1835 |
+
"attributes": {}
|
| 1836 |
+
}
|
| 1837 |
+
},
|
| 1838 |
+
"total_flos": 266165567225856.0,
|
| 1839 |
+
"train_batch_size": 8,
|
| 1840 |
+
"trial_name": null,
|
| 1841 |
+
"trial_params": null
|
| 1842 |
+
}
|