Upload folder using huggingface_hub
Browse files- README.md +202 -0
- adapter_config.json +380 -0
- adapter_model.safetensors +3 -0
- latest +1 -0
- qwen.tiktoken +0 -0
- rng_state_0.pth +3 -0
- rng_state_1.pth +3 -0
- rng_state_2.pth +3 -0
- rng_state_3.pth +3 -0
- scheduler.pt +3 -0
- special_tokens_map.json +3 -0
- tokenizer_config.json +14 -0
- trainer_state.json +1993 -0
- training_args.bin +3 -0
- zero_to_fp32.py +587 -0
README.md
ADDED
|
@@ -0,0 +1,202 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
library_name: peft
|
| 3 |
+
base_model: Qwen/Qwen-VL-Chat
|
| 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 0.10.0
|
adapter_config.json
ADDED
|
@@ -0,0 +1,380 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"alpha_pattern": {},
|
| 3 |
+
"auto_mapping": null,
|
| 4 |
+
"base_model_name_or_path": "Qwen/Qwen-VL-Chat",
|
| 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": 64,
|
| 20 |
+
"rank_pattern": {},
|
| 21 |
+
"revision": null,
|
| 22 |
+
"target_modules": [
|
| 23 |
+
"transformer.h.17.attn.c_proj",
|
| 24 |
+
"transformer.h.20.mlp.c_proj",
|
| 25 |
+
"transformer.visual.transformer.resblocks.1.attn.in_proj",
|
| 26 |
+
"transformer.h.3.attn.c_attn",
|
| 27 |
+
"transformer.visual.transformer.resblocks.12.attn.in_proj",
|
| 28 |
+
"transformer.visual.transformer.resblocks.47.attn.in_proj",
|
| 29 |
+
"transformer.h.28.mlp.w2",
|
| 30 |
+
"transformer.h.6.mlp.w2",
|
| 31 |
+
"transformer.h.13.mlp.w1",
|
| 32 |
+
"transformer.visual.transformer.resblocks.39.attn.out_proj",
|
| 33 |
+
"transformer.h.2.mlp.c_proj",
|
| 34 |
+
"transformer.visual.transformer.resblocks.3.attn.out_proj",
|
| 35 |
+
"transformer.visual.transformer.resblocks.0.attn.out_proj",
|
| 36 |
+
"transformer.h.4.attn.c_proj",
|
| 37 |
+
"transformer.h.22.mlp.c_proj",
|
| 38 |
+
"transformer.visual.transformer.resblocks.12.attn.out_proj",
|
| 39 |
+
"transformer.visual.transformer.resblocks.10.mlp.c_fc",
|
| 40 |
+
"transformer.visual.transformer.resblocks.43.attn.in_proj",
|
| 41 |
+
"transformer.visual.transformer.resblocks.0.attn.in_proj",
|
| 42 |
+
"transformer.visual.transformer.resblocks.26.mlp.c_fc",
|
| 43 |
+
"transformer.visual.transformer.resblocks.11.mlp.c_proj",
|
| 44 |
+
"transformer.h.0.attn.c_attn",
|
| 45 |
+
"transformer.h.19.mlp.w2",
|
| 46 |
+
"transformer.visual.transformer.resblocks.37.mlp.c_proj",
|
| 47 |
+
"transformer.visual.transformer.resblocks.40.mlp.c_proj",
|
| 48 |
+
"transformer.h.31.mlp.c_proj",
|
| 49 |
+
"transformer.visual.transformer.resblocks.32.mlp.c_fc",
|
| 50 |
+
"transformer.h.18.mlp.w1",
|
| 51 |
+
"transformer.h.23.mlp.w2",
|
| 52 |
+
"transformer.visual.transformer.resblocks.6.attn.out_proj",
|
| 53 |
+
"transformer.visual.transformer.resblocks.17.attn.in_proj",
|
| 54 |
+
"transformer.visual.transformer.resblocks.27.attn.out_proj",
|
| 55 |
+
"transformer.h.12.mlp.w2",
|
| 56 |
+
"transformer.h.23.mlp.c_proj",
|
| 57 |
+
"transformer.visual.transformer.resblocks.29.attn.in_proj",
|
| 58 |
+
"transformer.h.10.mlp.w1",
|
| 59 |
+
"transformer.visual.transformer.resblocks.18.attn.out_proj",
|
| 60 |
+
"transformer.visual.transformer.resblocks.4.attn.out_proj",
|
| 61 |
+
"transformer.visual.transformer.resblocks.8.mlp.c_fc",
|
| 62 |
+
"transformer.h.9.mlp.w1",
|
| 63 |
+
"transformer.visual.transformer.resblocks.38.mlp.c_proj",
|
| 64 |
+
"transformer.visual.transformer.resblocks.6.attn.in_proj",
|
| 65 |
+
"transformer.visual.transformer.resblocks.14.mlp.c_proj",
|
| 66 |
+
"transformer.visual.transformer.resblocks.22.attn.in_proj",
|
| 67 |
+
"transformer.visual.transformer.resblocks.25.mlp.c_proj",
|
| 68 |
+
"transformer.visual.transformer.resblocks.23.attn.out_proj",
|
| 69 |
+
"transformer.visual.transformer.resblocks.3.mlp.c_proj",
|
| 70 |
+
"transformer.visual.transformer.resblocks.41.mlp.c_proj",
|
| 71 |
+
"transformer.h.24.attn.c_proj",
|
| 72 |
+
"transformer.visual.transformer.resblocks.7.mlp.c_fc",
|
| 73 |
+
"transformer.visual.transformer.resblocks.38.mlp.c_fc",
|
| 74 |
+
"transformer.h.10.attn.c_attn",
|
| 75 |
+
"transformer.h.26.attn.c_attn",
|
| 76 |
+
"transformer.visual.transformer.resblocks.5.attn.in_proj",
|
| 77 |
+
"transformer.visual.transformer.resblocks.2.attn.out_proj",
|
| 78 |
+
"transformer.h.7.attn.c_proj",
|
| 79 |
+
"transformer.h.24.mlp.c_proj",
|
| 80 |
+
"transformer.visual.transformer.resblocks.34.mlp.c_proj",
|
| 81 |
+
"transformer.visual.transformer.resblocks.2.mlp.c_proj",
|
| 82 |
+
"transformer.h.12.mlp.c_proj",
|
| 83 |
+
"transformer.visual.transformer.resblocks.14.attn.out_proj",
|
| 84 |
+
"transformer.h.18.attn.c_attn",
|
| 85 |
+
"transformer.h.23.attn.c_proj",
|
| 86 |
+
"transformer.h.27.mlp.c_proj",
|
| 87 |
+
"transformer.visual.transformer.resblocks.26.mlp.c_proj",
|
| 88 |
+
"transformer.h.3.mlp.w1",
|
| 89 |
+
"transformer.h.2.mlp.w2",
|
| 90 |
+
"transformer.visual.transformer.resblocks.45.mlp.c_proj",
|
| 91 |
+
"transformer.visual.transformer.resblocks.25.mlp.c_fc",
|
| 92 |
+
"transformer.visual.transformer.resblocks.45.attn.out_proj",
|
| 93 |
+
"transformer.h.25.mlp.w1",
|
| 94 |
+
"transformer.visual.transformer.resblocks.15.mlp.c_proj",
|
| 95 |
+
"transformer.visual.transformer.resblocks.24.attn.in_proj",
|
| 96 |
+
"transformer.h.1.attn.c_proj",
|
| 97 |
+
"transformer.h.1.attn.c_attn",
|
| 98 |
+
"transformer.visual.transformer.resblocks.4.mlp.c_fc",
|
| 99 |
+
"transformer.visual.transformer.resblocks.9.mlp.c_fc",
|
| 100 |
+
"transformer.h.13.attn.c_attn",
|
| 101 |
+
"transformer.visual.transformer.resblocks.40.attn.out_proj",
|
| 102 |
+
"transformer.h.7.mlp.w2",
|
| 103 |
+
"transformer.h.9.attn.c_proj",
|
| 104 |
+
"transformer.h.15.attn.c_attn",
|
| 105 |
+
"transformer.visual.transformer.resblocks.0.mlp.c_fc",
|
| 106 |
+
"transformer.h.27.attn.c_attn",
|
| 107 |
+
"transformer.h.15.mlp.c_proj",
|
| 108 |
+
"transformer.h.21.mlp.w2",
|
| 109 |
+
"transformer.h.28.attn.c_proj",
|
| 110 |
+
"transformer.visual.transformer.resblocks.42.mlp.c_proj",
|
| 111 |
+
"transformer.visual.transformer.resblocks.16.attn.out_proj",
|
| 112 |
+
"transformer.h.9.mlp.w2",
|
| 113 |
+
"transformer.visual.transformer.resblocks.9.attn.in_proj",
|
| 114 |
+
"transformer.visual.transformer.resblocks.28.mlp.c_proj",
|
| 115 |
+
"transformer.visual.transformer.resblocks.6.mlp.c_proj",
|
| 116 |
+
"transformer.h.11.mlp.w1",
|
| 117 |
+
"transformer.visual.transformer.resblocks.18.attn.in_proj",
|
| 118 |
+
"transformer.h.10.attn.c_proj",
|
| 119 |
+
"transformer.visual.transformer.resblocks.42.mlp.c_fc",
|
| 120 |
+
"transformer.h.31.attn.c_attn",
|
| 121 |
+
"transformer.visual.transformer.resblocks.23.mlp.c_fc",
|
| 122 |
+
"transformer.visual.transformer.resblocks.21.attn.in_proj",
|
| 123 |
+
"transformer.h.24.mlp.w1",
|
| 124 |
+
"transformer.visual.transformer.resblocks.35.mlp.c_fc",
|
| 125 |
+
"transformer.visual.transformer.resblocks.7.mlp.c_proj",
|
| 126 |
+
"transformer.h.8.mlp.c_proj",
|
| 127 |
+
"transformer.visual.transformer.resblocks.12.mlp.c_fc",
|
| 128 |
+
"transformer.visual.transformer.resblocks.7.attn.out_proj",
|
| 129 |
+
"transformer.h.22.mlp.w2",
|
| 130 |
+
"transformer.h.29.mlp.w2",
|
| 131 |
+
"transformer.h.0.mlp.c_proj",
|
| 132 |
+
"transformer.visual.transformer.resblocks.38.attn.in_proj",
|
| 133 |
+
"transformer.h.8.mlp.w1",
|
| 134 |
+
"transformer.h.0.mlp.w1",
|
| 135 |
+
"transformer.h.26.mlp.w2",
|
| 136 |
+
"transformer.h.25.attn.c_proj",
|
| 137 |
+
"transformer.h.27.mlp.w1",
|
| 138 |
+
"transformer.visual.transformer.resblocks.21.attn.out_proj",
|
| 139 |
+
"transformer.visual.transformer.resblocks.44.attn.in_proj",
|
| 140 |
+
"transformer.visual.transformer.resblocks.43.attn.out_proj",
|
| 141 |
+
"transformer.h.29.attn.c_attn",
|
| 142 |
+
"transformer.h.24.attn.c_attn",
|
| 143 |
+
"transformer.visual.transformer.resblocks.17.attn.out_proj",
|
| 144 |
+
"transformer.h.2.attn.c_proj",
|
| 145 |
+
"transformer.visual.transformer.resblocks.15.mlp.c_fc",
|
| 146 |
+
"transformer.visual.transformer.resblocks.11.attn.in_proj",
|
| 147 |
+
"transformer.visual.transformer.resblocks.17.mlp.c_proj",
|
| 148 |
+
"transformer.h.11.mlp.c_proj",
|
| 149 |
+
"transformer.visual.transformer.resblocks.32.mlp.c_proj",
|
| 150 |
+
"transformer.visual.transformer.resblocks.6.mlp.c_fc",
|
| 151 |
+
"transformer.visual.transformer.resblocks.41.mlp.c_fc",
|
| 152 |
+
"transformer.visual.transformer.resblocks.5.mlp.c_fc",
|
| 153 |
+
"transformer.visual.transformer.resblocks.18.mlp.c_fc",
|
| 154 |
+
"transformer.visual.transformer.resblocks.24.mlp.c_proj",
|
| 155 |
+
"transformer.visual.transformer.resblocks.32.attn.out_proj",
|
| 156 |
+
"transformer.h.1.mlp.w2",
|
| 157 |
+
"transformer.h.21.mlp.c_proj",
|
| 158 |
+
"transformer.h.23.attn.c_attn",
|
| 159 |
+
"transformer.visual.transformer.resblocks.34.attn.out_proj",
|
| 160 |
+
"transformer.h.14.attn.c_attn",
|
| 161 |
+
"transformer.visual.transformer.resblocks.2.mlp.c_fc",
|
| 162 |
+
"transformer.visual.transformer.resblocks.31.attn.out_proj",
|
| 163 |
+
"transformer.visual.transformer.resblocks.30.mlp.c_proj",
|
| 164 |
+
"transformer.visual.transformer.resblocks.11.mlp.c_fc",
|
| 165 |
+
"transformer.visual.transformer.resblocks.31.attn.in_proj",
|
| 166 |
+
"transformer.visual.transformer.resblocks.39.mlp.c_proj",
|
| 167 |
+
"transformer.h.9.mlp.c_proj",
|
| 168 |
+
"transformer.visual.transformer.resblocks.20.attn.out_proj",
|
| 169 |
+
"transformer.h.18.mlp.c_proj",
|
| 170 |
+
"transformer.h.19.mlp.w1",
|
| 171 |
+
"transformer.h.9.attn.c_attn",
|
| 172 |
+
"transformer.visual.transformer.resblocks.36.attn.out_proj",
|
| 173 |
+
"transformer.visual.transformer.resblocks.7.attn.in_proj",
|
| 174 |
+
"transformer.visual.transformer.resblocks.30.attn.in_proj",
|
| 175 |
+
"transformer.visual.transformer.resblocks.47.attn.out_proj",
|
| 176 |
+
"transformer.visual.transformer.resblocks.0.mlp.c_proj",
|
| 177 |
+
"transformer.visual.transformer.resblocks.15.attn.in_proj",
|
| 178 |
+
"transformer.visual.transformer.resblocks.29.attn.out_proj",
|
| 179 |
+
"transformer.visual.transformer.resblocks.41.attn.in_proj",
|
| 180 |
+
"transformer.visual.transformer.resblocks.4.attn.in_proj",
|
| 181 |
+
"transformer.h.25.attn.c_attn",
|
| 182 |
+
"transformer.visual.transformer.resblocks.12.mlp.c_proj",
|
| 183 |
+
"transformer.h.16.mlp.w1",
|
| 184 |
+
"transformer.h.28.mlp.c_proj",
|
| 185 |
+
"transformer.visual.transformer.resblocks.27.attn.in_proj",
|
| 186 |
+
"transformer.visual.transformer.resblocks.13.mlp.c_proj",
|
| 187 |
+
"transformer.visual.transformer.resblocks.33.attn.in_proj",
|
| 188 |
+
"transformer.visual.transformer.resblocks.45.mlp.c_fc",
|
| 189 |
+
"transformer.visual.transformer.resblocks.46.mlp.c_proj",
|
| 190 |
+
"transformer.h.30.mlp.w1",
|
| 191 |
+
"transformer.visual.transformer.resblocks.43.mlp.c_fc",
|
| 192 |
+
"transformer.h.15.mlp.w1",
|
| 193 |
+
"transformer.h.16.attn.c_proj",
|
| 194 |
+
"transformer.h.20.mlp.w1",
|
| 195 |
+
"transformer.visual.transformer.resblocks.21.mlp.c_fc",
|
| 196 |
+
"transformer.visual.transformer.resblocks.10.mlp.c_proj",
|
| 197 |
+
"transformer.h.10.mlp.c_proj",
|
| 198 |
+
"transformer.visual.transformer.resblocks.35.attn.in_proj",
|
| 199 |
+
"transformer.h.13.mlp.w2",
|
| 200 |
+
"transformer.visual.transformer.resblocks.8.attn.out_proj",
|
| 201 |
+
"transformer.visual.transformer.resblocks.20.mlp.c_proj",
|
| 202 |
+
"transformer.h.22.attn.c_proj",
|
| 203 |
+
"transformer.h.6.mlp.w1",
|
| 204 |
+
"transformer.h.18.mlp.w2",
|
| 205 |
+
"transformer.h.4.mlp.c_proj",
|
| 206 |
+
"transformer.h.3.mlp.c_proj",
|
| 207 |
+
"transformer.visual.transformer.resblocks.42.attn.out_proj",
|
| 208 |
+
"transformer.visual.transformer.resblocks.36.attn.in_proj",
|
| 209 |
+
"transformer.visual.transformer.resblocks.17.mlp.c_fc",
|
| 210 |
+
"transformer.visual.transformer.resblocks.43.mlp.c_proj",
|
| 211 |
+
"transformer.visual.transformer.resblocks.37.attn.in_proj",
|
| 212 |
+
"transformer.visual.transformer.resblocks.1.attn.out_proj",
|
| 213 |
+
"transformer.visual.transformer.resblocks.22.mlp.c_fc",
|
| 214 |
+
"transformer.h.22.mlp.w1",
|
| 215 |
+
"transformer.visual.transformer.resblocks.44.mlp.c_fc",
|
| 216 |
+
"transformer.visual.transformer.resblocks.37.attn.out_proj",
|
| 217 |
+
"transformer.visual.transformer.resblocks.34.mlp.c_fc",
|
| 218 |
+
"transformer.visual.transformer.resblocks.29.mlp.c_fc",
|
| 219 |
+
"transformer.h.18.attn.c_proj",
|
| 220 |
+
"transformer.visual.transformer.resblocks.38.attn.out_proj",
|
| 221 |
+
"transformer.h.5.attn.c_attn",
|
| 222 |
+
"transformer.visual.transformer.resblocks.19.mlp.c_fc",
|
| 223 |
+
"transformer.visual.transformer.resblocks.15.attn.out_proj",
|
| 224 |
+
"transformer.visual.transformer.resblocks.37.mlp.c_fc",
|
| 225 |
+
"transformer.h.5.attn.c_proj",
|
| 226 |
+
"transformer.h.7.attn.c_attn",
|
| 227 |
+
"transformer.visual.transformer.resblocks.28.attn.out_proj",
|
| 228 |
+
"transformer.visual.transformer.resblocks.31.mlp.c_proj",
|
| 229 |
+
"transformer.h.29.mlp.c_proj",
|
| 230 |
+
"transformer.visual.transformer.resblocks.45.attn.in_proj",
|
| 231 |
+
"transformer.visual.transformer.resblocks.27.mlp.c_proj",
|
| 232 |
+
"transformer.visual.transformer.resblocks.10.attn.out_proj",
|
| 233 |
+
"transformer.visual.transformer.resblocks.40.attn.in_proj",
|
| 234 |
+
"transformer.h.23.mlp.w1",
|
| 235 |
+
"transformer.visual.transformer.resblocks.28.attn.in_proj",
|
| 236 |
+
"transformer.h.12.attn.c_proj",
|
| 237 |
+
"transformer.h.16.mlp.w2",
|
| 238 |
+
"transformer.h.27.mlp.w2",
|
| 239 |
+
"transformer.visual.transformer.resblocks.22.mlp.c_proj",
|
| 240 |
+
"transformer.visual.transformer.resblocks.9.mlp.c_proj",
|
| 241 |
+
"transformer.visual.transformer.resblocks.47.mlp.c_proj",
|
| 242 |
+
"transformer.h.26.attn.c_proj",
|
| 243 |
+
"transformer.visual.transformer.resblocks.40.mlp.c_fc",
|
| 244 |
+
"transformer.h.8.mlp.w2",
|
| 245 |
+
"transformer.visual.transformer.resblocks.27.mlp.c_fc",
|
| 246 |
+
"transformer.h.17.mlp.w1",
|
| 247 |
+
"transformer.h.31.mlp.w2",
|
| 248 |
+
"transformer.visual.transformer.resblocks.11.attn.out_proj",
|
| 249 |
+
"transformer.h.28.mlp.w1",
|
| 250 |
+
"transformer.visual.transformer.resblocks.10.attn.in_proj",
|
| 251 |
+
"transformer.h.12.mlp.w1",
|
| 252 |
+
"transformer.h.30.mlp.w2",
|
| 253 |
+
"transformer.visual.transformer.resblocks.13.attn.in_proj",
|
| 254 |
+
"transformer.h.6.attn.c_attn",
|
| 255 |
+
"transformer.h.5.mlp.c_proj",
|
| 256 |
+
"transformer.h.6.mlp.c_proj",
|
| 257 |
+
"transformer.h.22.attn.c_attn",
|
| 258 |
+
"transformer.h.13.attn.c_proj",
|
| 259 |
+
"transformer.visual.transformer.resblocks.46.mlp.c_fc",
|
| 260 |
+
"transformer.visual.transformer.resblocks.41.attn.out_proj",
|
| 261 |
+
"transformer.visual.transformer.resblocks.30.mlp.c_fc",
|
| 262 |
+
"transformer.h.17.mlp.c_proj",
|
| 263 |
+
"transformer.visual.transformer.resblocks.5.attn.out_proj",
|
| 264 |
+
"transformer.h.4.mlp.w2",
|
| 265 |
+
"transformer.visual.transformer.resblocks.1.mlp.c_proj",
|
| 266 |
+
"transformer.h.11.mlp.w2",
|
| 267 |
+
"transformer.h.19.attn.c_attn",
|
| 268 |
+
"transformer.h.14.mlp.w1",
|
| 269 |
+
"transformer.visual.transformer.resblocks.44.attn.out_proj",
|
| 270 |
+
"transformer.visual.transformer.resblocks.14.mlp.c_fc",
|
| 271 |
+
"transformer.h.21.attn.c_attn",
|
| 272 |
+
"transformer.visual.transformer.resblocks.36.mlp.c_proj",
|
| 273 |
+
"transformer.h.2.mlp.w1",
|
| 274 |
+
"transformer.h.14.attn.c_proj",
|
| 275 |
+
"transformer.visual.transformer.resblocks.46.attn.in_proj",
|
| 276 |
+
"transformer.h.6.attn.c_proj",
|
| 277 |
+
"transformer.h.0.mlp.w2",
|
| 278 |
+
"transformer.h.5.mlp.w1",
|
| 279 |
+
"transformer.h.30.attn.c_proj",
|
| 280 |
+
"transformer.h.24.mlp.w2",
|
| 281 |
+
"transformer.h.0.attn.c_proj",
|
| 282 |
+
"transformer.visual.transformer.resblocks.4.mlp.c_proj",
|
| 283 |
+
"transformer.visual.transformer.resblocks.22.attn.out_proj",
|
| 284 |
+
"transformer.h.10.mlp.w2",
|
| 285 |
+
"transformer.h.17.mlp.w2",
|
| 286 |
+
"transformer.visual.transformer.resblocks.23.attn.in_proj",
|
| 287 |
+
"transformer.visual.transformer.resblocks.36.mlp.c_fc",
|
| 288 |
+
"transformer.h.20.mlp.w2",
|
| 289 |
+
"transformer.visual.transformer.resblocks.9.attn.out_proj",
|
| 290 |
+
"transformer.h.29.mlp.w1",
|
| 291 |
+
"transformer.visual.transformer.resblocks.20.attn.in_proj",
|
| 292 |
+
"transformer.visual.transformer.resblocks.20.mlp.c_fc",
|
| 293 |
+
"transformer.h.15.attn.c_proj",
|
| 294 |
+
"transformer.h.3.mlp.w2",
|
| 295 |
+
"transformer.h.30.attn.c_attn",
|
| 296 |
+
"transformer.visual.transformer.resblocks.47.mlp.c_fc",
|
| 297 |
+
"transformer.visual.transformer.resblocks.16.mlp.c_proj",
|
| 298 |
+
"transformer.visual.transformer.resblocks.33.mlp.c_fc",
|
| 299 |
+
"transformer.visual.transformer.resblocks.39.mlp.c_fc",
|
| 300 |
+
"transformer.h.20.attn.c_attn",
|
| 301 |
+
"transformer.h.19.mlp.c_proj",
|
| 302 |
+
"transformer.visual.transformer.resblocks.46.attn.out_proj",
|
| 303 |
+
"transformer.visual.transformer.resblocks.29.mlp.c_proj",
|
| 304 |
+
"transformer.visual.transformer.resblocks.19.attn.out_proj",
|
| 305 |
+
"transformer.visual.transformer.resblocks.26.attn.in_proj",
|
| 306 |
+
"transformer.visual.transformer.resblocks.16.mlp.c_fc",
|
| 307 |
+
"transformer.h.11.attn.c_proj",
|
| 308 |
+
"transformer.h.12.attn.c_attn",
|
| 309 |
+
"transformer.visual.conv1",
|
| 310 |
+
"transformer.visual.transformer.resblocks.35.attn.out_proj",
|
| 311 |
+
"transformer.h.25.mlp.c_proj",
|
| 312 |
+
"transformer.visual.transformer.resblocks.14.attn.in_proj",
|
| 313 |
+
"transformer.h.26.mlp.w1",
|
| 314 |
+
"transformer.visual.transformer.resblocks.1.mlp.c_fc",
|
| 315 |
+
"transformer.h.7.mlp.c_proj",
|
| 316 |
+
"transformer.h.29.attn.c_proj",
|
| 317 |
+
"transformer.h.1.mlp.c_proj",
|
| 318 |
+
"transformer.visual.transformer.resblocks.33.mlp.c_proj",
|
| 319 |
+
"transformer.h.14.mlp.c_proj",
|
| 320 |
+
"transformer.h.3.attn.c_proj",
|
| 321 |
+
"transformer.h.25.mlp.w2",
|
| 322 |
+
"transformer.h.20.attn.c_proj",
|
| 323 |
+
"transformer.h.16.mlp.c_proj",
|
| 324 |
+
"transformer.visual.transformer.resblocks.3.attn.in_proj",
|
| 325 |
+
"transformer.h.17.attn.c_attn",
|
| 326 |
+
"transformer.h.14.mlp.w2",
|
| 327 |
+
"transformer.visual.transformer.resblocks.2.attn.in_proj",
|
| 328 |
+
"transformer.visual.transformer.resblocks.5.mlp.c_proj",
|
| 329 |
+
"transformer.visual.transformer.resblocks.3.mlp.c_fc",
|
| 330 |
+
"transformer.visual.transformer.resblocks.33.attn.out_proj",
|
| 331 |
+
"transformer.h.15.mlp.w2",
|
| 332 |
+
"transformer.h.4.attn.c_attn",
|
| 333 |
+
"transformer.h.31.mlp.w1",
|
| 334 |
+
"transformer.h.11.attn.c_attn",
|
| 335 |
+
"transformer.visual.transformer.resblocks.23.mlp.c_proj",
|
| 336 |
+
"transformer.h.7.mlp.w1",
|
| 337 |
+
"transformer.visual.transformer.resblocks.34.attn.in_proj",
|
| 338 |
+
"transformer.h.1.mlp.w1",
|
| 339 |
+
"transformer.visual.transformer.resblocks.28.mlp.c_fc",
|
| 340 |
+
"transformer.h.21.attn.c_proj",
|
| 341 |
+
"transformer.h.30.mlp.c_proj",
|
| 342 |
+
"transformer.h.21.mlp.w1",
|
| 343 |
+
"transformer.visual.transformer.resblocks.30.attn.out_proj",
|
| 344 |
+
"transformer.visual.transformer.resblocks.42.attn.in_proj",
|
| 345 |
+
"transformer.visual.transformer.resblocks.25.attn.out_proj",
|
| 346 |
+
"transformer.visual.transformer.resblocks.19.mlp.c_proj",
|
| 347 |
+
"transformer.visual.transformer.resblocks.39.attn.in_proj",
|
| 348 |
+
"transformer.visual.transformer.resblocks.19.attn.in_proj",
|
| 349 |
+
"transformer.visual.transformer.resblocks.13.mlp.c_fc",
|
| 350 |
+
"transformer.h.13.mlp.c_proj",
|
| 351 |
+
"transformer.visual.transformer.resblocks.25.attn.in_proj",
|
| 352 |
+
"transformer.visual.transformer.resblocks.31.mlp.c_fc",
|
| 353 |
+
"transformer.visual.transformer.resblocks.24.attn.out_proj",
|
| 354 |
+
"transformer.visual.transformer.resblocks.24.mlp.c_fc",
|
| 355 |
+
"transformer.h.4.mlp.w1",
|
| 356 |
+
"transformer.h.8.attn.c_attn",
|
| 357 |
+
"transformer.visual.transformer.resblocks.21.mlp.c_proj",
|
| 358 |
+
"transformer.visual.transformer.resblocks.44.mlp.c_proj",
|
| 359 |
+
"transformer.h.28.attn.c_attn",
|
| 360 |
+
"transformer.visual.transformer.resblocks.18.mlp.c_proj",
|
| 361 |
+
"transformer.visual.transformer.resblocks.32.attn.in_proj",
|
| 362 |
+
"transformer.h.19.attn.c_proj",
|
| 363 |
+
"transformer.h.2.attn.c_attn",
|
| 364 |
+
"transformer.visual.transformer.resblocks.35.mlp.c_proj",
|
| 365 |
+
"transformer.h.26.mlp.c_proj",
|
| 366 |
+
"transformer.h.8.attn.c_proj",
|
| 367 |
+
"transformer.h.27.attn.c_proj",
|
| 368 |
+
"transformer.visual.transformer.resblocks.13.attn.out_proj",
|
| 369 |
+
"transformer.h.16.attn.c_attn",
|
| 370 |
+
"transformer.visual.transformer.resblocks.16.attn.in_proj",
|
| 371 |
+
"transformer.visual.transformer.resblocks.8.attn.in_proj",
|
| 372 |
+
"transformer.visual.transformer.resblocks.26.attn.out_proj",
|
| 373 |
+
"transformer.h.31.attn.c_proj",
|
| 374 |
+
"transformer.h.5.mlp.w2",
|
| 375 |
+
"transformer.visual.transformer.resblocks.8.mlp.c_proj"
|
| 376 |
+
],
|
| 377 |
+
"task_type": "CAUSAL_LM",
|
| 378 |
+
"use_dora": false,
|
| 379 |
+
"use_rslora": false
|
| 380 |
+
}
|
adapter_model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a535f8faf62ff0f1b19a0a003566aab062d0b4ed66d7d12b1a4b2541b92a0ae8
|
| 3 |
+
size 469105640
|
latest
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
global_step2800
|
qwen.tiktoken
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
rng_state_0.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:68f840911fbc7051ece0e1c060c40b82d45b8edf23e2a3872c8e02ecebf67a26
|
| 3 |
+
size 14960
|
rng_state_1.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:68b321aa3d71576dec15af0375b3693729175e49283eea87928462724dfcf745
|
| 3 |
+
size 14960
|
rng_state_2.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:73b44b88d1bba3405d2bd906b256777b38a454b132494d9962cf11750a697099
|
| 3 |
+
size 14960
|
rng_state_3.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:916d8bb82a2aad59d5a38354a8f69f886f5bfbe84e1431e6bea937f401cec5dd
|
| 3 |
+
size 14960
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b28152862c82ab8bed7327b476edab940c3e9927d66ceaa0dba25e1157b3ff89
|
| 3 |
+
size 1064
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"pad_token": "<|endoftext|>"
|
| 3 |
+
}
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,14 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"added_tokens_decoder": {},
|
| 3 |
+
"auto_map": {
|
| 4 |
+
"AutoTokenizer": [
|
| 5 |
+
"Qwen/Qwen-VL-Chat--tokenization_qwen.QWenTokenizer",
|
| 6 |
+
null
|
| 7 |
+
]
|
| 8 |
+
},
|
| 9 |
+
"clean_up_tokenization_spaces": true,
|
| 10 |
+
"model_max_length": 1280,
|
| 11 |
+
"pad_token": "<|endoftext|>",
|
| 12 |
+
"padding_side": "right",
|
| 13 |
+
"tokenizer_class": "QWenTokenizer"
|
| 14 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1993 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.4711425206124853,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 2800,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.0016826518593303045,
|
| 13 |
+
"grad_norm": 5.367858933563703,
|
| 14 |
+
"learning_rate": 4.9999999999999996e-06,
|
| 15 |
+
"loss": 0.9537,
|
| 16 |
+
"step": 10
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.003365303718660609,
|
| 20 |
+
"grad_norm": 9.386746384686745,
|
| 21 |
+
"learning_rate": 9.999999999999999e-06,
|
| 22 |
+
"loss": 0.943,
|
| 23 |
+
"step": 20
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.005047955577990914,
|
| 27 |
+
"grad_norm": 7.387362447577942,
|
| 28 |
+
"learning_rate": 1.5e-05,
|
| 29 |
+
"loss": 0.934,
|
| 30 |
+
"step": 30
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.006730607437321218,
|
| 34 |
+
"grad_norm": 6.9256319824932655,
|
| 35 |
+
"learning_rate": 1.9999999999999998e-05,
|
| 36 |
+
"loss": 0.8376,
|
| 37 |
+
"step": 40
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.008413259296651522,
|
| 41 |
+
"grad_norm": 9.1148382590838,
|
| 42 |
+
"learning_rate": 2.5e-05,
|
| 43 |
+
"loss": 0.8484,
|
| 44 |
+
"step": 50
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.010095911155981827,
|
| 48 |
+
"grad_norm": 3.9989232759892426,
|
| 49 |
+
"learning_rate": 3e-05,
|
| 50 |
+
"loss": 0.8097,
|
| 51 |
+
"step": 60
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.011778563015312132,
|
| 55 |
+
"grad_norm": 3.892371218590039,
|
| 56 |
+
"learning_rate": 2.9999786123888308e-05,
|
| 57 |
+
"loss": 0.7811,
|
| 58 |
+
"step": 70
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.013461214874642436,
|
| 62 |
+
"grad_norm": 8.096662196282066,
|
| 63 |
+
"learning_rate": 2.9999144501652298e-05,
|
| 64 |
+
"loss": 0.7446,
|
| 65 |
+
"step": 80
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.01514386673397274,
|
| 69 |
+
"grad_norm": 1.5769306611206149,
|
| 70 |
+
"learning_rate": 2.9998075151588992e-05,
|
| 71 |
+
"loss": 0.7258,
|
| 72 |
+
"step": 90
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.016826518593303044,
|
| 76 |
+
"grad_norm": 8.47430485487969,
|
| 77 |
+
"learning_rate": 2.999657810419285e-05,
|
| 78 |
+
"loss": 0.7052,
|
| 79 |
+
"step": 100
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.01850917045263335,
|
| 83 |
+
"grad_norm": 2.363071299913598,
|
| 84 |
+
"learning_rate": 2.999465340215489e-05,
|
| 85 |
+
"loss": 0.7657,
|
| 86 |
+
"step": 110
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.020191822311963654,
|
| 90 |
+
"grad_norm": 1.9252385425154874,
|
| 91 |
+
"learning_rate": 2.999230110036149e-05,
|
| 92 |
+
"loss": 0.7329,
|
| 93 |
+
"step": 120
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.02187447417129396,
|
| 97 |
+
"grad_norm": 8.946028475031488,
|
| 98 |
+
"learning_rate": 2.99895212658928e-05,
|
| 99 |
+
"loss": 0.7304,
|
| 100 |
+
"step": 130
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.023557126030624265,
|
| 104 |
+
"grad_norm": 6.877609312630206,
|
| 105 |
+
"learning_rate": 2.9986313978020846e-05,
|
| 106 |
+
"loss": 0.7453,
|
| 107 |
+
"step": 140
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.02523977788995457,
|
| 111 |
+
"grad_norm": 2.5256324882367993,
|
| 112 |
+
"learning_rate": 2.9982679328207262e-05,
|
| 113 |
+
"loss": 0.7366,
|
| 114 |
+
"step": 150
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.02692242974928487,
|
| 118 |
+
"grad_norm": 2.709550398238738,
|
| 119 |
+
"learning_rate": 2.9978617420100692e-05,
|
| 120 |
+
"loss": 0.7258,
|
| 121 |
+
"step": 160
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.028605081608615177,
|
| 125 |
+
"grad_norm": 1.543550019689774,
|
| 126 |
+
"learning_rate": 2.9974128369533805e-05,
|
| 127 |
+
"loss": 0.7372,
|
| 128 |
+
"step": 170
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.03028773346794548,
|
| 132 |
+
"grad_norm": 3.3453966881155504,
|
| 133 |
+
"learning_rate": 2.9969212304520034e-05,
|
| 134 |
+
"loss": 0.743,
|
| 135 |
+
"step": 180
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.03197038532727579,
|
| 139 |
+
"grad_norm": 1.922001656181265,
|
| 140 |
+
"learning_rate": 2.9963869365249895e-05,
|
| 141 |
+
"loss": 0.7819,
|
| 142 |
+
"step": 190
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.03365303718660609,
|
| 146 |
+
"grad_norm": 2.0611188483400036,
|
| 147 |
+
"learning_rate": 2.995809970408699e-05,
|
| 148 |
+
"loss": 0.7155,
|
| 149 |
+
"step": 200
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.0353356890459364,
|
| 153 |
+
"grad_norm": 1.5313041833127259,
|
| 154 |
+
"learning_rate": 2.9951903485563685e-05,
|
| 155 |
+
"loss": 0.7322,
|
| 156 |
+
"step": 210
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.0370183409052667,
|
| 160 |
+
"grad_norm": 2.0124191694435085,
|
| 161 |
+
"learning_rate": 2.99452808863764e-05,
|
| 162 |
+
"loss": 0.6759,
|
| 163 |
+
"step": 220
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.03870099276459701,
|
| 167 |
+
"grad_norm": 3.182123324389477,
|
| 168 |
+
"learning_rate": 2.993823209538056e-05,
|
| 169 |
+
"loss": 0.6953,
|
| 170 |
+
"step": 230
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.04038364462392731,
|
| 174 |
+
"grad_norm": 1.6122782177661379,
|
| 175 |
+
"learning_rate": 2.9930757313585238e-05,
|
| 176 |
+
"loss": 0.6953,
|
| 177 |
+
"step": 240
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.04206629648325761,
|
| 181 |
+
"grad_norm": 2.2027482596695647,
|
| 182 |
+
"learning_rate": 2.9922856754147406e-05,
|
| 183 |
+
"loss": 0.7301,
|
| 184 |
+
"step": 250
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.04374894834258792,
|
| 188 |
+
"grad_norm": 2.6782477155989213,
|
| 189 |
+
"learning_rate": 2.9914530642365852e-05,
|
| 190 |
+
"loss": 0.6891,
|
| 191 |
+
"step": 260
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.04543160020191822,
|
| 195 |
+
"grad_norm": 1.9740401144541417,
|
| 196 |
+
"learning_rate": 2.990577921567476e-05,
|
| 197 |
+
"loss": 0.7231,
|
| 198 |
+
"step": 270
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.04711425206124853,
|
| 202 |
+
"grad_norm": 1.719874620968932,
|
| 203 |
+
"learning_rate": 2.989660272363696e-05,
|
| 204 |
+
"loss": 0.7505,
|
| 205 |
+
"step": 280
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.04879690392057883,
|
| 209 |
+
"grad_norm": 1.3138364164203409,
|
| 210 |
+
"learning_rate": 2.988700142793676e-05,
|
| 211 |
+
"loss": 0.7116,
|
| 212 |
+
"step": 290
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.05047955577990914,
|
| 216 |
+
"grad_norm": 5.853627389344256,
|
| 217 |
+
"learning_rate": 2.9876975602372536e-05,
|
| 218 |
+
"loss": 0.719,
|
| 219 |
+
"step": 300
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.05216220763923944,
|
| 223 |
+
"grad_norm": 2.347259437170711,
|
| 224 |
+
"learning_rate": 2.9866525532848906e-05,
|
| 225 |
+
"loss": 0.6803,
|
| 226 |
+
"step": 310
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.05384485949856974,
|
| 230 |
+
"grad_norm": 1.937679220955038,
|
| 231 |
+
"learning_rate": 2.9855651517368567e-05,
|
| 232 |
+
"loss": 0.7461,
|
| 233 |
+
"step": 320
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.05552751135790005,
|
| 237 |
+
"grad_norm": 1.6661300351569575,
|
| 238 |
+
"learning_rate": 2.9844353866023802e-05,
|
| 239 |
+
"loss": 0.7472,
|
| 240 |
+
"step": 330
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.05721016321723035,
|
| 244 |
+
"grad_norm": 2.357915869204484,
|
| 245 |
+
"learning_rate": 2.9832632900987642e-05,
|
| 246 |
+
"loss": 0.7148,
|
| 247 |
+
"step": 340
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.05889281507656066,
|
| 251 |
+
"grad_norm": 4.398815186243292,
|
| 252 |
+
"learning_rate": 2.982048895650468e-05,
|
| 253 |
+
"loss": 0.6992,
|
| 254 |
+
"step": 350
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.06057546693589096,
|
| 258 |
+
"grad_norm": 12.662682224480092,
|
| 259 |
+
"learning_rate": 2.9807922378881537e-05,
|
| 260 |
+
"loss": 0.7539,
|
| 261 |
+
"step": 360
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.06225811879522127,
|
| 265 |
+
"grad_norm": 0.8642696401357872,
|
| 266 |
+
"learning_rate": 2.979493352647697e-05,
|
| 267 |
+
"loss": 0.7212,
|
| 268 |
+
"step": 370
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.06394077065455157,
|
| 272 |
+
"grad_norm": 27.047937858232604,
|
| 273 |
+
"learning_rate": 2.9781522769691686e-05,
|
| 274 |
+
"loss": 0.722,
|
| 275 |
+
"step": 380
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.06562342251388188,
|
| 279 |
+
"grad_norm": 2.598805292448644,
|
| 280 |
+
"learning_rate": 2.9767690490957758e-05,
|
| 281 |
+
"loss": 0.7065,
|
| 282 |
+
"step": 390
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.06730607437321218,
|
| 286 |
+
"grad_norm": 1.2314762895092763,
|
| 287 |
+
"learning_rate": 2.9753437084727713e-05,
|
| 288 |
+
"loss": 0.7498,
|
| 289 |
+
"step": 400
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.06898872623254249,
|
| 293 |
+
"grad_norm": 1.6421909669790502,
|
| 294 |
+
"learning_rate": 2.9738762957463292e-05,
|
| 295 |
+
"loss": 0.6992,
|
| 296 |
+
"step": 410
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.0706713780918728,
|
| 300 |
+
"grad_norm": 2.023552968622588,
|
| 301 |
+
"learning_rate": 2.9723668527623877e-05,
|
| 302 |
+
"loss": 0.6943,
|
| 303 |
+
"step": 420
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.0723540299512031,
|
| 307 |
+
"grad_norm": 1.5172337910969138,
|
| 308 |
+
"learning_rate": 2.9708154225654526e-05,
|
| 309 |
+
"loss": 0.6987,
|
| 310 |
+
"step": 430
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.0740366818105334,
|
| 314 |
+
"grad_norm": 1.197852135730745,
|
| 315 |
+
"learning_rate": 2.9692220493973712e-05,
|
| 316 |
+
"loss": 0.7302,
|
| 317 |
+
"step": 440
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.0757193336698637,
|
| 321 |
+
"grad_norm": 2.4396443837967183,
|
| 322 |
+
"learning_rate": 2.9675867786960718e-05,
|
| 323 |
+
"loss": 0.7318,
|
| 324 |
+
"step": 450
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.07740198552919401,
|
| 328 |
+
"grad_norm": 1.4599851880563282,
|
| 329 |
+
"learning_rate": 2.9659096570942654e-05,
|
| 330 |
+
"loss": 0.6941,
|
| 331 |
+
"step": 460
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.07908463738852431,
|
| 335 |
+
"grad_norm": 1.117755825364562,
|
| 336 |
+
"learning_rate": 2.9641907324181194e-05,
|
| 337 |
+
"loss": 0.7399,
|
| 338 |
+
"step": 470
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.08076728924785462,
|
| 342 |
+
"grad_norm": 2.9235378164576242,
|
| 343 |
+
"learning_rate": 2.96243005368589e-05,
|
| 344 |
+
"loss": 0.7207,
|
| 345 |
+
"step": 480
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.08244994110718493,
|
| 349 |
+
"grad_norm": 7.308883163781362,
|
| 350 |
+
"learning_rate": 2.960627671106527e-05,
|
| 351 |
+
"loss": 0.682,
|
| 352 |
+
"step": 490
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.08413259296651522,
|
| 356 |
+
"grad_norm": 3.4394827932955234,
|
| 357 |
+
"learning_rate": 2.9587836360782405e-05,
|
| 358 |
+
"loss": 0.708,
|
| 359 |
+
"step": 500
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.08581524482584553,
|
| 363 |
+
"grad_norm": 3.2314529856927634,
|
| 364 |
+
"learning_rate": 2.9568980011870357e-05,
|
| 365 |
+
"loss": 0.7335,
|
| 366 |
+
"step": 510
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.08749789668517584,
|
| 370 |
+
"grad_norm": 1.825724533695325,
|
| 371 |
+
"learning_rate": 2.954970820205214e-05,
|
| 372 |
+
"loss": 0.6951,
|
| 373 |
+
"step": 520
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.08918054854450615,
|
| 377 |
+
"grad_norm": 3.3231741746640076,
|
| 378 |
+
"learning_rate": 2.9530021480898393e-05,
|
| 379 |
+
"loss": 0.7793,
|
| 380 |
+
"step": 530
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.09086320040383644,
|
| 384 |
+
"grad_norm": 1.3097651462571123,
|
| 385 |
+
"learning_rate": 2.9509920409811696e-05,
|
| 386 |
+
"loss": 0.7087,
|
| 387 |
+
"step": 540
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.09254585226316675,
|
| 391 |
+
"grad_norm": 6.685911471215255,
|
| 392 |
+
"learning_rate": 2.9489405562010565e-05,
|
| 393 |
+
"loss": 0.6906,
|
| 394 |
+
"step": 550
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.09422850412249706,
|
| 398 |
+
"grad_norm": 2.870746617513948,
|
| 399 |
+
"learning_rate": 2.9468477522513132e-05,
|
| 400 |
+
"loss": 0.7028,
|
| 401 |
+
"step": 560
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.09591115598182735,
|
| 405 |
+
"grad_norm": 1.782555352805469,
|
| 406 |
+
"learning_rate": 2.9447136888120408e-05,
|
| 407 |
+
"loss": 0.6901,
|
| 408 |
+
"step": 570
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.09759380784115766,
|
| 412 |
+
"grad_norm": 2.336519711000487,
|
| 413 |
+
"learning_rate": 2.9425384267399327e-05,
|
| 414 |
+
"loss": 0.7779,
|
| 415 |
+
"step": 580
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.09927645970048797,
|
| 419 |
+
"grad_norm": 8.935574410818228,
|
| 420 |
+
"learning_rate": 2.940322028066534e-05,
|
| 421 |
+
"loss": 0.7503,
|
| 422 |
+
"step": 590
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.10095911155981828,
|
| 426 |
+
"grad_norm": 2.754713786882031,
|
| 427 |
+
"learning_rate": 2.938064555996476e-05,
|
| 428 |
+
"loss": 0.7208,
|
| 429 |
+
"step": 600
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.10264176341914857,
|
| 433 |
+
"grad_norm": 1.5082503557652136,
|
| 434 |
+
"learning_rate": 2.9357660749056713e-05,
|
| 435 |
+
"loss": 0.7169,
|
| 436 |
+
"step": 610
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.10432441527847888,
|
| 440 |
+
"grad_norm": 9.04522194526273,
|
| 441 |
+
"learning_rate": 2.9334266503394803e-05,
|
| 442 |
+
"loss": 0.6927,
|
| 443 |
+
"step": 620
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.10600706713780919,
|
| 447 |
+
"grad_norm": 55.28278686388287,
|
| 448 |
+
"learning_rate": 2.9310463490108397e-05,
|
| 449 |
+
"loss": 0.7107,
|
| 450 |
+
"step": 630
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.10768971899713949,
|
| 454 |
+
"grad_norm": 3.721916069105249,
|
| 455 |
+
"learning_rate": 2.928625238798362e-05,
|
| 456 |
+
"loss": 0.6951,
|
| 457 |
+
"step": 640
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.1093723708564698,
|
| 461 |
+
"grad_norm": 2.5040797323750112,
|
| 462 |
+
"learning_rate": 2.9261633887443993e-05,
|
| 463 |
+
"loss": 0.6916,
|
| 464 |
+
"step": 650
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.1110550227158001,
|
| 468 |
+
"grad_norm": 3.5468924769840617,
|
| 469 |
+
"learning_rate": 2.9236608690530738e-05,
|
| 470 |
+
"loss": 0.7077,
|
| 471 |
+
"step": 660
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.11273767457513041,
|
| 475 |
+
"grad_norm": 3.0266819778200746,
|
| 476 |
+
"learning_rate": 2.921117751088276e-05,
|
| 477 |
+
"loss": 0.6952,
|
| 478 |
+
"step": 670
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.1144203264344607,
|
| 482 |
+
"grad_norm": 1.634743894298146,
|
| 483 |
+
"learning_rate": 2.91853410737163e-05,
|
| 484 |
+
"loss": 0.6936,
|
| 485 |
+
"step": 680
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.11610297829379101,
|
| 489 |
+
"grad_norm": 1.0925365801520501,
|
| 490 |
+
"learning_rate": 2.915910011580426e-05,
|
| 491 |
+
"loss": 0.7317,
|
| 492 |
+
"step": 690
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 0.11778563015312132,
|
| 496 |
+
"grad_norm": 1.6959112138540386,
|
| 497 |
+
"learning_rate": 2.9132455385455176e-05,
|
| 498 |
+
"loss": 0.6917,
|
| 499 |
+
"step": 700
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.11946828201245162,
|
| 503 |
+
"grad_norm": 1.9723433746891168,
|
| 504 |
+
"learning_rate": 2.9105407642491895e-05,
|
| 505 |
+
"loss": 0.7209,
|
| 506 |
+
"step": 710
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.12115093387178193,
|
| 510 |
+
"grad_norm": 2.1537215293733833,
|
| 511 |
+
"learning_rate": 2.907795765822989e-05,
|
| 512 |
+
"loss": 0.7488,
|
| 513 |
+
"step": 720
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.12283358573111224,
|
| 517 |
+
"grad_norm": 3.227101869737169,
|
| 518 |
+
"learning_rate": 2.9050106215455283e-05,
|
| 519 |
+
"loss": 0.7152,
|
| 520 |
+
"step": 730
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.12451623759044254,
|
| 524 |
+
"grad_norm": 2.7222358893572554,
|
| 525 |
+
"learning_rate": 2.9021854108402516e-05,
|
| 526 |
+
"loss": 0.708,
|
| 527 |
+
"step": 740
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.12619888944977284,
|
| 531 |
+
"grad_norm": 2.1054843767538136,
|
| 532 |
+
"learning_rate": 2.8993202142731693e-05,
|
| 533 |
+
"loss": 0.7251,
|
| 534 |
+
"step": 750
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.12788154130910315,
|
| 538 |
+
"grad_norm": 2.11845883419618,
|
| 539 |
+
"learning_rate": 2.8964151135505616e-05,
|
| 540 |
+
"loss": 0.7405,
|
| 541 |
+
"step": 760
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.12956419316843346,
|
| 545 |
+
"grad_norm": 13.171512404187755,
|
| 546 |
+
"learning_rate": 2.8934701915166477e-05,
|
| 547 |
+
"loss": 0.6844,
|
| 548 |
+
"step": 770
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.13124684502776376,
|
| 552 |
+
"grad_norm": 2.7633375632879127,
|
| 553 |
+
"learning_rate": 2.890485532151225e-05,
|
| 554 |
+
"loss": 0.6766,
|
| 555 |
+
"step": 780
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.13292949688709407,
|
| 559 |
+
"grad_norm": 1.8420785342693768,
|
| 560 |
+
"learning_rate": 2.887461220567271e-05,
|
| 561 |
+
"loss": 0.7037,
|
| 562 |
+
"step": 790
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 0.13461214874642435,
|
| 566 |
+
"grad_norm": 1.5557447509529954,
|
| 567 |
+
"learning_rate": 2.8843973430085204e-05,
|
| 568 |
+
"loss": 0.6991,
|
| 569 |
+
"step": 800
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.13629480060575466,
|
| 573 |
+
"grad_norm": 1.9295826624758823,
|
| 574 |
+
"learning_rate": 2.8812939868470016e-05,
|
| 575 |
+
"loss": 0.6956,
|
| 576 |
+
"step": 810
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 0.13797745246508497,
|
| 580 |
+
"grad_norm": 3.3211216557707126,
|
| 581 |
+
"learning_rate": 2.878151240580548e-05,
|
| 582 |
+
"loss": 0.6774,
|
| 583 |
+
"step": 820
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 0.13966010432441528,
|
| 587 |
+
"grad_norm": 4.196064403930616,
|
| 588 |
+
"learning_rate": 2.874969193830274e-05,
|
| 589 |
+
"loss": 0.6752,
|
| 590 |
+
"step": 830
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.1413427561837456,
|
| 594 |
+
"grad_norm": 5.574976270137628,
|
| 595 |
+
"learning_rate": 2.871747937338016e-05,
|
| 596 |
+
"loss": 0.6553,
|
| 597 |
+
"step": 840
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 0.1430254080430759,
|
| 601 |
+
"grad_norm": 1.6494038718740478,
|
| 602 |
+
"learning_rate": 2.8684875629637505e-05,
|
| 603 |
+
"loss": 0.7152,
|
| 604 |
+
"step": 850
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 0.1447080599024062,
|
| 608 |
+
"grad_norm": 1.3061892609414858,
|
| 609 |
+
"learning_rate": 2.8651881636829698e-05,
|
| 610 |
+
"loss": 0.7462,
|
| 611 |
+
"step": 860
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 0.1463907117617365,
|
| 615 |
+
"grad_norm": 4.321044418392694,
|
| 616 |
+
"learning_rate": 2.861849833584032e-05,
|
| 617 |
+
"loss": 0.6902,
|
| 618 |
+
"step": 870
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 0.1480733636210668,
|
| 622 |
+
"grad_norm": 2.9444722968009764,
|
| 623 |
+
"learning_rate": 2.8584726678654787e-05,
|
| 624 |
+
"loss": 0.6813,
|
| 625 |
+
"step": 880
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.1497560154803971,
|
| 629 |
+
"grad_norm": 1.4940245340163587,
|
| 630 |
+
"learning_rate": 2.85505676283332e-05,
|
| 631 |
+
"loss": 0.689,
|
| 632 |
+
"step": 890
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.1514386673397274,
|
| 636 |
+
"grad_norm": 3.3704010040589565,
|
| 637 |
+
"learning_rate": 2.851602215898287e-05,
|
| 638 |
+
"loss": 0.6953,
|
| 639 |
+
"step": 900
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.15312131919905772,
|
| 643 |
+
"grad_norm": 1.6597144402924948,
|
| 644 |
+
"learning_rate": 2.8481091255730552e-05,
|
| 645 |
+
"loss": 0.7277,
|
| 646 |
+
"step": 910
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 0.15480397105838803,
|
| 650 |
+
"grad_norm": 10.969872224353953,
|
| 651 |
+
"learning_rate": 2.844577591469435e-05,
|
| 652 |
+
"loss": 0.7142,
|
| 653 |
+
"step": 920
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 0.15648662291771834,
|
| 657 |
+
"grad_norm": 8.45616831264245,
|
| 658 |
+
"learning_rate": 2.8410077142955304e-05,
|
| 659 |
+
"loss": 0.7197,
|
| 660 |
+
"step": 930
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 0.15816927477704862,
|
| 664 |
+
"grad_norm": 2.9594258901214427,
|
| 665 |
+
"learning_rate": 2.8373995958528683e-05,
|
| 666 |
+
"loss": 0.7351,
|
| 667 |
+
"step": 940
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 0.15985192663637893,
|
| 671 |
+
"grad_norm": 2.168676312428759,
|
| 672 |
+
"learning_rate": 2.8337533390334942e-05,
|
| 673 |
+
"loss": 0.7544,
|
| 674 |
+
"step": 950
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 0.16153457849570924,
|
| 678 |
+
"grad_norm": 7.898767360662744,
|
| 679 |
+
"learning_rate": 2.8300690478170388e-05,
|
| 680 |
+
"loss": 0.7015,
|
| 681 |
+
"step": 960
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 0.16321723035503954,
|
| 685 |
+
"grad_norm": 16.83650212945308,
|
| 686 |
+
"learning_rate": 2.826346827267753e-05,
|
| 687 |
+
"loss": 0.7139,
|
| 688 |
+
"step": 970
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 0.16489988221436985,
|
| 692 |
+
"grad_norm": 2.3791337429068977,
|
| 693 |
+
"learning_rate": 2.8225867835315114e-05,
|
| 694 |
+
"loss": 0.7053,
|
| 695 |
+
"step": 980
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 0.16658253407370016,
|
| 699 |
+
"grad_norm": 1.9679363325295285,
|
| 700 |
+
"learning_rate": 2.8187890238327842e-05,
|
| 701 |
+
"loss": 0.7313,
|
| 702 |
+
"step": 990
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 0.16826518593303044,
|
| 706 |
+
"grad_norm": 1.4822625638777076,
|
| 707 |
+
"learning_rate": 2.814953656471583e-05,
|
| 708 |
+
"loss": 0.7085,
|
| 709 |
+
"step": 1000
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 0.16994783779236075,
|
| 713 |
+
"grad_norm": 2.647291447509443,
|
| 714 |
+
"learning_rate": 2.8110807908203682e-05,
|
| 715 |
+
"loss": 0.6638,
|
| 716 |
+
"step": 1010
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 0.17163048965169106,
|
| 720 |
+
"grad_norm": 2.969379719654364,
|
| 721 |
+
"learning_rate": 2.8071705373209328e-05,
|
| 722 |
+
"loss": 0.6884,
|
| 723 |
+
"step": 1020
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 0.17331314151102137,
|
| 727 |
+
"grad_norm": 1.1163745403124403,
|
| 728 |
+
"learning_rate": 2.803223007481252e-05,
|
| 729 |
+
"loss": 0.6885,
|
| 730 |
+
"step": 1030
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 0.17499579337035168,
|
| 734 |
+
"grad_norm": 1.2686557979094786,
|
| 735 |
+
"learning_rate": 2.7992383138723034e-05,
|
| 736 |
+
"loss": 0.7037,
|
| 737 |
+
"step": 1040
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 0.17667844522968199,
|
| 741 |
+
"grad_norm": 4.648945448875594,
|
| 742 |
+
"learning_rate": 2.7952165701248573e-05,
|
| 743 |
+
"loss": 0.6933,
|
| 744 |
+
"step": 1050
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 0.1783610970890123,
|
| 748 |
+
"grad_norm": 4.723564874595428,
|
| 749 |
+
"learning_rate": 2.7911578909262353e-05,
|
| 750 |
+
"loss": 0.7144,
|
| 751 |
+
"step": 1060
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 0.18004374894834257,
|
| 755 |
+
"grad_norm": 5.211806926801946,
|
| 756 |
+
"learning_rate": 2.787062392017041e-05,
|
| 757 |
+
"loss": 0.7266,
|
| 758 |
+
"step": 1070
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 0.18172640080767288,
|
| 762 |
+
"grad_norm": 1.3725560316172503,
|
| 763 |
+
"learning_rate": 2.7829301901878592e-05,
|
| 764 |
+
"loss": 0.7445,
|
| 765 |
+
"step": 1080
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 0.1834090526670032,
|
| 769 |
+
"grad_norm": 0.9012241436004484,
|
| 770 |
+
"learning_rate": 2.7787614032759243e-05,
|
| 771 |
+
"loss": 0.6986,
|
| 772 |
+
"step": 1090
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 0.1850917045263335,
|
| 776 |
+
"grad_norm": 2.912544243603394,
|
| 777 |
+
"learning_rate": 2.7745561501617605e-05,
|
| 778 |
+
"loss": 0.7173,
|
| 779 |
+
"step": 1100
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 0.1867743563856638,
|
| 783 |
+
"grad_norm": 1.4248442614931247,
|
| 784 |
+
"learning_rate": 2.7703145507657923e-05,
|
| 785 |
+
"loss": 0.7035,
|
| 786 |
+
"step": 1110
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 0.18845700824499412,
|
| 790 |
+
"grad_norm": 2.186609904533333,
|
| 791 |
+
"learning_rate": 2.766036726044926e-05,
|
| 792 |
+
"loss": 0.7371,
|
| 793 |
+
"step": 1120
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 0.19013966010432443,
|
| 797 |
+
"grad_norm": 2.0524595532166603,
|
| 798 |
+
"learning_rate": 2.7617227979890957e-05,
|
| 799 |
+
"loss": 0.6986,
|
| 800 |
+
"step": 1130
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 0.1918223119636547,
|
| 804 |
+
"grad_norm": 1.8227045280907195,
|
| 805 |
+
"learning_rate": 2.7573728896177897e-05,
|
| 806 |
+
"loss": 0.7075,
|
| 807 |
+
"step": 1140
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 0.19350496382298502,
|
| 811 |
+
"grad_norm": 1.8425998009576734,
|
| 812 |
+
"learning_rate": 2.7529871249765397e-05,
|
| 813 |
+
"loss": 0.6897,
|
| 814 |
+
"step": 1150
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 0.19518761568231532,
|
| 818 |
+
"grad_norm": 5.3035191638420836,
|
| 819 |
+
"learning_rate": 2.7485656291333845e-05,
|
| 820 |
+
"loss": 0.7027,
|
| 821 |
+
"step": 1160
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 0.19687026754164563,
|
| 825 |
+
"grad_norm": 3.3228474353685504,
|
| 826 |
+
"learning_rate": 2.7441085281753028e-05,
|
| 827 |
+
"loss": 0.7091,
|
| 828 |
+
"step": 1170
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 0.19855291940097594,
|
| 832 |
+
"grad_norm": 3.5016968564731283,
|
| 833 |
+
"learning_rate": 2.739615949204617e-05,
|
| 834 |
+
"loss": 0.7241,
|
| 835 |
+
"step": 1180
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 0.20023557126030625,
|
| 839 |
+
"grad_norm": 1.7190048028902127,
|
| 840 |
+
"learning_rate": 2.7350880203353703e-05,
|
| 841 |
+
"loss": 0.7192,
|
| 842 |
+
"step": 1190
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 0.20191822311963656,
|
| 846 |
+
"grad_norm": 3.7186824247487515,
|
| 847 |
+
"learning_rate": 2.7305248706896722e-05,
|
| 848 |
+
"loss": 0.7063,
|
| 849 |
+
"step": 1200
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 0.20360087497896684,
|
| 853 |
+
"grad_norm": 4.1717869895766935,
|
| 854 |
+
"learning_rate": 2.7259266303940164e-05,
|
| 855 |
+
"loss": 0.7088,
|
| 856 |
+
"step": 1210
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 0.20528352683829715,
|
| 860 |
+
"grad_norm": 2.5124857963805804,
|
| 861 |
+
"learning_rate": 2.7212934305755697e-05,
|
| 862 |
+
"loss": 0.7198,
|
| 863 |
+
"step": 1220
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 0.20696617869762746,
|
| 867 |
+
"grad_norm": 2.095136268936366,
|
| 868 |
+
"learning_rate": 2.7166254033584343e-05,
|
| 869 |
+
"loss": 0.753,
|
| 870 |
+
"step": 1230
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 0.20864883055695777,
|
| 874 |
+
"grad_norm": 3.2661098868577256,
|
| 875 |
+
"learning_rate": 2.7119226818598784e-05,
|
| 876 |
+
"loss": 0.6779,
|
| 877 |
+
"step": 1240
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 0.21033148241628807,
|
| 881 |
+
"grad_norm": 3.055506603735091,
|
| 882 |
+
"learning_rate": 2.7071854001865402e-05,
|
| 883 |
+
"loss": 0.7013,
|
| 884 |
+
"step": 1250
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 0.21201413427561838,
|
| 888 |
+
"grad_norm": 12.522953778477769,
|
| 889 |
+
"learning_rate": 2.702413693430604e-05,
|
| 890 |
+
"loss": 0.7088,
|
| 891 |
+
"step": 1260
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 0.2136967861349487,
|
| 895 |
+
"grad_norm": 3.476240301739368,
|
| 896 |
+
"learning_rate": 2.697607697665948e-05,
|
| 897 |
+
"loss": 0.689,
|
| 898 |
+
"step": 1270
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 0.21537943799427897,
|
| 902 |
+
"grad_norm": 1.1862686197570156,
|
| 903 |
+
"learning_rate": 2.6927675499442648e-05,
|
| 904 |
+
"loss": 0.7243,
|
| 905 |
+
"step": 1280
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 0.21706208985360928,
|
| 909 |
+
"grad_norm": 1.6505042403801382,
|
| 910 |
+
"learning_rate": 2.68789338829115e-05,
|
| 911 |
+
"loss": 0.7083,
|
| 912 |
+
"step": 1290
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 0.2187447417129396,
|
| 916 |
+
"grad_norm": 4.74071740077375,
|
| 917 |
+
"learning_rate": 2.6829853517021698e-05,
|
| 918 |
+
"loss": 0.7016,
|
| 919 |
+
"step": 1300
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 0.2204273935722699,
|
| 923 |
+
"grad_norm": 4.124079283639458,
|
| 924 |
+
"learning_rate": 2.6780435801388945e-05,
|
| 925 |
+
"loss": 0.7077,
|
| 926 |
+
"step": 1310
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 0.2221100454316002,
|
| 930 |
+
"grad_norm": 1.9487864410536297,
|
| 931 |
+
"learning_rate": 2.6730682145249093e-05,
|
| 932 |
+
"loss": 0.7355,
|
| 933 |
+
"step": 1320
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 0.22379269729093051,
|
| 937 |
+
"grad_norm": 2.4839241050514733,
|
| 938 |
+
"learning_rate": 2.668059396741795e-05,
|
| 939 |
+
"loss": 0.7092,
|
| 940 |
+
"step": 1330
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 0.22547534915026082,
|
| 944 |
+
"grad_norm": 2.841913657394254,
|
| 945 |
+
"learning_rate": 2.6630172696250804e-05,
|
| 946 |
+
"loss": 0.7303,
|
| 947 |
+
"step": 1340
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 0.2271580010095911,
|
| 951 |
+
"grad_norm": 2.7442870185873347,
|
| 952 |
+
"learning_rate": 2.6579419769601715e-05,
|
| 953 |
+
"loss": 0.6739,
|
| 954 |
+
"step": 1350
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 0.2288406528689214,
|
| 958 |
+
"grad_norm": 1.3854365909071105,
|
| 959 |
+
"learning_rate": 2.6528336634782493e-05,
|
| 960 |
+
"loss": 0.7073,
|
| 961 |
+
"step": 1360
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 0.23052330472825172,
|
| 965 |
+
"grad_norm": 3.115941001607779,
|
| 966 |
+
"learning_rate": 2.6476924748521443e-05,
|
| 967 |
+
"loss": 0.7267,
|
| 968 |
+
"step": 1370
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 0.23220595658758203,
|
| 972 |
+
"grad_norm": 6.9185951332741,
|
| 973 |
+
"learning_rate": 2.6425185576921812e-05,
|
| 974 |
+
"loss": 0.7456,
|
| 975 |
+
"step": 1380
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 0.23388860844691234,
|
| 979 |
+
"grad_norm": 2.378601355345996,
|
| 980 |
+
"learning_rate": 2.637312059541997e-05,
|
| 981 |
+
"loss": 0.6912,
|
| 982 |
+
"step": 1390
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 0.23557126030624265,
|
| 986 |
+
"grad_norm": 2.7929947858543906,
|
| 987 |
+
"learning_rate": 2.632073128874336e-05,
|
| 988 |
+
"loss": 0.7184,
|
| 989 |
+
"step": 1400
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 0.23725391216557296,
|
| 993 |
+
"grad_norm": 1.5382855773213957,
|
| 994 |
+
"learning_rate": 2.6268019150868144e-05,
|
| 995 |
+
"loss": 0.7099,
|
| 996 |
+
"step": 1410
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 0.23893656402490324,
|
| 1000 |
+
"grad_norm": 6.1010563795570025,
|
| 1001 |
+
"learning_rate": 2.62149856849766e-05,
|
| 1002 |
+
"loss": 0.6895,
|
| 1003 |
+
"step": 1420
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 0.24061921588423354,
|
| 1007 |
+
"grad_norm": 5.999491987974443,
|
| 1008 |
+
"learning_rate": 2.616163240341426e-05,
|
| 1009 |
+
"loss": 0.7493,
|
| 1010 |
+
"step": 1430
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 0.24230186774356385,
|
| 1014 |
+
"grad_norm": 2.837037600849311,
|
| 1015 |
+
"learning_rate": 2.6107960827646774e-05,
|
| 1016 |
+
"loss": 0.7176,
|
| 1017 |
+
"step": 1440
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 0.24398451960289416,
|
| 1021 |
+
"grad_norm": 1.7029089834427125,
|
| 1022 |
+
"learning_rate": 2.6053972488216538e-05,
|
| 1023 |
+
"loss": 0.6852,
|
| 1024 |
+
"step": 1450
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 0.24566717146222447,
|
| 1028 |
+
"grad_norm": 1.382189249222589,
|
| 1029 |
+
"learning_rate": 2.5999668924699035e-05,
|
| 1030 |
+
"loss": 0.685,
|
| 1031 |
+
"step": 1460
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 0.24734982332155478,
|
| 1035 |
+
"grad_norm": 1.9496045543050813,
|
| 1036 |
+
"learning_rate": 2.5945051685658923e-05,
|
| 1037 |
+
"loss": 0.6591,
|
| 1038 |
+
"step": 1470
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 0.2490324751808851,
|
| 1042 |
+
"grad_norm": 5.479390805764353,
|
| 1043 |
+
"learning_rate": 2.5890122328605908e-05,
|
| 1044 |
+
"loss": 0.7085,
|
| 1045 |
+
"step": 1480
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 0.25071512704021537,
|
| 1049 |
+
"grad_norm": 1.7567995670915637,
|
| 1050 |
+
"learning_rate": 2.5834882419950295e-05,
|
| 1051 |
+
"loss": 0.7091,
|
| 1052 |
+
"step": 1490
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 0.2523977788995457,
|
| 1056 |
+
"grad_norm": 1.9685911084195309,
|
| 1057 |
+
"learning_rate": 2.577933353495833e-05,
|
| 1058 |
+
"loss": 0.7218,
|
| 1059 |
+
"step": 1500
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 0.254080430758876,
|
| 1063 |
+
"grad_norm": 3.400633915540874,
|
| 1064 |
+
"learning_rate": 2.5723477257707293e-05,
|
| 1065 |
+
"loss": 0.7148,
|
| 1066 |
+
"step": 1510
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 0.2557630826182063,
|
| 1070 |
+
"grad_norm": 1.2116738326443663,
|
| 1071 |
+
"learning_rate": 2.566731518104029e-05,
|
| 1072 |
+
"loss": 0.7321,
|
| 1073 |
+
"step": 1520
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 0.2574457344775366,
|
| 1077 |
+
"grad_norm": 1.3376343864594256,
|
| 1078 |
+
"learning_rate": 2.5610848906520878e-05,
|
| 1079 |
+
"loss": 0.748,
|
| 1080 |
+
"step": 1530
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 0.2591283863368669,
|
| 1084 |
+
"grad_norm": 2.6089861003232055,
|
| 1085 |
+
"learning_rate": 2.5554080044387344e-05,
|
| 1086 |
+
"loss": 0.7127,
|
| 1087 |
+
"step": 1540
|
| 1088 |
+
},
|
| 1089 |
+
{
|
| 1090 |
+
"epoch": 0.2608110381961972,
|
| 1091 |
+
"grad_norm": 3.2047926120640526,
|
| 1092 |
+
"learning_rate": 2.5497010213506825e-05,
|
| 1093 |
+
"loss": 0.7262,
|
| 1094 |
+
"step": 1550
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 0.26249369005552753,
|
| 1098 |
+
"grad_norm": 1.4899957348295265,
|
| 1099 |
+
"learning_rate": 2.5439641041329128e-05,
|
| 1100 |
+
"loss": 0.7122,
|
| 1101 |
+
"step": 1560
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 0.26417634191485784,
|
| 1105 |
+
"grad_norm": 3.595968473922136,
|
| 1106 |
+
"learning_rate": 2.5381974163840313e-05,
|
| 1107 |
+
"loss": 0.7092,
|
| 1108 |
+
"step": 1570
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"epoch": 0.26585899377418815,
|
| 1112 |
+
"grad_norm": 3.5232117574234003,
|
| 1113 |
+
"learning_rate": 2.532401122551605e-05,
|
| 1114 |
+
"loss": 0.6924,
|
| 1115 |
+
"step": 1580
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 0.2675416456335184,
|
| 1119 |
+
"grad_norm": 2.618947453668302,
|
| 1120 |
+
"learning_rate": 2.526575387927473e-05,
|
| 1121 |
+
"loss": 0.7067,
|
| 1122 |
+
"step": 1590
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 0.2692242974928487,
|
| 1126 |
+
"grad_norm": 3.6282673284589566,
|
| 1127 |
+
"learning_rate": 2.52072037864303e-05,
|
| 1128 |
+
"loss": 0.6945,
|
| 1129 |
+
"step": 1600
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 0.270906949352179,
|
| 1133 |
+
"grad_norm": 2.2274379147013,
|
| 1134 |
+
"learning_rate": 2.5148362616644926e-05,
|
| 1135 |
+
"loss": 0.6727,
|
| 1136 |
+
"step": 1610
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 0.2725896012115093,
|
| 1140 |
+
"grad_norm": 2.823867881580523,
|
| 1141 |
+
"learning_rate": 2.508923204788135e-05,
|
| 1142 |
+
"loss": 0.7158,
|
| 1143 |
+
"step": 1620
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 0.27427225307083963,
|
| 1147 |
+
"grad_norm": 2.0118901151982245,
|
| 1148 |
+
"learning_rate": 2.5029813766355062e-05,
|
| 1149 |
+
"loss": 0.7422,
|
| 1150 |
+
"step": 1630
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 0.27595490493016994,
|
| 1154 |
+
"grad_norm": 1.2843584175617246,
|
| 1155 |
+
"learning_rate": 2.4970109466486202e-05,
|
| 1156 |
+
"loss": 0.7099,
|
| 1157 |
+
"step": 1640
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 0.27763755678950025,
|
| 1161 |
+
"grad_norm": 3.5059277881120914,
|
| 1162 |
+
"learning_rate": 2.491012085085122e-05,
|
| 1163 |
+
"loss": 0.7164,
|
| 1164 |
+
"step": 1650
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 0.27932020864883056,
|
| 1168 |
+
"grad_norm": 1.7458993688338285,
|
| 1169 |
+
"learning_rate": 2.4849849630134384e-05,
|
| 1170 |
+
"loss": 0.6901,
|
| 1171 |
+
"step": 1660
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 0.28100286050816087,
|
| 1175 |
+
"grad_norm": 5.813346226937464,
|
| 1176 |
+
"learning_rate": 2.4789297523078924e-05,
|
| 1177 |
+
"loss": 0.7181,
|
| 1178 |
+
"step": 1670
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 0.2826855123674912,
|
| 1182 |
+
"grad_norm": 2.0515286491489237,
|
| 1183 |
+
"learning_rate": 2.4728466256438072e-05,
|
| 1184 |
+
"loss": 0.7431,
|
| 1185 |
+
"step": 1680
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 0.2843681642268215,
|
| 1189 |
+
"grad_norm": 2.6702746679350375,
|
| 1190 |
+
"learning_rate": 2.4667357564925798e-05,
|
| 1191 |
+
"loss": 0.701,
|
| 1192 |
+
"step": 1690
|
| 1193 |
+
},
|
| 1194 |
+
{
|
| 1195 |
+
"epoch": 0.2860508160861518,
|
| 1196 |
+
"grad_norm": 2.707565805299449,
|
| 1197 |
+
"learning_rate": 2.460597319116735e-05,
|
| 1198 |
+
"loss": 0.6725,
|
| 1199 |
+
"step": 1700
|
| 1200 |
+
},
|
| 1201 |
+
{
|
| 1202 |
+
"epoch": 0.2877334679454821,
|
| 1203 |
+
"grad_norm": 1.7994267796032153,
|
| 1204 |
+
"learning_rate": 2.4544314885649552e-05,
|
| 1205 |
+
"loss": 0.7043,
|
| 1206 |
+
"step": 1710
|
| 1207 |
+
},
|
| 1208 |
+
{
|
| 1209 |
+
"epoch": 0.2894161198048124,
|
| 1210 |
+
"grad_norm": 2.240627477157692,
|
| 1211 |
+
"learning_rate": 2.4482384406670883e-05,
|
| 1212 |
+
"loss": 0.7337,
|
| 1213 |
+
"step": 1720
|
| 1214 |
+
},
|
| 1215 |
+
{
|
| 1216 |
+
"epoch": 0.29109877166414266,
|
| 1217 |
+
"grad_norm": 1.4093208691675285,
|
| 1218 |
+
"learning_rate": 2.4420183520291354e-05,
|
| 1219 |
+
"loss": 0.706,
|
| 1220 |
+
"step": 1730
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 0.292781423523473,
|
| 1224 |
+
"grad_norm": 1.5799653304195502,
|
| 1225 |
+
"learning_rate": 2.4357714000282127e-05,
|
| 1226 |
+
"loss": 0.7254,
|
| 1227 |
+
"step": 1740
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"epoch": 0.2944640753828033,
|
| 1231 |
+
"grad_norm": 1.8282839714116759,
|
| 1232 |
+
"learning_rate": 2.4294977628074938e-05,
|
| 1233 |
+
"loss": 0.68,
|
| 1234 |
+
"step": 1750
|
| 1235 |
+
},
|
| 1236 |
+
{
|
| 1237 |
+
"epoch": 0.2961467272421336,
|
| 1238 |
+
"grad_norm": 13.490769798309381,
|
| 1239 |
+
"learning_rate": 2.42319761927113e-05,
|
| 1240 |
+
"loss": 0.6984,
|
| 1241 |
+
"step": 1760
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 0.2978293791014639,
|
| 1245 |
+
"grad_norm": 1.1660842236351188,
|
| 1246 |
+
"learning_rate": 2.4168711490791484e-05,
|
| 1247 |
+
"loss": 0.6893,
|
| 1248 |
+
"step": 1770
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 0.2995120309607942,
|
| 1252 |
+
"grad_norm": 1.4880113732457052,
|
| 1253 |
+
"learning_rate": 2.4105185326423286e-05,
|
| 1254 |
+
"loss": 0.7371,
|
| 1255 |
+
"step": 1780
|
| 1256 |
+
},
|
| 1257 |
+
{
|
| 1258 |
+
"epoch": 0.3011946828201245,
|
| 1259 |
+
"grad_norm": 1.9796491202207207,
|
| 1260 |
+
"learning_rate": 2.4041399511170574e-05,
|
| 1261 |
+
"loss": 0.7372,
|
| 1262 |
+
"step": 1790
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 0.3028773346794548,
|
| 1266 |
+
"grad_norm": 3.2861914347482846,
|
| 1267 |
+
"learning_rate": 2.3977355864001635e-05,
|
| 1268 |
+
"loss": 0.7145,
|
| 1269 |
+
"step": 1800
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 0.30455998653878513,
|
| 1273 |
+
"grad_norm": 3.8536888582450595,
|
| 1274 |
+
"learning_rate": 2.3913056211237304e-05,
|
| 1275 |
+
"loss": 0.7244,
|
| 1276 |
+
"step": 1810
|
| 1277 |
+
},
|
| 1278 |
+
{
|
| 1279 |
+
"epoch": 0.30624263839811544,
|
| 1280 |
+
"grad_norm": 2.250827213388724,
|
| 1281 |
+
"learning_rate": 2.3848502386498866e-05,
|
| 1282 |
+
"loss": 0.7444,
|
| 1283 |
+
"step": 1820
|
| 1284 |
+
},
|
| 1285 |
+
{
|
| 1286 |
+
"epoch": 0.30792529025744575,
|
| 1287 |
+
"grad_norm": 1.6760548188250846,
|
| 1288 |
+
"learning_rate": 2.3783696230655802e-05,
|
| 1289 |
+
"loss": 0.7415,
|
| 1290 |
+
"step": 1830
|
| 1291 |
+
},
|
| 1292 |
+
{
|
| 1293 |
+
"epoch": 0.30960794211677606,
|
| 1294 |
+
"grad_norm": 2.83690011157284,
|
| 1295 |
+
"learning_rate": 2.371863959177326e-05,
|
| 1296 |
+
"loss": 0.6769,
|
| 1297 |
+
"step": 1840
|
| 1298 |
+
},
|
| 1299 |
+
{
|
| 1300 |
+
"epoch": 0.31129059397610637,
|
| 1301 |
+
"grad_norm": 3.6586666108883037,
|
| 1302 |
+
"learning_rate": 2.365333432505937e-05,
|
| 1303 |
+
"loss": 0.6981,
|
| 1304 |
+
"step": 1850
|
| 1305 |
+
},
|
| 1306 |
+
{
|
| 1307 |
+
"epoch": 0.3129732458354367,
|
| 1308 |
+
"grad_norm": 2.967916913846329,
|
| 1309 |
+
"learning_rate": 2.3587782292812323e-05,
|
| 1310 |
+
"loss": 0.7235,
|
| 1311 |
+
"step": 1860
|
| 1312 |
+
},
|
| 1313 |
+
{
|
| 1314 |
+
"epoch": 0.31465589769476693,
|
| 1315 |
+
"grad_norm": 2.7607388194454607,
|
| 1316 |
+
"learning_rate": 2.35219853643673e-05,
|
| 1317 |
+
"loss": 0.7202,
|
| 1318 |
+
"step": 1870
|
| 1319 |
+
},
|
| 1320 |
+
{
|
| 1321 |
+
"epoch": 0.31633854955409724,
|
| 1322 |
+
"grad_norm": 2.5793375573884925,
|
| 1323 |
+
"learning_rate": 2.3455945416043132e-05,
|
| 1324 |
+
"loss": 0.7437,
|
| 1325 |
+
"step": 1880
|
| 1326 |
+
},
|
| 1327 |
+
{
|
| 1328 |
+
"epoch": 0.31802120141342755,
|
| 1329 |
+
"grad_norm": 1.6474727320404343,
|
| 1330 |
+
"learning_rate": 2.338966433108879e-05,
|
| 1331 |
+
"loss": 0.6664,
|
| 1332 |
+
"step": 1890
|
| 1333 |
+
},
|
| 1334 |
+
{
|
| 1335 |
+
"epoch": 0.31970385327275785,
|
| 1336 |
+
"grad_norm": 2.8252072958720102,
|
| 1337 |
+
"learning_rate": 2.3323143999629712e-05,
|
| 1338 |
+
"loss": 0.6641,
|
| 1339 |
+
"step": 1900
|
| 1340 |
+
},
|
| 1341 |
+
{
|
| 1342 |
+
"epoch": 0.32138650513208816,
|
| 1343 |
+
"grad_norm": 1.8240997471681801,
|
| 1344 |
+
"learning_rate": 2.3256386318613877e-05,
|
| 1345 |
+
"loss": 0.7029,
|
| 1346 |
+
"step": 1910
|
| 1347 |
+
},
|
| 1348 |
+
{
|
| 1349 |
+
"epoch": 0.32306915699141847,
|
| 1350 |
+
"grad_norm": 1.7867386563705459,
|
| 1351 |
+
"learning_rate": 2.318939319175771e-05,
|
| 1352 |
+
"loss": 0.6806,
|
| 1353 |
+
"step": 1920
|
| 1354 |
+
},
|
| 1355 |
+
{
|
| 1356 |
+
"epoch": 0.3247518088507488,
|
| 1357 |
+
"grad_norm": 2.519605910503542,
|
| 1358 |
+
"learning_rate": 2.3122166529491822e-05,
|
| 1359 |
+
"loss": 0.6837,
|
| 1360 |
+
"step": 1930
|
| 1361 |
+
},
|
| 1362 |
+
{
|
| 1363 |
+
"epoch": 0.3264344607100791,
|
| 1364 |
+
"grad_norm": 1.5090617010699425,
|
| 1365 |
+
"learning_rate": 2.3054708248906483e-05,
|
| 1366 |
+
"loss": 0.7201,
|
| 1367 |
+
"step": 1940
|
| 1368 |
+
},
|
| 1369 |
+
{
|
| 1370 |
+
"epoch": 0.3281171125694094,
|
| 1371 |
+
"grad_norm": 1.85373627743108,
|
| 1372 |
+
"learning_rate": 2.2987020273696996e-05,
|
| 1373 |
+
"loss": 0.7007,
|
| 1374 |
+
"step": 1950
|
| 1375 |
+
},
|
| 1376 |
+
{
|
| 1377 |
+
"epoch": 0.3297997644287397,
|
| 1378 |
+
"grad_norm": 3.1668783585579714,
|
| 1379 |
+
"learning_rate": 2.2919104534108825e-05,
|
| 1380 |
+
"loss": 0.6827,
|
| 1381 |
+
"step": 1960
|
| 1382 |
+
},
|
| 1383 |
+
{
|
| 1384 |
+
"epoch": 0.33148241628807,
|
| 1385 |
+
"grad_norm": 2.802801151344103,
|
| 1386 |
+
"learning_rate": 2.2850962966882547e-05,
|
| 1387 |
+
"loss": 0.733,
|
| 1388 |
+
"step": 1970
|
| 1389 |
+
},
|
| 1390 |
+
{
|
| 1391 |
+
"epoch": 0.3331650681474003,
|
| 1392 |
+
"grad_norm": 4.351080547606847,
|
| 1393 |
+
"learning_rate": 2.278259751519861e-05,
|
| 1394 |
+
"loss": 0.7125,
|
| 1395 |
+
"step": 1980
|
| 1396 |
+
},
|
| 1397 |
+
{
|
| 1398 |
+
"epoch": 0.33484772000673063,
|
| 1399 |
+
"grad_norm": 1.4284076903376268,
|
| 1400 |
+
"learning_rate": 2.2714010128621957e-05,
|
| 1401 |
+
"loss": 0.7166,
|
| 1402 |
+
"step": 1990
|
| 1403 |
+
},
|
| 1404 |
+
{
|
| 1405 |
+
"epoch": 0.3365303718660609,
|
| 1406 |
+
"grad_norm": 1.4047557097137526,
|
| 1407 |
+
"learning_rate": 2.2645202763046385e-05,
|
| 1408 |
+
"loss": 0.7306,
|
| 1409 |
+
"step": 2000
|
| 1410 |
+
},
|
| 1411 |
+
{
|
| 1412 |
+
"epoch": 0.3382130237253912,
|
| 1413 |
+
"grad_norm": 1.7792478988054032,
|
| 1414 |
+
"learning_rate": 2.2576177380638808e-05,
|
| 1415 |
+
"loss": 0.6819,
|
| 1416 |
+
"step": 2010
|
| 1417 |
+
},
|
| 1418 |
+
{
|
| 1419 |
+
"epoch": 0.3398956755847215,
|
| 1420 |
+
"grad_norm": 1.541280861427235,
|
| 1421 |
+
"learning_rate": 2.2506935949783277e-05,
|
| 1422 |
+
"loss": 0.7188,
|
| 1423 |
+
"step": 2020
|
| 1424 |
+
},
|
| 1425 |
+
{
|
| 1426 |
+
"epoch": 0.3415783274440518,
|
| 1427 |
+
"grad_norm": 3.1010621866032717,
|
| 1428 |
+
"learning_rate": 2.243748044502485e-05,
|
| 1429 |
+
"loss": 0.7262,
|
| 1430 |
+
"step": 2030
|
| 1431 |
+
},
|
| 1432 |
+
{
|
| 1433 |
+
"epoch": 0.3432609793033821,
|
| 1434 |
+
"grad_norm": 2.3564311476470245,
|
| 1435 |
+
"learning_rate": 2.236781284701332e-05,
|
| 1436 |
+
"loss": 0.6862,
|
| 1437 |
+
"step": 2040
|
| 1438 |
+
},
|
| 1439 |
+
{
|
| 1440 |
+
"epoch": 0.3449436311627124,
|
| 1441 |
+
"grad_norm": 2.277474191898381,
|
| 1442 |
+
"learning_rate": 2.229793514244666e-05,
|
| 1443 |
+
"loss": 0.7086,
|
| 1444 |
+
"step": 2050
|
| 1445 |
+
},
|
| 1446 |
+
{
|
| 1447 |
+
"epoch": 0.34662628302204274,
|
| 1448 |
+
"grad_norm": 1.9189792309740905,
|
| 1449 |
+
"learning_rate": 2.222784932401445e-05,
|
| 1450 |
+
"loss": 0.6785,
|
| 1451 |
+
"step": 2060
|
| 1452 |
+
},
|
| 1453 |
+
{
|
| 1454 |
+
"epoch": 0.34830893488137304,
|
| 1455 |
+
"grad_norm": 8.557409847942546,
|
| 1456 |
+
"learning_rate": 2.2157557390341e-05,
|
| 1457 |
+
"loss": 0.7162,
|
| 1458 |
+
"step": 2070
|
| 1459 |
+
},
|
| 1460 |
+
{
|
| 1461 |
+
"epoch": 0.34999158674070335,
|
| 1462 |
+
"grad_norm": 2.0580462138378324,
|
| 1463 |
+
"learning_rate": 2.2087061345928375e-05,
|
| 1464 |
+
"loss": 0.663,
|
| 1465 |
+
"step": 2080
|
| 1466 |
+
},
|
| 1467 |
+
{
|
| 1468 |
+
"epoch": 0.35167423860003366,
|
| 1469 |
+
"grad_norm": 11.262172640870563,
|
| 1470 |
+
"learning_rate": 2.2016363201099205e-05,
|
| 1471 |
+
"loss": 0.7363,
|
| 1472 |
+
"step": 2090
|
| 1473 |
+
},
|
| 1474 |
+
{
|
| 1475 |
+
"epoch": 0.35335689045936397,
|
| 1476 |
+
"grad_norm": 12.967220942427355,
|
| 1477 |
+
"learning_rate": 2.1945464971939424e-05,
|
| 1478 |
+
"loss": 0.7169,
|
| 1479 |
+
"step": 2100
|
| 1480 |
+
},
|
| 1481 |
+
{
|
| 1482 |
+
"epoch": 0.3550395423186943,
|
| 1483 |
+
"grad_norm": 31.131995950709616,
|
| 1484 |
+
"learning_rate": 2.1874368680240692e-05,
|
| 1485 |
+
"loss": 0.7109,
|
| 1486 |
+
"step": 2110
|
| 1487 |
+
},
|
| 1488 |
+
{
|
| 1489 |
+
"epoch": 0.3567221941780246,
|
| 1490 |
+
"grad_norm": 3.759967527046392,
|
| 1491 |
+
"learning_rate": 2.1803076353442806e-05,
|
| 1492 |
+
"loss": 0.72,
|
| 1493 |
+
"step": 2120
|
| 1494 |
+
},
|
| 1495 |
+
{
|
| 1496 |
+
"epoch": 0.3584048460373549,
|
| 1497 |
+
"grad_norm": 2.0854209231683494,
|
| 1498 |
+
"learning_rate": 2.1731590024575848e-05,
|
| 1499 |
+
"loss": 0.7443,
|
| 1500 |
+
"step": 2130
|
| 1501 |
+
},
|
| 1502 |
+
{
|
| 1503 |
+
"epoch": 0.36008749789668515,
|
| 1504 |
+
"grad_norm": 1.550491190575623,
|
| 1505 |
+
"learning_rate": 2.165991173220223e-05,
|
| 1506 |
+
"loss": 0.7272,
|
| 1507 |
+
"step": 2140
|
| 1508 |
+
},
|
| 1509 |
+
{
|
| 1510 |
+
"epoch": 0.36177014975601546,
|
| 1511 |
+
"grad_norm": 3.184378797474582,
|
| 1512 |
+
"learning_rate": 2.158804352035855e-05,
|
| 1513 |
+
"loss": 0.6796,
|
| 1514 |
+
"step": 2150
|
| 1515 |
+
},
|
| 1516 |
+
{
|
| 1517 |
+
"epoch": 0.36345280161534577,
|
| 1518 |
+
"grad_norm": 2.2017388277936623,
|
| 1519 |
+
"learning_rate": 2.1515987438497295e-05,
|
| 1520 |
+
"loss": 0.7079,
|
| 1521 |
+
"step": 2160
|
| 1522 |
+
},
|
| 1523 |
+
{
|
| 1524 |
+
"epoch": 0.3651354534746761,
|
| 1525 |
+
"grad_norm": 1.8785352439752243,
|
| 1526 |
+
"learning_rate": 2.1443745541428416e-05,
|
| 1527 |
+
"loss": 0.7157,
|
| 1528 |
+
"step": 2170
|
| 1529 |
+
},
|
| 1530 |
+
{
|
| 1531 |
+
"epoch": 0.3668181053340064,
|
| 1532 |
+
"grad_norm": 5.799376408169486,
|
| 1533 |
+
"learning_rate": 2.137131988926072e-05,
|
| 1534 |
+
"loss": 0.7147,
|
| 1535 |
+
"step": 2180
|
| 1536 |
+
},
|
| 1537 |
+
{
|
| 1538 |
+
"epoch": 0.3685007571933367,
|
| 1539 |
+
"grad_norm": 1.463705775425984,
|
| 1540 |
+
"learning_rate": 2.129871254734312e-05,
|
| 1541 |
+
"loss": 0.6965,
|
| 1542 |
+
"step": 2190
|
| 1543 |
+
},
|
| 1544 |
+
{
|
| 1545 |
+
"epoch": 0.370183409052667,
|
| 1546 |
+
"grad_norm": 2.8778655656985435,
|
| 1547 |
+
"learning_rate": 2.122592558620575e-05,
|
| 1548 |
+
"loss": 0.6853,
|
| 1549 |
+
"step": 2200
|
| 1550 |
+
},
|
| 1551 |
+
{
|
| 1552 |
+
"epoch": 0.3718660609119973,
|
| 1553 |
+
"grad_norm": 2.830232793716089,
|
| 1554 |
+
"learning_rate": 2.1152961081500906e-05,
|
| 1555 |
+
"loss": 0.6682,
|
| 1556 |
+
"step": 2210
|
| 1557 |
+
},
|
| 1558 |
+
{
|
| 1559 |
+
"epoch": 0.3735487127713276,
|
| 1560 |
+
"grad_norm": 7.790226305121679,
|
| 1561 |
+
"learning_rate": 2.1079821113943863e-05,
|
| 1562 |
+
"loss": 0.6855,
|
| 1563 |
+
"step": 2220
|
| 1564 |
+
},
|
| 1565 |
+
{
|
| 1566 |
+
"epoch": 0.3752313646306579,
|
| 1567 |
+
"grad_norm": 1.635629077245975,
|
| 1568 |
+
"learning_rate": 2.100650776925353e-05,
|
| 1569 |
+
"loss": 0.7156,
|
| 1570 |
+
"step": 2230
|
| 1571 |
+
},
|
| 1572 |
+
{
|
| 1573 |
+
"epoch": 0.37691401648998824,
|
| 1574 |
+
"grad_norm": 1.6353332190474046,
|
| 1575 |
+
"learning_rate": 2.0933023138092995e-05,
|
| 1576 |
+
"loss": 0.7083,
|
| 1577 |
+
"step": 2240
|
| 1578 |
+
},
|
| 1579 |
+
{
|
| 1580 |
+
"epoch": 0.37859666834931854,
|
| 1581 |
+
"grad_norm": 2.4020219788955237,
|
| 1582 |
+
"learning_rate": 2.0859369316009877e-05,
|
| 1583 |
+
"loss": 0.6873,
|
| 1584 |
+
"step": 2250
|
| 1585 |
+
},
|
| 1586 |
+
{
|
| 1587 |
+
"epoch": 0.38027932020864885,
|
| 1588 |
+
"grad_norm": 1.5183766349594623,
|
| 1589 |
+
"learning_rate": 2.0785548403376592e-05,
|
| 1590 |
+
"loss": 0.6882,
|
| 1591 |
+
"step": 2260
|
| 1592 |
+
},
|
| 1593 |
+
{
|
| 1594 |
+
"epoch": 0.38196197206797916,
|
| 1595 |
+
"grad_norm": 3.9509980884297313,
|
| 1596 |
+
"learning_rate": 2.0711562505330437e-05,
|
| 1597 |
+
"loss": 0.7037,
|
| 1598 |
+
"step": 2270
|
| 1599 |
+
},
|
| 1600 |
+
{
|
| 1601 |
+
"epoch": 0.3836446239273094,
|
| 1602 |
+
"grad_norm": 1.080804508141519,
|
| 1603 |
+
"learning_rate": 2.063741373171357e-05,
|
| 1604 |
+
"loss": 0.7124,
|
| 1605 |
+
"step": 2280
|
| 1606 |
+
},
|
| 1607 |
+
{
|
| 1608 |
+
"epoch": 0.3853272757866397,
|
| 1609 |
+
"grad_norm": 2.4771901968792975,
|
| 1610 |
+
"learning_rate": 2.0563104197012847e-05,
|
| 1611 |
+
"loss": 0.7348,
|
| 1612 |
+
"step": 2290
|
| 1613 |
+
},
|
| 1614 |
+
{
|
| 1615 |
+
"epoch": 0.38700992764597003,
|
| 1616 |
+
"grad_norm": 1.4012276716265397,
|
| 1617 |
+
"learning_rate": 2.048863602029951e-05,
|
| 1618 |
+
"loss": 0.7157,
|
| 1619 |
+
"step": 2300
|
| 1620 |
+
},
|
| 1621 |
+
{
|
| 1622 |
+
"epoch": 0.38869257950530034,
|
| 1623 |
+
"grad_norm": 1.8764316814985353,
|
| 1624 |
+
"learning_rate": 2.0414011325168777e-05,
|
| 1625 |
+
"loss": 0.6932,
|
| 1626 |
+
"step": 2310
|
| 1627 |
+
},
|
| 1628 |
+
{
|
| 1629 |
+
"epoch": 0.39037523136463065,
|
| 1630 |
+
"grad_norm": 3.2158348437646183,
|
| 1631 |
+
"learning_rate": 2.0339232239679252e-05,
|
| 1632 |
+
"loss": 0.6789,
|
| 1633 |
+
"step": 2320
|
| 1634 |
+
},
|
| 1635 |
+
{
|
| 1636 |
+
"epoch": 0.39205788322396096,
|
| 1637 |
+
"grad_norm": 2.468508068736803,
|
| 1638 |
+
"learning_rate": 2.026430089629229e-05,
|
| 1639 |
+
"loss": 0.7163,
|
| 1640 |
+
"step": 2330
|
| 1641 |
+
},
|
| 1642 |
+
{
|
| 1643 |
+
"epoch": 0.39374053508329127,
|
| 1644 |
+
"grad_norm": 2.0179845911955656,
|
| 1645 |
+
"learning_rate": 2.0189219431811123e-05,
|
| 1646 |
+
"loss": 0.6566,
|
| 1647 |
+
"step": 2340
|
| 1648 |
+
},
|
| 1649 |
+
{
|
| 1650 |
+
"epoch": 0.3954231869426216,
|
| 1651 |
+
"grad_norm": 0.9653345936734927,
|
| 1652 |
+
"learning_rate": 2.0113989987319988e-05,
|
| 1653 |
+
"loss": 0.6939,
|
| 1654 |
+
"step": 2350
|
| 1655 |
+
},
|
| 1656 |
+
{
|
| 1657 |
+
"epoch": 0.3971058388019519,
|
| 1658 |
+
"grad_norm": 1.568403155522187,
|
| 1659 |
+
"learning_rate": 2.0038614708123023e-05,
|
| 1660 |
+
"loss": 0.7288,
|
| 1661 |
+
"step": 2360
|
| 1662 |
+
},
|
| 1663 |
+
{
|
| 1664 |
+
"epoch": 0.3987884906612822,
|
| 1665 |
+
"grad_norm": 5.098482519541986,
|
| 1666 |
+
"learning_rate": 1.996309574368311e-05,
|
| 1667 |
+
"loss": 0.7217,
|
| 1668 |
+
"step": 2370
|
| 1669 |
+
},
|
| 1670 |
+
{
|
| 1671 |
+
"epoch": 0.4004711425206125,
|
| 1672 |
+
"grad_norm": 1.656770228609968,
|
| 1673 |
+
"learning_rate": 1.9887435247560586e-05,
|
| 1674 |
+
"loss": 0.743,
|
| 1675 |
+
"step": 2380
|
| 1676 |
+
},
|
| 1677 |
+
{
|
| 1678 |
+
"epoch": 0.4021537943799428,
|
| 1679 |
+
"grad_norm": 1.679661595344395,
|
| 1680 |
+
"learning_rate": 1.981163537735181e-05,
|
| 1681 |
+
"loss": 0.724,
|
| 1682 |
+
"step": 2390
|
| 1683 |
+
},
|
| 1684 |
+
{
|
| 1685 |
+
"epoch": 0.4038364462392731,
|
| 1686 |
+
"grad_norm": 2.057009585071239,
|
| 1687 |
+
"learning_rate": 1.9735698294627644e-05,
|
| 1688 |
+
"loss": 0.7228,
|
| 1689 |
+
"step": 2400
|
| 1690 |
+
},
|
| 1691 |
+
{
|
| 1692 |
+
"epoch": 0.4055190980986034,
|
| 1693 |
+
"grad_norm": 1.6262124746744704,
|
| 1694 |
+
"learning_rate": 1.9659626164871828e-05,
|
| 1695 |
+
"loss": 0.6621,
|
| 1696 |
+
"step": 2410
|
| 1697 |
+
},
|
| 1698 |
+
{
|
| 1699 |
+
"epoch": 0.4072017499579337,
|
| 1700 |
+
"grad_norm": 8.087231459721274,
|
| 1701 |
+
"learning_rate": 1.95834211574192e-05,
|
| 1702 |
+
"loss": 0.6988,
|
| 1703 |
+
"step": 2420
|
| 1704 |
+
},
|
| 1705 |
+
{
|
| 1706 |
+
"epoch": 0.408884401817264,
|
| 1707 |
+
"grad_norm": 2.053339294676776,
|
| 1708 |
+
"learning_rate": 1.9507085445393855e-05,
|
| 1709 |
+
"loss": 0.7056,
|
| 1710 |
+
"step": 2430
|
| 1711 |
+
},
|
| 1712 |
+
{
|
| 1713 |
+
"epoch": 0.4105670536765943,
|
| 1714 |
+
"grad_norm": 2.295299690475712,
|
| 1715 |
+
"learning_rate": 1.9430621205647156e-05,
|
| 1716 |
+
"loss": 0.7231,
|
| 1717 |
+
"step": 2440
|
| 1718 |
+
},
|
| 1719 |
+
{
|
| 1720 |
+
"epoch": 0.4122497055359246,
|
| 1721 |
+
"grad_norm": 2.419261777687713,
|
| 1722 |
+
"learning_rate": 1.935403061869568e-05,
|
| 1723 |
+
"loss": 0.7786,
|
| 1724 |
+
"step": 2450
|
| 1725 |
+
},
|
| 1726 |
+
{
|
| 1727 |
+
"epoch": 0.4139323573952549,
|
| 1728 |
+
"grad_norm": 1.2971369467313154,
|
| 1729 |
+
"learning_rate": 1.9277315868659017e-05,
|
| 1730 |
+
"loss": 0.6857,
|
| 1731 |
+
"step": 2460
|
| 1732 |
+
},
|
| 1733 |
+
{
|
| 1734 |
+
"epoch": 0.4156150092545852,
|
| 1735 |
+
"grad_norm": 3.89724254450292,
|
| 1736 |
+
"learning_rate": 1.920047914319749e-05,
|
| 1737 |
+
"loss": 0.6795,
|
| 1738 |
+
"step": 2470
|
| 1739 |
+
},
|
| 1740 |
+
{
|
| 1741 |
+
"epoch": 0.41729766111391553,
|
| 1742 |
+
"grad_norm": 1.8276445458438462,
|
| 1743 |
+
"learning_rate": 1.9123522633449772e-05,
|
| 1744 |
+
"loss": 0.7249,
|
| 1745 |
+
"step": 2480
|
| 1746 |
+
},
|
| 1747 |
+
{
|
| 1748 |
+
"epoch": 0.41898031297324584,
|
| 1749 |
+
"grad_norm": 1.7262471459079933,
|
| 1750 |
+
"learning_rate": 1.9046448533970423e-05,
|
| 1751 |
+
"loss": 0.7055,
|
| 1752 |
+
"step": 2490
|
| 1753 |
+
},
|
| 1754 |
+
{
|
| 1755 |
+
"epoch": 0.42066296483257615,
|
| 1756 |
+
"grad_norm": 1.5815796760993692,
|
| 1757 |
+
"learning_rate": 1.8969259042667255e-05,
|
| 1758 |
+
"loss": 0.7012,
|
| 1759 |
+
"step": 2500
|
| 1760 |
+
},
|
| 1761 |
+
{
|
| 1762 |
+
"epoch": 0.42234561669190646,
|
| 1763 |
+
"grad_norm": 1.0102816357132662,
|
| 1764 |
+
"learning_rate": 1.8891956360738706e-05,
|
| 1765 |
+
"loss": 0.6803,
|
| 1766 |
+
"step": 2510
|
| 1767 |
+
},
|
| 1768 |
+
{
|
| 1769 |
+
"epoch": 0.42402826855123676,
|
| 1770 |
+
"grad_norm": 8.978060365810029,
|
| 1771 |
+
"learning_rate": 1.881454269261105e-05,
|
| 1772 |
+
"loss": 0.6791,
|
| 1773 |
+
"step": 2520
|
| 1774 |
+
},
|
| 1775 |
+
{
|
| 1776 |
+
"epoch": 0.4257109204105671,
|
| 1777 |
+
"grad_norm": 1.7740952536546388,
|
| 1778 |
+
"learning_rate": 1.873702024587553e-05,
|
| 1779 |
+
"loss": 0.7118,
|
| 1780 |
+
"step": 2530
|
| 1781 |
+
},
|
| 1782 |
+
{
|
| 1783 |
+
"epoch": 0.4273935722698974,
|
| 1784 |
+
"grad_norm": 16.141334008508057,
|
| 1785 |
+
"learning_rate": 1.865939123122541e-05,
|
| 1786 |
+
"loss": 0.7053,
|
| 1787 |
+
"step": 2540
|
| 1788 |
+
},
|
| 1789 |
+
{
|
| 1790 |
+
"epoch": 0.4290762241292277,
|
| 1791 |
+
"grad_norm": 1.7964747107424808,
|
| 1792 |
+
"learning_rate": 1.858165786239293e-05,
|
| 1793 |
+
"loss": 0.706,
|
| 1794 |
+
"step": 2550
|
| 1795 |
+
},
|
| 1796 |
+
{
|
| 1797 |
+
"epoch": 0.43075887598855794,
|
| 1798 |
+
"grad_norm": 2.1026202521481743,
|
| 1799 |
+
"learning_rate": 1.850382235608618e-05,
|
| 1800 |
+
"loss": 0.7057,
|
| 1801 |
+
"step": 2560
|
| 1802 |
+
},
|
| 1803 |
+
{
|
| 1804 |
+
"epoch": 0.43244152784788825,
|
| 1805 |
+
"grad_norm": 2.8681027149657226,
|
| 1806 |
+
"learning_rate": 1.842588693192589e-05,
|
| 1807 |
+
"loss": 0.7111,
|
| 1808 |
+
"step": 2570
|
| 1809 |
+
},
|
| 1810 |
+
{
|
| 1811 |
+
"epoch": 0.43412417970721856,
|
| 1812 |
+
"grad_norm": 1.2285580771772646,
|
| 1813 |
+
"learning_rate": 1.834785381238212e-05,
|
| 1814 |
+
"loss": 0.7023,
|
| 1815 |
+
"step": 2580
|
| 1816 |
+
},
|
| 1817 |
+
{
|
| 1818 |
+
"epoch": 0.43580683156654887,
|
| 1819 |
+
"grad_norm": 1.86184784397222,
|
| 1820 |
+
"learning_rate": 1.8269725222710895e-05,
|
| 1821 |
+
"loss": 0.6614,
|
| 1822 |
+
"step": 2590
|
| 1823 |
+
},
|
| 1824 |
+
{
|
| 1825 |
+
"epoch": 0.4374894834258792,
|
| 1826 |
+
"grad_norm": 1.7349607835894336,
|
| 1827 |
+
"learning_rate": 1.8191503390890745e-05,
|
| 1828 |
+
"loss": 0.6976,
|
| 1829 |
+
"step": 2600
|
| 1830 |
+
},
|
| 1831 |
+
{
|
| 1832 |
+
"epoch": 0.4391721352852095,
|
| 1833 |
+
"grad_norm": 4.623472425426236,
|
| 1834 |
+
"learning_rate": 1.8113190547559167e-05,
|
| 1835 |
+
"loss": 0.7147,
|
| 1836 |
+
"step": 2610
|
| 1837 |
+
},
|
| 1838 |
+
{
|
| 1839 |
+
"epoch": 0.4408547871445398,
|
| 1840 |
+
"grad_norm": 1.71339692622417,
|
| 1841 |
+
"learning_rate": 1.8034788925949024e-05,
|
| 1842 |
+
"loss": 0.6664,
|
| 1843 |
+
"step": 2620
|
| 1844 |
+
},
|
| 1845 |
+
{
|
| 1846 |
+
"epoch": 0.4425374390038701,
|
| 1847 |
+
"grad_norm": 3.0050627586678975,
|
| 1848 |
+
"learning_rate": 1.795630076182484e-05,
|
| 1849 |
+
"loss": 0.6802,
|
| 1850 |
+
"step": 2630
|
| 1851 |
+
},
|
| 1852 |
+
{
|
| 1853 |
+
"epoch": 0.4442200908632004,
|
| 1854 |
+
"grad_norm": 3.1349481547325992,
|
| 1855 |
+
"learning_rate": 1.7877728293419067e-05,
|
| 1856 |
+
"loss": 0.6981,
|
| 1857 |
+
"step": 2640
|
| 1858 |
+
},
|
| 1859 |
+
{
|
| 1860 |
+
"epoch": 0.4459027427225307,
|
| 1861 |
+
"grad_norm": 3.3593498761324336,
|
| 1862 |
+
"learning_rate": 1.7799073761368236e-05,
|
| 1863 |
+
"loss": 0.7219,
|
| 1864 |
+
"step": 2650
|
| 1865 |
+
},
|
| 1866 |
+
{
|
| 1867 |
+
"epoch": 0.44758539458186103,
|
| 1868 |
+
"grad_norm": 1.7981704686011195,
|
| 1869 |
+
"learning_rate": 1.7720339408649084e-05,
|
| 1870 |
+
"loss": 0.7163,
|
| 1871 |
+
"step": 2660
|
| 1872 |
+
},
|
| 1873 |
+
{
|
| 1874 |
+
"epoch": 0.44926804644119134,
|
| 1875 |
+
"grad_norm": 1.2590076147308225,
|
| 1876 |
+
"learning_rate": 1.7641527480514575e-05,
|
| 1877 |
+
"loss": 0.7086,
|
| 1878 |
+
"step": 2670
|
| 1879 |
+
},
|
| 1880 |
+
{
|
| 1881 |
+
"epoch": 0.45095069830052165,
|
| 1882 |
+
"grad_norm": 5.385441773268162,
|
| 1883 |
+
"learning_rate": 1.756264022442987e-05,
|
| 1884 |
+
"loss": 0.6695,
|
| 1885 |
+
"step": 2680
|
| 1886 |
+
},
|
| 1887 |
+
{
|
| 1888 |
+
"epoch": 0.4526333501598519,
|
| 1889 |
+
"grad_norm": 1.6481056783365962,
|
| 1890 |
+
"learning_rate": 1.7483679890008247e-05,
|
| 1891 |
+
"loss": 0.7314,
|
| 1892 |
+
"step": 2690
|
| 1893 |
+
},
|
| 1894 |
+
{
|
| 1895 |
+
"epoch": 0.4543160020191822,
|
| 1896 |
+
"grad_norm": 1.6160788306327938,
|
| 1897 |
+
"learning_rate": 1.740464872894695e-05,
|
| 1898 |
+
"loss": 0.7336,
|
| 1899 |
+
"step": 2700
|
| 1900 |
+
},
|
| 1901 |
+
{
|
| 1902 |
+
"epoch": 0.4559986538785125,
|
| 1903 |
+
"grad_norm": 1.4049942343651247,
|
| 1904 |
+
"learning_rate": 1.732554899496297e-05,
|
| 1905 |
+
"loss": 0.6399,
|
| 1906 |
+
"step": 2710
|
| 1907 |
+
},
|
| 1908 |
+
{
|
| 1909 |
+
"epoch": 0.4576813057378428,
|
| 1910 |
+
"grad_norm": 1.0231254626105863,
|
| 1911 |
+
"learning_rate": 1.7246382943728777e-05,
|
| 1912 |
+
"loss": 0.7213,
|
| 1913 |
+
"step": 2720
|
| 1914 |
+
},
|
| 1915 |
+
{
|
| 1916 |
+
"epoch": 0.45936395759717313,
|
| 1917 |
+
"grad_norm": 1.938670327906363,
|
| 1918 |
+
"learning_rate": 1.7167152832807997e-05,
|
| 1919 |
+
"loss": 0.6879,
|
| 1920 |
+
"step": 2730
|
| 1921 |
+
},
|
| 1922 |
+
{
|
| 1923 |
+
"epoch": 0.46104660945650344,
|
| 1924 |
+
"grad_norm": 3.044969011471669,
|
| 1925 |
+
"learning_rate": 1.7087860921591047e-05,
|
| 1926 |
+
"loss": 0.7113,
|
| 1927 |
+
"step": 2740
|
| 1928 |
+
},
|
| 1929 |
+
{
|
| 1930 |
+
"epoch": 0.46272926131583375,
|
| 1931 |
+
"grad_norm": 2.2385420988451394,
|
| 1932 |
+
"learning_rate": 1.7008509471230673e-05,
|
| 1933 |
+
"loss": 0.6928,
|
| 1934 |
+
"step": 2750
|
| 1935 |
+
},
|
| 1936 |
+
{
|
| 1937 |
+
"epoch": 0.46441191317516406,
|
| 1938 |
+
"grad_norm": 2.281718793802536,
|
| 1939 |
+
"learning_rate": 1.69291007445775e-05,
|
| 1940 |
+
"loss": 0.7608,
|
| 1941 |
+
"step": 2760
|
| 1942 |
+
},
|
| 1943 |
+
{
|
| 1944 |
+
"epoch": 0.46609456503449437,
|
| 1945 |
+
"grad_norm": 2.588995378875439,
|
| 1946 |
+
"learning_rate": 1.684963700611548e-05,
|
| 1947 |
+
"loss": 0.6455,
|
| 1948 |
+
"step": 2770
|
| 1949 |
+
},
|
| 1950 |
+
{
|
| 1951 |
+
"epoch": 0.4677772168938247,
|
| 1952 |
+
"grad_norm": 2.8099788810235027,
|
| 1953 |
+
"learning_rate": 1.677012052189734e-05,
|
| 1954 |
+
"loss": 0.7263,
|
| 1955 |
+
"step": 2780
|
| 1956 |
+
},
|
| 1957 |
+
{
|
| 1958 |
+
"epoch": 0.469459868753155,
|
| 1959 |
+
"grad_norm": 4.221047912947499,
|
| 1960 |
+
"learning_rate": 1.6690553559479946e-05,
|
| 1961 |
+
"loss": 0.6781,
|
| 1962 |
+
"step": 2790
|
| 1963 |
+
},
|
| 1964 |
+
{
|
| 1965 |
+
"epoch": 0.4711425206124853,
|
| 1966 |
+
"grad_norm": 2.1824366142795304,
|
| 1967 |
+
"learning_rate": 1.6610938387859623e-05,
|
| 1968 |
+
"loss": 0.6756,
|
| 1969 |
+
"step": 2800
|
| 1970 |
+
}
|
| 1971 |
+
],
|
| 1972 |
+
"logging_steps": 10,
|
| 1973 |
+
"max_steps": 5943,
|
| 1974 |
+
"num_input_tokens_seen": 0,
|
| 1975 |
+
"num_train_epochs": 1,
|
| 1976 |
+
"save_steps": 400,
|
| 1977 |
+
"stateful_callbacks": {
|
| 1978 |
+
"TrainerControl": {
|
| 1979 |
+
"args": {
|
| 1980 |
+
"should_epoch_stop": false,
|
| 1981 |
+
"should_evaluate": false,
|
| 1982 |
+
"should_log": false,
|
| 1983 |
+
"should_save": true,
|
| 1984 |
+
"should_training_stop": false
|
| 1985 |
+
},
|
| 1986 |
+
"attributes": {}
|
| 1987 |
+
}
|
| 1988 |
+
},
|
| 1989 |
+
"total_flos": 1.2756662754476032e+19,
|
| 1990 |
+
"train_batch_size": 4,
|
| 1991 |
+
"trial_name": null,
|
| 1992 |
+
"trial_params": null
|
| 1993 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e95a8f5e7f8a0f6f3e1f415e9606de2bf6f80315b55f9012ea921093e8d88264
|
| 3 |
+
size 6520
|
zero_to_fp32.py
ADDED
|
@@ -0,0 +1,587 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
#!/usr/bin/env python
|
| 2 |
+
|
| 3 |
+
# Copyright (c) Microsoft Corporation.
|
| 4 |
+
# SPDX-License-Identifier: Apache-2.0
|
| 5 |
+
|
| 6 |
+
# DeepSpeed Team
|
| 7 |
+
|
| 8 |
+
# This script extracts fp32 consolidated weights from a zero 1, 2 and 3 DeepSpeed checkpoints. It gets
|
| 9 |
+
# copied into the top level checkpoint dir, so the user can easily do the conversion at any point in
|
| 10 |
+
# the future. Once extracted, the weights don't require DeepSpeed and can be used in any
|
| 11 |
+
# application.
|
| 12 |
+
#
|
| 13 |
+
# example: python zero_to_fp32.py . pytorch_model.bin
|
| 14 |
+
|
| 15 |
+
import argparse
|
| 16 |
+
import torch
|
| 17 |
+
import glob
|
| 18 |
+
import math
|
| 19 |
+
import os
|
| 20 |
+
import re
|
| 21 |
+
from collections import OrderedDict
|
| 22 |
+
from dataclasses import dataclass
|
| 23 |
+
|
| 24 |
+
# while this script doesn't use deepspeed to recover data, since the checkpoints are pickled with
|
| 25 |
+
# DeepSpeed data structures it has to be available in the current python environment.
|
| 26 |
+
from deepspeed.utils import logger
|
| 27 |
+
from deepspeed.checkpoint.constants import (DS_VERSION, OPTIMIZER_STATE_DICT, SINGLE_PARTITION_OF_FP32_GROUPS,
|
| 28 |
+
FP32_FLAT_GROUPS, ZERO_STAGE, PARTITION_COUNT, PARAM_SHAPES, BUFFER_NAMES,
|
| 29 |
+
FROZEN_PARAM_SHAPES, FROZEN_PARAM_FRAGMENTS)
|
| 30 |
+
|
| 31 |
+
|
| 32 |
+
@dataclass
|
| 33 |
+
class zero_model_state:
|
| 34 |
+
buffers: dict()
|
| 35 |
+
param_shapes: dict()
|
| 36 |
+
shared_params: list
|
| 37 |
+
ds_version: int
|
| 38 |
+
frozen_param_shapes: dict()
|
| 39 |
+
frozen_param_fragments: dict()
|
| 40 |
+
|
| 41 |
+
|
| 42 |
+
debug = 0
|
| 43 |
+
|
| 44 |
+
# load to cpu
|
| 45 |
+
device = torch.device('cpu')
|
| 46 |
+
|
| 47 |
+
|
| 48 |
+
def atoi(text):
|
| 49 |
+
return int(text) if text.isdigit() else text
|
| 50 |
+
|
| 51 |
+
|
| 52 |
+
def natural_keys(text):
|
| 53 |
+
'''
|
| 54 |
+
alist.sort(key=natural_keys) sorts in human order
|
| 55 |
+
http://nedbatchelder.com/blog/200712/human_sorting.html
|
| 56 |
+
(See Toothy's implementation in the comments)
|
| 57 |
+
'''
|
| 58 |
+
return [atoi(c) for c in re.split(r'(\d+)', text)]
|
| 59 |
+
|
| 60 |
+
|
| 61 |
+
def get_model_state_file(checkpoint_dir, zero_stage):
|
| 62 |
+
if not os.path.isdir(checkpoint_dir):
|
| 63 |
+
raise FileNotFoundError(f"Directory '{checkpoint_dir}' doesn't exist")
|
| 64 |
+
|
| 65 |
+
# there should be only one file
|
| 66 |
+
if zero_stage <= 2:
|
| 67 |
+
file = os.path.join(checkpoint_dir, "mp_rank_00_model_states.pt")
|
| 68 |
+
elif zero_stage == 3:
|
| 69 |
+
file = os.path.join(checkpoint_dir, "zero_pp_rank_0_mp_rank_00_model_states.pt")
|
| 70 |
+
|
| 71 |
+
if not os.path.exists(file):
|
| 72 |
+
raise FileNotFoundError(f"can't find model states file at '{file}'")
|
| 73 |
+
|
| 74 |
+
return file
|
| 75 |
+
|
| 76 |
+
|
| 77 |
+
def get_checkpoint_files(checkpoint_dir, glob_pattern):
|
| 78 |
+
# XXX: need to test that this simple glob rule works for multi-node setup too
|
| 79 |
+
ckpt_files = sorted(glob.glob(os.path.join(checkpoint_dir, glob_pattern)), key=natural_keys)
|
| 80 |
+
|
| 81 |
+
if len(ckpt_files) == 0:
|
| 82 |
+
raise FileNotFoundError(f"can't find {glob_pattern} files in directory '{checkpoint_dir}'")
|
| 83 |
+
|
| 84 |
+
return ckpt_files
|
| 85 |
+
|
| 86 |
+
|
| 87 |
+
def get_optim_files(checkpoint_dir):
|
| 88 |
+
return get_checkpoint_files(checkpoint_dir, "*_optim_states.pt")
|
| 89 |
+
|
| 90 |
+
|
| 91 |
+
def get_model_state_files(checkpoint_dir):
|
| 92 |
+
return get_checkpoint_files(checkpoint_dir, "*_model_states.pt")
|
| 93 |
+
|
| 94 |
+
|
| 95 |
+
def parse_model_states(files):
|
| 96 |
+
zero_model_states = []
|
| 97 |
+
for file in files:
|
| 98 |
+
state_dict = torch.load(file, map_location=device)
|
| 99 |
+
|
| 100 |
+
if BUFFER_NAMES not in state_dict:
|
| 101 |
+
raise ValueError(f"{file} is not a model state checkpoint")
|
| 102 |
+
buffer_names = state_dict[BUFFER_NAMES]
|
| 103 |
+
if debug:
|
| 104 |
+
print("Found buffers:", buffer_names)
|
| 105 |
+
|
| 106 |
+
# recover just the buffers while restoring them to fp32 if they were saved in fp16
|
| 107 |
+
buffers = {k: v.float() for k, v in state_dict["module"].items() if k in buffer_names}
|
| 108 |
+
param_shapes = state_dict[PARAM_SHAPES]
|
| 109 |
+
|
| 110 |
+
# collect parameters that are included in param_shapes
|
| 111 |
+
param_names = []
|
| 112 |
+
for s in param_shapes:
|
| 113 |
+
for name in s.keys():
|
| 114 |
+
param_names.append(name)
|
| 115 |
+
|
| 116 |
+
# update with frozen parameters
|
| 117 |
+
frozen_param_shapes = state_dict.get(FROZEN_PARAM_SHAPES, None)
|
| 118 |
+
if frozen_param_shapes is not None:
|
| 119 |
+
if debug:
|
| 120 |
+
print(f"Found frozen_param_shapes: {frozen_param_shapes}")
|
| 121 |
+
param_names += list(frozen_param_shapes.keys())
|
| 122 |
+
|
| 123 |
+
# handle shared params
|
| 124 |
+
shared_params = [[k, v] for k, v in state_dict["shared_params"].items()]
|
| 125 |
+
|
| 126 |
+
ds_version = state_dict.get(DS_VERSION, None)
|
| 127 |
+
|
| 128 |
+
frozen_param_fragments = state_dict.get(FROZEN_PARAM_FRAGMENTS, None)
|
| 129 |
+
|
| 130 |
+
z_model_state = zero_model_state(buffers=buffers,
|
| 131 |
+
param_shapes=param_shapes,
|
| 132 |
+
shared_params=shared_params,
|
| 133 |
+
ds_version=ds_version,
|
| 134 |
+
frozen_param_shapes=frozen_param_shapes,
|
| 135 |
+
frozen_param_fragments=frozen_param_fragments)
|
| 136 |
+
zero_model_states.append(z_model_state)
|
| 137 |
+
|
| 138 |
+
return zero_model_states
|
| 139 |
+
|
| 140 |
+
|
| 141 |
+
def parse_optim_states(files, ds_checkpoint_dir):
|
| 142 |
+
|
| 143 |
+
total_files = len(files)
|
| 144 |
+
state_dicts = []
|
| 145 |
+
for f in files:
|
| 146 |
+
state_dict = torch.load(f, map_location=device)
|
| 147 |
+
# immediately discard the potentially huge 2 optimizer states as we only care for fp32 master weights
|
| 148 |
+
# and also handle the case where it was already removed by another helper script
|
| 149 |
+
state_dict["optimizer_state_dict"].pop("optimizer_state_dict", None)
|
| 150 |
+
state_dicts.append(state_dict)
|
| 151 |
+
|
| 152 |
+
if not ZERO_STAGE in state_dicts[0][OPTIMIZER_STATE_DICT]:
|
| 153 |
+
raise ValueError(f"{files[0]} is not a zero checkpoint")
|
| 154 |
+
zero_stage = state_dicts[0][OPTIMIZER_STATE_DICT][ZERO_STAGE]
|
| 155 |
+
world_size = state_dicts[0][OPTIMIZER_STATE_DICT][PARTITION_COUNT]
|
| 156 |
+
|
| 157 |
+
# For ZeRO-2 each param group can have different partition_count as data parallelism for expert
|
| 158 |
+
# parameters can be different from data parallelism for non-expert parameters. So we can just
|
| 159 |
+
# use the max of the partition_count to get the dp world_size.
|
| 160 |
+
|
| 161 |
+
if type(world_size) is list:
|
| 162 |
+
world_size = max(world_size)
|
| 163 |
+
|
| 164 |
+
if world_size != total_files:
|
| 165 |
+
raise ValueError(
|
| 166 |
+
f"Expected {world_size} of '*_optim_states.pt' under '{ds_checkpoint_dir}' but found {total_files} files. "
|
| 167 |
+
"Possibly due to an overwrite of an old checkpoint, or a checkpoint didn't get saved by one or more processes."
|
| 168 |
+
)
|
| 169 |
+
|
| 170 |
+
# the groups are named differently in each stage
|
| 171 |
+
if zero_stage <= 2:
|
| 172 |
+
fp32_groups_key = SINGLE_PARTITION_OF_FP32_GROUPS
|
| 173 |
+
elif zero_stage == 3:
|
| 174 |
+
fp32_groups_key = FP32_FLAT_GROUPS
|
| 175 |
+
else:
|
| 176 |
+
raise ValueError(f"unknown zero stage {zero_stage}")
|
| 177 |
+
|
| 178 |
+
if zero_stage <= 2:
|
| 179 |
+
fp32_flat_groups = [state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key] for i in range(len(state_dicts))]
|
| 180 |
+
elif zero_stage == 3:
|
| 181 |
+
# if there is more than one param group, there will be multiple flattened tensors - one
|
| 182 |
+
# flattened tensor per group - for simplicity merge them into a single tensor
|
| 183 |
+
#
|
| 184 |
+
# XXX: could make the script more memory efficient for when there are multiple groups - it
|
| 185 |
+
# will require matching the sub-lists of param_shapes for each param group flattened tensor
|
| 186 |
+
|
| 187 |
+
fp32_flat_groups = [
|
| 188 |
+
torch.cat(state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key], 0) for i in range(len(state_dicts))
|
| 189 |
+
]
|
| 190 |
+
|
| 191 |
+
return zero_stage, world_size, fp32_flat_groups
|
| 192 |
+
|
| 193 |
+
|
| 194 |
+
def _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir):
|
| 195 |
+
"""
|
| 196 |
+
Returns fp32 state_dict reconstructed from ds checkpoint
|
| 197 |
+
|
| 198 |
+
Args:
|
| 199 |
+
- ``ds_checkpoint_dir``: path to the deepspeed checkpoint folder (where the optimizer files are)
|
| 200 |
+
|
| 201 |
+
"""
|
| 202 |
+
print(f"Processing zero checkpoint '{ds_checkpoint_dir}'")
|
| 203 |
+
|
| 204 |
+
optim_files = get_optim_files(ds_checkpoint_dir)
|
| 205 |
+
zero_stage, world_size, fp32_flat_groups = parse_optim_states(optim_files, ds_checkpoint_dir)
|
| 206 |
+
print(f"Detected checkpoint of type zero stage {zero_stage}, world_size: {world_size}")
|
| 207 |
+
|
| 208 |
+
model_files = get_model_state_files(ds_checkpoint_dir)
|
| 209 |
+
|
| 210 |
+
zero_model_states = parse_model_states(model_files)
|
| 211 |
+
print(f'Parsing checkpoint created by deepspeed=={zero_model_states[0].ds_version}')
|
| 212 |
+
|
| 213 |
+
if zero_stage <= 2:
|
| 214 |
+
return _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states)
|
| 215 |
+
elif zero_stage == 3:
|
| 216 |
+
return _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states)
|
| 217 |
+
|
| 218 |
+
|
| 219 |
+
def _zero2_merge_frozen_params(state_dict, zero_model_states):
|
| 220 |
+
if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0:
|
| 221 |
+
return
|
| 222 |
+
|
| 223 |
+
frozen_param_shapes = zero_model_states[0].frozen_param_shapes
|
| 224 |
+
frozen_param_fragments = zero_model_states[0].frozen_param_fragments
|
| 225 |
+
|
| 226 |
+
if debug:
|
| 227 |
+
num_elem = sum(s.numel() for s in frozen_param_shapes.values())
|
| 228 |
+
print(f'rank 0: {FROZEN_PARAM_SHAPES}.numel = {num_elem}')
|
| 229 |
+
|
| 230 |
+
wanted_params = len(frozen_param_shapes)
|
| 231 |
+
wanted_numel = sum(s.numel() for s in frozen_param_shapes.values())
|
| 232 |
+
avail_numel = sum([p.numel() for p in frozen_param_fragments.values()])
|
| 233 |
+
print(f'Frozen params: Have {avail_numel} numels to process.')
|
| 234 |
+
print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params')
|
| 235 |
+
|
| 236 |
+
total_params = 0
|
| 237 |
+
total_numel = 0
|
| 238 |
+
for name, shape in frozen_param_shapes.items():
|
| 239 |
+
total_params += 1
|
| 240 |
+
unpartitioned_numel = shape.numel()
|
| 241 |
+
total_numel += unpartitioned_numel
|
| 242 |
+
|
| 243 |
+
state_dict[name] = frozen_param_fragments[name]
|
| 244 |
+
|
| 245 |
+
if debug:
|
| 246 |
+
print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ")
|
| 247 |
+
|
| 248 |
+
print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements")
|
| 249 |
+
|
| 250 |
+
|
| 251 |
+
def _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states):
|
| 252 |
+
param_shapes = zero_model_states[0].param_shapes
|
| 253 |
+
|
| 254 |
+
# Reconstruction protocol:
|
| 255 |
+
#
|
| 256 |
+
# XXX: document this
|
| 257 |
+
|
| 258 |
+
if debug:
|
| 259 |
+
for i in range(world_size):
|
| 260 |
+
for j in range(len(fp32_flat_groups[0])):
|
| 261 |
+
print(f"{FP32_FLAT_GROUPS}[{i}][{j}].shape={fp32_flat_groups[i][j].shape}")
|
| 262 |
+
|
| 263 |
+
# XXX: memory usage doubles here (zero2)
|
| 264 |
+
num_param_groups = len(fp32_flat_groups[0])
|
| 265 |
+
merged_single_partition_of_fp32_groups = []
|
| 266 |
+
for i in range(num_param_groups):
|
| 267 |
+
merged_partitions = [sd[i] for sd in fp32_flat_groups]
|
| 268 |
+
full_single_fp32_vector = torch.cat(merged_partitions, 0)
|
| 269 |
+
merged_single_partition_of_fp32_groups.append(full_single_fp32_vector)
|
| 270 |
+
avail_numel = sum(
|
| 271 |
+
[full_single_fp32_vector.numel() for full_single_fp32_vector in merged_single_partition_of_fp32_groups])
|
| 272 |
+
|
| 273 |
+
if debug:
|
| 274 |
+
wanted_params = sum([len(shapes) for shapes in param_shapes])
|
| 275 |
+
wanted_numel = sum([sum(shape.numel() for shape in shapes.values()) for shapes in param_shapes])
|
| 276 |
+
# not asserting if there is a mismatch due to possible padding
|
| 277 |
+
print(f"Have {avail_numel} numels to process.")
|
| 278 |
+
print(f"Need {wanted_numel} numels in {wanted_params} params.")
|
| 279 |
+
|
| 280 |
+
# params
|
| 281 |
+
# XXX: for huge models that can't fit into the host's RAM we will have to recode this to support
|
| 282 |
+
# out-of-core computing solution
|
| 283 |
+
total_numel = 0
|
| 284 |
+
total_params = 0
|
| 285 |
+
for shapes, full_single_fp32_vector in zip(param_shapes, merged_single_partition_of_fp32_groups):
|
| 286 |
+
offset = 0
|
| 287 |
+
avail_numel = full_single_fp32_vector.numel()
|
| 288 |
+
for name, shape in shapes.items():
|
| 289 |
+
|
| 290 |
+
unpartitioned_numel = shape.numel()
|
| 291 |
+
total_numel += unpartitioned_numel
|
| 292 |
+
total_params += 1
|
| 293 |
+
|
| 294 |
+
if debug:
|
| 295 |
+
print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ")
|
| 296 |
+
state_dict[name] = full_single_fp32_vector.narrow(0, offset, unpartitioned_numel).view(shape)
|
| 297 |
+
offset += unpartitioned_numel
|
| 298 |
+
|
| 299 |
+
# Z2 started to align to 2*world_size to improve nccl performance. Therefore both offset and
|
| 300 |
+
# avail_numel can differ by anywhere between 0..2*world_size. Due to two unrelated complex
|
| 301 |
+
# paddings performed in the code it's almost impossible to predict the exact numbers w/o the
|
| 302 |
+
# live optimizer object, so we are checking that the numbers are within the right range
|
| 303 |
+
align_to = 2 * world_size
|
| 304 |
+
|
| 305 |
+
def zero2_align(x):
|
| 306 |
+
return align_to * math.ceil(x / align_to)
|
| 307 |
+
|
| 308 |
+
if debug:
|
| 309 |
+
print(f"original offset={offset}, avail_numel={avail_numel}")
|
| 310 |
+
|
| 311 |
+
offset = zero2_align(offset)
|
| 312 |
+
avail_numel = zero2_align(avail_numel)
|
| 313 |
+
|
| 314 |
+
if debug:
|
| 315 |
+
print(f"aligned offset={offset}, avail_numel={avail_numel}")
|
| 316 |
+
|
| 317 |
+
# Sanity check
|
| 318 |
+
if offset != avail_numel:
|
| 319 |
+
raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong")
|
| 320 |
+
|
| 321 |
+
print(f"Reconstructed fp32 state dict with {total_params} params {total_numel} elements")
|
| 322 |
+
|
| 323 |
+
|
| 324 |
+
def _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states):
|
| 325 |
+
state_dict = OrderedDict()
|
| 326 |
+
|
| 327 |
+
# buffers
|
| 328 |
+
buffers = zero_model_states[0].buffers
|
| 329 |
+
state_dict.update(buffers)
|
| 330 |
+
if debug:
|
| 331 |
+
print(f"added {len(buffers)} buffers")
|
| 332 |
+
|
| 333 |
+
_zero2_merge_frozen_params(state_dict, zero_model_states)
|
| 334 |
+
|
| 335 |
+
_zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states)
|
| 336 |
+
|
| 337 |
+
# recover shared parameters
|
| 338 |
+
for pair in zero_model_states[0].shared_params:
|
| 339 |
+
if pair[1] in state_dict:
|
| 340 |
+
state_dict[pair[0]] = state_dict[pair[1]]
|
| 341 |
+
|
| 342 |
+
return state_dict
|
| 343 |
+
|
| 344 |
+
|
| 345 |
+
def zero3_partitioned_param_info(unpartitioned_numel, world_size):
|
| 346 |
+
remainder = unpartitioned_numel % world_size
|
| 347 |
+
padding_numel = (world_size - remainder) if remainder else 0
|
| 348 |
+
partitioned_numel = math.ceil(unpartitioned_numel / world_size)
|
| 349 |
+
return partitioned_numel, padding_numel
|
| 350 |
+
|
| 351 |
+
|
| 352 |
+
def _zero3_merge_frozen_params(state_dict, world_size, zero_model_states):
|
| 353 |
+
if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0:
|
| 354 |
+
return
|
| 355 |
+
|
| 356 |
+
if debug:
|
| 357 |
+
for i in range(world_size):
|
| 358 |
+
num_elem = sum(s.numel() for s in zero_model_states[i].frozen_param_fragments.values())
|
| 359 |
+
print(f'rank {i}: {FROZEN_PARAM_SHAPES}.numel = {num_elem}')
|
| 360 |
+
|
| 361 |
+
frozen_param_shapes = zero_model_states[0].frozen_param_shapes
|
| 362 |
+
wanted_params = len(frozen_param_shapes)
|
| 363 |
+
wanted_numel = sum(s.numel() for s in frozen_param_shapes.values())
|
| 364 |
+
avail_numel = sum([p.numel() for p in zero_model_states[0].frozen_param_fragments.values()]) * world_size
|
| 365 |
+
print(f'Frozen params: Have {avail_numel} numels to process.')
|
| 366 |
+
print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params')
|
| 367 |
+
|
| 368 |
+
total_params = 0
|
| 369 |
+
total_numel = 0
|
| 370 |
+
for name, shape in zero_model_states[0].frozen_param_shapes.items():
|
| 371 |
+
total_params += 1
|
| 372 |
+
unpartitioned_numel = shape.numel()
|
| 373 |
+
total_numel += unpartitioned_numel
|
| 374 |
+
|
| 375 |
+
param_frags = tuple(model_state.frozen_param_fragments[name] for model_state in zero_model_states)
|
| 376 |
+
state_dict[name] = torch.cat(param_frags, 0).narrow(0, 0, unpartitioned_numel).view(shape)
|
| 377 |
+
|
| 378 |
+
partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size)
|
| 379 |
+
|
| 380 |
+
if debug:
|
| 381 |
+
print(
|
| 382 |
+
f"Frozen params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}"
|
| 383 |
+
)
|
| 384 |
+
|
| 385 |
+
print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements")
|
| 386 |
+
|
| 387 |
+
|
| 388 |
+
def _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states):
|
| 389 |
+
param_shapes = zero_model_states[0].param_shapes
|
| 390 |
+
avail_numel = fp32_flat_groups[0].numel() * world_size
|
| 391 |
+
# Reconstruction protocol: For zero3 we need to zip the partitions together at boundary of each
|
| 392 |
+
# param, re-consolidating each param, while dealing with padding if any
|
| 393 |
+
|
| 394 |
+
# merge list of dicts, preserving order
|
| 395 |
+
param_shapes = {k: v for d in param_shapes for k, v in d.items()}
|
| 396 |
+
|
| 397 |
+
if debug:
|
| 398 |
+
for i in range(world_size):
|
| 399 |
+
print(f"{FP32_FLAT_GROUPS}[{i}].shape={fp32_flat_groups[i].shape}")
|
| 400 |
+
|
| 401 |
+
wanted_params = len(param_shapes)
|
| 402 |
+
wanted_numel = sum(shape.numel() for shape in param_shapes.values())
|
| 403 |
+
# not asserting if there is a mismatch due to possible padding
|
| 404 |
+
avail_numel = fp32_flat_groups[0].numel() * world_size
|
| 405 |
+
print(f"Trainable params: Have {avail_numel} numels to process.")
|
| 406 |
+
print(f"Trainable params: Need {wanted_numel} numels in {wanted_params} params.")
|
| 407 |
+
|
| 408 |
+
# params
|
| 409 |
+
# XXX: for huge models that can't fit into the host's RAM we will have to recode this to support
|
| 410 |
+
# out-of-core computing solution
|
| 411 |
+
offset = 0
|
| 412 |
+
total_numel = 0
|
| 413 |
+
total_params = 0
|
| 414 |
+
for name, shape in param_shapes.items():
|
| 415 |
+
|
| 416 |
+
unpartitioned_numel = shape.numel()
|
| 417 |
+
total_numel += unpartitioned_numel
|
| 418 |
+
total_params += 1
|
| 419 |
+
|
| 420 |
+
partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size)
|
| 421 |
+
|
| 422 |
+
if debug:
|
| 423 |
+
print(
|
| 424 |
+
f"Trainable params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}"
|
| 425 |
+
)
|
| 426 |
+
|
| 427 |
+
# XXX: memory usage doubles here
|
| 428 |
+
state_dict[name] = torch.cat(
|
| 429 |
+
tuple(fp32_flat_groups[i].narrow(0, offset, partitioned_numel) for i in range(world_size)),
|
| 430 |
+
0).narrow(0, 0, unpartitioned_numel).view(shape)
|
| 431 |
+
offset += partitioned_numel
|
| 432 |
+
|
| 433 |
+
offset *= world_size
|
| 434 |
+
|
| 435 |
+
# Sanity check
|
| 436 |
+
if offset != avail_numel:
|
| 437 |
+
raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong")
|
| 438 |
+
|
| 439 |
+
print(f"Reconstructed Trainable fp32 state dict with {total_params} params {total_numel} elements")
|
| 440 |
+
|
| 441 |
+
|
| 442 |
+
def _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states):
|
| 443 |
+
state_dict = OrderedDict()
|
| 444 |
+
|
| 445 |
+
# buffers
|
| 446 |
+
buffers = zero_model_states[0].buffers
|
| 447 |
+
state_dict.update(buffers)
|
| 448 |
+
if debug:
|
| 449 |
+
print(f"added {len(buffers)} buffers")
|
| 450 |
+
|
| 451 |
+
_zero3_merge_frozen_params(state_dict, world_size, zero_model_states)
|
| 452 |
+
|
| 453 |
+
_zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states)
|
| 454 |
+
|
| 455 |
+
# recover shared parameters
|
| 456 |
+
for pair in zero_model_states[0].shared_params:
|
| 457 |
+
if pair[1] in state_dict:
|
| 458 |
+
state_dict[pair[0]] = state_dict[pair[1]]
|
| 459 |
+
|
| 460 |
+
return state_dict
|
| 461 |
+
|
| 462 |
+
|
| 463 |
+
def get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag=None):
|
| 464 |
+
"""
|
| 465 |
+
Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated state_dict that can be loaded with
|
| 466 |
+
``load_state_dict()`` and used for training without DeepSpeed or shared with others, for example
|
| 467 |
+
via a model hub.
|
| 468 |
+
|
| 469 |
+
Args:
|
| 470 |
+
- ``checkpoint_dir``: path to the desired checkpoint folder
|
| 471 |
+
- ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in 'latest' file. e.g., ``global_step14``
|
| 472 |
+
|
| 473 |
+
Returns:
|
| 474 |
+
- pytorch ``state_dict``
|
| 475 |
+
|
| 476 |
+
Note: this approach may not work if your application doesn't have sufficient free CPU memory and
|
| 477 |
+
you may need to use the offline approach using the ``zero_to_fp32.py`` script that is saved with
|
| 478 |
+
the checkpoint.
|
| 479 |
+
|
| 480 |
+
A typical usage might be ::
|
| 481 |
+
|
| 482 |
+
from deepspeed.utils.zero_to_fp32 import get_fp32_state_dict_from_zero_checkpoint
|
| 483 |
+
# do the training and checkpoint saving
|
| 484 |
+
state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir) # already on cpu
|
| 485 |
+
model = model.cpu() # move to cpu
|
| 486 |
+
model.load_state_dict(state_dict)
|
| 487 |
+
# submit to model hub or save the model to share with others
|
| 488 |
+
|
| 489 |
+
In this example the ``model`` will no longer be usable in the deepspeed context of the same
|
| 490 |
+
application. i.e. you will need to re-initialize the deepspeed engine, since
|
| 491 |
+
``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it.
|
| 492 |
+
|
| 493 |
+
If you want it all done for you, use ``load_state_dict_from_zero_checkpoint`` instead.
|
| 494 |
+
|
| 495 |
+
"""
|
| 496 |
+
if tag is None:
|
| 497 |
+
latest_path = os.path.join(checkpoint_dir, 'latest')
|
| 498 |
+
if os.path.isfile(latest_path):
|
| 499 |
+
with open(latest_path, 'r') as fd:
|
| 500 |
+
tag = fd.read().strip()
|
| 501 |
+
else:
|
| 502 |
+
raise ValueError(f"Unable to find 'latest' file at {latest_path}")
|
| 503 |
+
|
| 504 |
+
ds_checkpoint_dir = os.path.join(checkpoint_dir, tag)
|
| 505 |
+
|
| 506 |
+
if not os.path.isdir(ds_checkpoint_dir):
|
| 507 |
+
raise FileNotFoundError(f"Directory '{ds_checkpoint_dir}' doesn't exist")
|
| 508 |
+
|
| 509 |
+
return _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir)
|
| 510 |
+
|
| 511 |
+
|
| 512 |
+
def convert_zero_checkpoint_to_fp32_state_dict(checkpoint_dir, output_file, tag=None):
|
| 513 |
+
"""
|
| 514 |
+
Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict`` file that can be
|
| 515 |
+
loaded with ``torch.load(file)`` + ``load_state_dict()`` and used for training without DeepSpeed.
|
| 516 |
+
|
| 517 |
+
Args:
|
| 518 |
+
- ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``)
|
| 519 |
+
- ``output_file``: path to the pytorch fp32 state_dict output file (e.g. path/pytorch_model.bin)
|
| 520 |
+
- ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14``
|
| 521 |
+
"""
|
| 522 |
+
|
| 523 |
+
state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag)
|
| 524 |
+
print(f"Saving fp32 state dict to {output_file}")
|
| 525 |
+
torch.save(state_dict, output_file)
|
| 526 |
+
|
| 527 |
+
|
| 528 |
+
def load_state_dict_from_zero_checkpoint(model, checkpoint_dir, tag=None):
|
| 529 |
+
"""
|
| 530 |
+
1. Put the provided model to cpu
|
| 531 |
+
2. Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict``
|
| 532 |
+
3. Load it into the provided model
|
| 533 |
+
|
| 534 |
+
Args:
|
| 535 |
+
- ``model``: the model object to update
|
| 536 |
+
- ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``)
|
| 537 |
+
- ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14``
|
| 538 |
+
|
| 539 |
+
Returns:
|
| 540 |
+
- ``model`: modified model
|
| 541 |
+
|
| 542 |
+
Make sure you have plenty of CPU memory available before you call this function. If you don't
|
| 543 |
+
have enough use the ``zero_to_fp32.py`` utility to do the conversion. You will find it
|
| 544 |
+
conveniently placed for you in the checkpoint folder.
|
| 545 |
+
|
| 546 |
+
A typical usage might be ::
|
| 547 |
+
|
| 548 |
+
from deepspeed.utils.zero_to_fp32 import load_state_dict_from_zero_checkpoint
|
| 549 |
+
model = load_state_dict_from_zero_checkpoint(trainer.model, checkpoint_dir)
|
| 550 |
+
# submit to model hub or save the model to share with others
|
| 551 |
+
|
| 552 |
+
Note, that once this was run, the ``model`` will no longer be usable in the deepspeed context
|
| 553 |
+
of the same application. i.e. you will need to re-initialize the deepspeed engine, since
|
| 554 |
+
``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it.
|
| 555 |
+
|
| 556 |
+
"""
|
| 557 |
+
logger.info(f"Extracting fp32 weights")
|
| 558 |
+
state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag)
|
| 559 |
+
|
| 560 |
+
logger.info(f"Overwriting model with fp32 weights")
|
| 561 |
+
model = model.cpu()
|
| 562 |
+
model.load_state_dict(state_dict, strict=False)
|
| 563 |
+
|
| 564 |
+
return model
|
| 565 |
+
|
| 566 |
+
|
| 567 |
+
if __name__ == "__main__":
|
| 568 |
+
|
| 569 |
+
parser = argparse.ArgumentParser()
|
| 570 |
+
parser.add_argument("checkpoint_dir",
|
| 571 |
+
type=str,
|
| 572 |
+
help="path to the desired checkpoint folder, e.g., path/checkpoint-12")
|
| 573 |
+
parser.add_argument(
|
| 574 |
+
"output_file",
|
| 575 |
+
type=str,
|
| 576 |
+
help="path to the pytorch fp32 state_dict output file (e.g. path/checkpoint-12/pytorch_model.bin)")
|
| 577 |
+
parser.add_argument("-t",
|
| 578 |
+
"--tag",
|
| 579 |
+
type=str,
|
| 580 |
+
default=None,
|
| 581 |
+
help="checkpoint tag used as a unique identifier for checkpoint. e.g., global_step1")
|
| 582 |
+
parser.add_argument("-d", "--debug", action='store_true', help="enable debug")
|
| 583 |
+
args = parser.parse_args()
|
| 584 |
+
|
| 585 |
+
debug = args.debug
|
| 586 |
+
|
| 587 |
+
convert_zero_checkpoint_to_fp32_state_dict(args.checkpoint_dir, args.output_file, tag=args.tag)
|