Upload folder using huggingface_hub
Browse files- README.md +202 -0
- adapter_config.json +37 -0
- adapter_model.safetensors +3 -0
- global_step3310/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt +3 -0
- global_step3310/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt +3 -0
- global_step3310/zero_pp_rank_0_mp_rank_00_model_states.pt +3 -0
- global_step3310/zero_pp_rank_1_mp_rank_00_model_states.pt +3 -0
- latest +1 -0
- rng_state_0.pth +3 -0
- rng_state_1.pth +3 -0
- special_tokens_map.json +24 -0
- tokenizer.json +0 -0
- tokenizer_config.json +42 -0
- trainer_state.json +2350 -0
- training_args.bin +3 -0
- zero_to_fp32.py +760 -0
README.md
ADDED
|
@@ -0,0 +1,202 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
base_model: meta-llama/Llama-2-7b-hf
|
| 3 |
+
library_name: peft
|
| 4 |
+
---
|
| 5 |
+
|
| 6 |
+
# Model Card for Model ID
|
| 7 |
+
|
| 8 |
+
<!-- Provide a quick summary of what the model is/does. -->
|
| 9 |
+
|
| 10 |
+
|
| 11 |
+
|
| 12 |
+
## Model Details
|
| 13 |
+
|
| 14 |
+
### Model Description
|
| 15 |
+
|
| 16 |
+
<!-- Provide a longer summary of what this model is. -->
|
| 17 |
+
|
| 18 |
+
|
| 19 |
+
|
| 20 |
+
- **Developed by:** [More Information Needed]
|
| 21 |
+
- **Funded by [optional]:** [More Information Needed]
|
| 22 |
+
- **Shared by [optional]:** [More Information Needed]
|
| 23 |
+
- **Model type:** [More Information Needed]
|
| 24 |
+
- **Language(s) (NLP):** [More Information Needed]
|
| 25 |
+
- **License:** [More Information Needed]
|
| 26 |
+
- **Finetuned from model [optional]:** [More Information Needed]
|
| 27 |
+
|
| 28 |
+
### Model Sources [optional]
|
| 29 |
+
|
| 30 |
+
<!-- Provide the basic links for the model. -->
|
| 31 |
+
|
| 32 |
+
- **Repository:** [More Information Needed]
|
| 33 |
+
- **Paper [optional]:** [More Information Needed]
|
| 34 |
+
- **Demo [optional]:** [More Information Needed]
|
| 35 |
+
|
| 36 |
+
## Uses
|
| 37 |
+
|
| 38 |
+
<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
|
| 39 |
+
|
| 40 |
+
### Direct Use
|
| 41 |
+
|
| 42 |
+
<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
|
| 43 |
+
|
| 44 |
+
[More Information Needed]
|
| 45 |
+
|
| 46 |
+
### Downstream Use [optional]
|
| 47 |
+
|
| 48 |
+
<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
|
| 49 |
+
|
| 50 |
+
[More Information Needed]
|
| 51 |
+
|
| 52 |
+
### Out-of-Scope Use
|
| 53 |
+
|
| 54 |
+
<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
|
| 55 |
+
|
| 56 |
+
[More Information Needed]
|
| 57 |
+
|
| 58 |
+
## Bias, Risks, and Limitations
|
| 59 |
+
|
| 60 |
+
<!-- This section is meant to convey both technical and sociotechnical limitations. -->
|
| 61 |
+
|
| 62 |
+
[More Information Needed]
|
| 63 |
+
|
| 64 |
+
### Recommendations
|
| 65 |
+
|
| 66 |
+
<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
|
| 67 |
+
|
| 68 |
+
Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
|
| 69 |
+
|
| 70 |
+
## How to Get Started with the Model
|
| 71 |
+
|
| 72 |
+
Use the code below to get started with the model.
|
| 73 |
+
|
| 74 |
+
[More Information Needed]
|
| 75 |
+
|
| 76 |
+
## Training Details
|
| 77 |
+
|
| 78 |
+
### Training Data
|
| 79 |
+
|
| 80 |
+
<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
|
| 81 |
+
|
| 82 |
+
[More Information Needed]
|
| 83 |
+
|
| 84 |
+
### Training Procedure
|
| 85 |
+
|
| 86 |
+
<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
|
| 87 |
+
|
| 88 |
+
#### Preprocessing [optional]
|
| 89 |
+
|
| 90 |
+
[More Information Needed]
|
| 91 |
+
|
| 92 |
+
|
| 93 |
+
#### Training Hyperparameters
|
| 94 |
+
|
| 95 |
+
- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
|
| 96 |
+
|
| 97 |
+
#### Speeds, Sizes, Times [optional]
|
| 98 |
+
|
| 99 |
+
<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
|
| 100 |
+
|
| 101 |
+
[More Information Needed]
|
| 102 |
+
|
| 103 |
+
## Evaluation
|
| 104 |
+
|
| 105 |
+
<!-- This section describes the evaluation protocols and provides the results. -->
|
| 106 |
+
|
| 107 |
+
### Testing Data, Factors & Metrics
|
| 108 |
+
|
| 109 |
+
#### Testing Data
|
| 110 |
+
|
| 111 |
+
<!-- This should link to a Dataset Card if possible. -->
|
| 112 |
+
|
| 113 |
+
[More Information Needed]
|
| 114 |
+
|
| 115 |
+
#### Factors
|
| 116 |
+
|
| 117 |
+
<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
|
| 118 |
+
|
| 119 |
+
[More Information Needed]
|
| 120 |
+
|
| 121 |
+
#### Metrics
|
| 122 |
+
|
| 123 |
+
<!-- These are the evaluation metrics being used, ideally with a description of why. -->
|
| 124 |
+
|
| 125 |
+
[More Information Needed]
|
| 126 |
+
|
| 127 |
+
### Results
|
| 128 |
+
|
| 129 |
+
[More Information Needed]
|
| 130 |
+
|
| 131 |
+
#### Summary
|
| 132 |
+
|
| 133 |
+
|
| 134 |
+
|
| 135 |
+
## Model Examination [optional]
|
| 136 |
+
|
| 137 |
+
<!-- Relevant interpretability work for the model goes here -->
|
| 138 |
+
|
| 139 |
+
[More Information Needed]
|
| 140 |
+
|
| 141 |
+
## Environmental Impact
|
| 142 |
+
|
| 143 |
+
<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
|
| 144 |
+
|
| 145 |
+
Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
|
| 146 |
+
|
| 147 |
+
- **Hardware Type:** [More Information Needed]
|
| 148 |
+
- **Hours used:** [More Information Needed]
|
| 149 |
+
- **Cloud Provider:** [More Information Needed]
|
| 150 |
+
- **Compute Region:** [More Information Needed]
|
| 151 |
+
- **Carbon Emitted:** [More Information Needed]
|
| 152 |
+
|
| 153 |
+
## Technical Specifications [optional]
|
| 154 |
+
|
| 155 |
+
### Model Architecture and Objective
|
| 156 |
+
|
| 157 |
+
[More Information Needed]
|
| 158 |
+
|
| 159 |
+
### Compute Infrastructure
|
| 160 |
+
|
| 161 |
+
[More Information Needed]
|
| 162 |
+
|
| 163 |
+
#### Hardware
|
| 164 |
+
|
| 165 |
+
[More Information Needed]
|
| 166 |
+
|
| 167 |
+
#### Software
|
| 168 |
+
|
| 169 |
+
[More Information Needed]
|
| 170 |
+
|
| 171 |
+
## Citation [optional]
|
| 172 |
+
|
| 173 |
+
<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
|
| 174 |
+
|
| 175 |
+
**BibTeX:**
|
| 176 |
+
|
| 177 |
+
[More Information Needed]
|
| 178 |
+
|
| 179 |
+
**APA:**
|
| 180 |
+
|
| 181 |
+
[More Information Needed]
|
| 182 |
+
|
| 183 |
+
## Glossary [optional]
|
| 184 |
+
|
| 185 |
+
<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
|
| 186 |
+
|
| 187 |
+
[More Information Needed]
|
| 188 |
+
|
| 189 |
+
## More Information [optional]
|
| 190 |
+
|
| 191 |
+
[More Information Needed]
|
| 192 |
+
|
| 193 |
+
## Model Card Authors [optional]
|
| 194 |
+
|
| 195 |
+
[More Information Needed]
|
| 196 |
+
|
| 197 |
+
## Model Card Contact
|
| 198 |
+
|
| 199 |
+
[More Information Needed]
|
| 200 |
+
### Framework versions
|
| 201 |
+
|
| 202 |
+
- PEFT 0.14.0
|
adapter_config.json
ADDED
|
@@ -0,0 +1,37 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"alpha_pattern": {},
|
| 3 |
+
"auto_mapping": null,
|
| 4 |
+
"base_model_name_or_path": "meta-llama/Llama-2-7b-hf",
|
| 5 |
+
"bias": "none",
|
| 6 |
+
"eva_config": null,
|
| 7 |
+
"exclude_modules": null,
|
| 8 |
+
"fan_in_fan_out": false,
|
| 9 |
+
"inference_mode": true,
|
| 10 |
+
"init_lora_weights": true,
|
| 11 |
+
"layer_replication": null,
|
| 12 |
+
"layers_pattern": null,
|
| 13 |
+
"layers_to_transform": null,
|
| 14 |
+
"loftq_config": {},
|
| 15 |
+
"lora_alpha": 64,
|
| 16 |
+
"lora_bias": false,
|
| 17 |
+
"lora_dropout": 0.1,
|
| 18 |
+
"megatron_config": null,
|
| 19 |
+
"megatron_core": "megatron.core",
|
| 20 |
+
"modules_to_save": null,
|
| 21 |
+
"peft_type": "LORA",
|
| 22 |
+
"r": 32,
|
| 23 |
+
"rank_pattern": {},
|
| 24 |
+
"revision": null,
|
| 25 |
+
"target_modules": [
|
| 26 |
+
"down_proj",
|
| 27 |
+
"gate_proj",
|
| 28 |
+
"k_proj",
|
| 29 |
+
"up_proj",
|
| 30 |
+
"q_proj",
|
| 31 |
+
"v_proj",
|
| 32 |
+
"o_proj"
|
| 33 |
+
],
|
| 34 |
+
"task_type": "FEATURE_EXTRACTION",
|
| 35 |
+
"use_dora": false,
|
| 36 |
+
"use_rslora": false
|
| 37 |
+
}
|
adapter_model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:52b1e17579ae7a201521e14c7cf7f18e6daf9b7ce5590919d41c7d87fd057b74
|
| 3 |
+
size 159965640
|
global_step3310/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7b51b2bca142cab6c6a3e360deedbeefae15e9b3b75829d1282afadb59e96c3e
|
| 3 |
+
size 479727769
|
global_step3310/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e258499bc9490b893b6e721c1fa983904aac0ddc17ca9e81de083487cf4342a4
|
| 3 |
+
size 479727769
|
global_step3310/zero_pp_rank_0_mp_rank_00_model_states.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b7b0abbf9825e669e9d08645c5261c88118487430b64f0a3aec260199b163e07
|
| 3 |
+
size 6608028566
|
global_step3310/zero_pp_rank_1_mp_rank_00_model_states.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e1ff1fc44edfb16759ea0664e4ed7c49a0b6272fc526da8c2f2ac4fcd441d4b7
|
| 3 |
+
size 6608028566
|
latest
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
global_step3310
|
rng_state_0.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:176d3aaf52ecb5d14e2795e92bfc4db4bbf1f941c2eac3d3ff17ed1fdff7bb0e
|
| 3 |
+
size 14713
|
rng_state_1.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ccf24e52455ea91253371d1efb37ffc454e4f33bbdda7434be1dd756c3f71f4a
|
| 3 |
+
size 14713
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1,24 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": {
|
| 3 |
+
"content": "<s>",
|
| 4 |
+
"lstrip": false,
|
| 5 |
+
"normalized": false,
|
| 6 |
+
"rstrip": false,
|
| 7 |
+
"single_word": false
|
| 8 |
+
},
|
| 9 |
+
"eos_token": {
|
| 10 |
+
"content": "</s>",
|
| 11 |
+
"lstrip": false,
|
| 12 |
+
"normalized": false,
|
| 13 |
+
"rstrip": false,
|
| 14 |
+
"single_word": false
|
| 15 |
+
},
|
| 16 |
+
"pad_token": "</s>",
|
| 17 |
+
"unk_token": {
|
| 18 |
+
"content": "<unk>",
|
| 19 |
+
"lstrip": false,
|
| 20 |
+
"normalized": false,
|
| 21 |
+
"rstrip": false,
|
| 22 |
+
"single_word": false
|
| 23 |
+
}
|
| 24 |
+
}
|
tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,42 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_bos_token": true,
|
| 3 |
+
"add_eos_token": false,
|
| 4 |
+
"add_prefix_space": null,
|
| 5 |
+
"added_tokens_decoder": {
|
| 6 |
+
"0": {
|
| 7 |
+
"content": "<unk>",
|
| 8 |
+
"lstrip": false,
|
| 9 |
+
"normalized": false,
|
| 10 |
+
"rstrip": false,
|
| 11 |
+
"single_word": false,
|
| 12 |
+
"special": true
|
| 13 |
+
},
|
| 14 |
+
"1": {
|
| 15 |
+
"content": "<s>",
|
| 16 |
+
"lstrip": false,
|
| 17 |
+
"normalized": false,
|
| 18 |
+
"rstrip": false,
|
| 19 |
+
"single_word": false,
|
| 20 |
+
"special": true
|
| 21 |
+
},
|
| 22 |
+
"2": {
|
| 23 |
+
"content": "</s>",
|
| 24 |
+
"lstrip": false,
|
| 25 |
+
"normalized": false,
|
| 26 |
+
"rstrip": false,
|
| 27 |
+
"single_word": false,
|
| 28 |
+
"special": true
|
| 29 |
+
}
|
| 30 |
+
},
|
| 31 |
+
"bos_token": "<s>",
|
| 32 |
+
"clean_up_tokenization_spaces": false,
|
| 33 |
+
"eos_token": "</s>",
|
| 34 |
+
"legacy": false,
|
| 35 |
+
"model_max_length": 1000000000000000019884624838656,
|
| 36 |
+
"pad_token": "</s>",
|
| 37 |
+
"padding_side": "right",
|
| 38 |
+
"sp_model_kwargs": {},
|
| 39 |
+
"tokenizer_class": "LlamaTokenizer",
|
| 40 |
+
"unk_token": "<unk>",
|
| 41 |
+
"use_default_system_prompt": false
|
| 42 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,2350 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.9998489654130795,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 3310,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.0030206917384080953,
|
| 13 |
+
"grad_norm": 6.613835485331974,
|
| 14 |
+
"learning_rate": 4.9999999999999996e-05,
|
| 15 |
+
"loss": 6.2562,
|
| 16 |
+
"step": 10
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.006041383476816191,
|
| 20 |
+
"grad_norm": 4.759929469878327,
|
| 21 |
+
"learning_rate": 6.505149978319905e-05,
|
| 22 |
+
"loss": 3.3105,
|
| 23 |
+
"step": 20
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.009062075215224286,
|
| 27 |
+
"grad_norm": 3.4302773742415433,
|
| 28 |
+
"learning_rate": 7.385606273598311e-05,
|
| 29 |
+
"loss": 1.6293,
|
| 30 |
+
"step": 30
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.012082766953632381,
|
| 34 |
+
"grad_norm": 2.200390666433873,
|
| 35 |
+
"learning_rate": 8.01029995663981e-05,
|
| 36 |
+
"loss": 1.2723,
|
| 37 |
+
"step": 40
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.015103458692040477,
|
| 41 |
+
"grad_norm": 1.9872314808738405,
|
| 42 |
+
"learning_rate": 8.494850021680092e-05,
|
| 43 |
+
"loss": 1.1221,
|
| 44 |
+
"step": 50
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.01812415043044857,
|
| 48 |
+
"grad_norm": 2.0585565038414173,
|
| 49 |
+
"learning_rate": 8.890756251918216e-05,
|
| 50 |
+
"loss": 1.1238,
|
| 51 |
+
"step": 60
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.02114484216885667,
|
| 55 |
+
"grad_norm": 1.5179740873933507,
|
| 56 |
+
"learning_rate": 9.225490200071284e-05,
|
| 57 |
+
"loss": 1.0437,
|
| 58 |
+
"step": 70
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.024165533907264763,
|
| 62 |
+
"grad_norm": 1.7872492138745553,
|
| 63 |
+
"learning_rate": 9.515449934959716e-05,
|
| 64 |
+
"loss": 0.9537,
|
| 65 |
+
"step": 80
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.02718622564567286,
|
| 69 |
+
"grad_norm": 1.59848853171592,
|
| 70 |
+
"learning_rate": 9.771212547196623e-05,
|
| 71 |
+
"loss": 0.9299,
|
| 72 |
+
"step": 90
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.030206917384080954,
|
| 76 |
+
"grad_norm": 1.4459867462911404,
|
| 77 |
+
"learning_rate": 9.999999999999999e-05,
|
| 78 |
+
"loss": 0.8744,
|
| 79 |
+
"step": 100
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.03322760912248905,
|
| 83 |
+
"grad_norm": 1.2666954985573535,
|
| 84 |
+
"learning_rate": 9.97196261682243e-05,
|
| 85 |
+
"loss": 0.9279,
|
| 86 |
+
"step": 110
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.03624830086089714,
|
| 90 |
+
"grad_norm": 1.1375696912612132,
|
| 91 |
+
"learning_rate": 9.940809968847352e-05,
|
| 92 |
+
"loss": 0.943,
|
| 93 |
+
"step": 120
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.03926899259930524,
|
| 97 |
+
"grad_norm": 1.5887317329510864,
|
| 98 |
+
"learning_rate": 9.909657320872274e-05,
|
| 99 |
+
"loss": 0.8703,
|
| 100 |
+
"step": 130
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.04228968433771334,
|
| 104 |
+
"grad_norm": 1.5838324888073332,
|
| 105 |
+
"learning_rate": 9.878504672897197e-05,
|
| 106 |
+
"loss": 0.8943,
|
| 107 |
+
"step": 140
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.045310376076121435,
|
| 111 |
+
"grad_norm": 1.247226983069568,
|
| 112 |
+
"learning_rate": 9.847352024922119e-05,
|
| 113 |
+
"loss": 0.8818,
|
| 114 |
+
"step": 150
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.048331067814529526,
|
| 118 |
+
"grad_norm": 1.6083164873540423,
|
| 119 |
+
"learning_rate": 9.816199376947041e-05,
|
| 120 |
+
"loss": 0.8748,
|
| 121 |
+
"step": 160
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.05135175955293762,
|
| 125 |
+
"grad_norm": 1.0686893898752028,
|
| 126 |
+
"learning_rate": 9.785046728971962e-05,
|
| 127 |
+
"loss": 0.875,
|
| 128 |
+
"step": 170
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.05437245129134572,
|
| 132 |
+
"grad_norm": 1.3766933277373585,
|
| 133 |
+
"learning_rate": 9.753894080996884e-05,
|
| 134 |
+
"loss": 0.8324,
|
| 135 |
+
"step": 180
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.05739314302975381,
|
| 139 |
+
"grad_norm": 1.0332571782362285,
|
| 140 |
+
"learning_rate": 9.722741433021807e-05,
|
| 141 |
+
"loss": 0.8414,
|
| 142 |
+
"step": 190
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.06041383476816191,
|
| 146 |
+
"grad_norm": 1.2849086192566328,
|
| 147 |
+
"learning_rate": 9.691588785046729e-05,
|
| 148 |
+
"loss": 0.825,
|
| 149 |
+
"step": 200
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.06343452650657,
|
| 153 |
+
"grad_norm": 1.125015534848599,
|
| 154 |
+
"learning_rate": 9.660436137071651e-05,
|
| 155 |
+
"loss": 0.827,
|
| 156 |
+
"step": 210
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.0664552182449781,
|
| 160 |
+
"grad_norm": 1.0924381371211556,
|
| 161 |
+
"learning_rate": 9.629283489096573e-05,
|
| 162 |
+
"loss": 0.8381,
|
| 163 |
+
"step": 220
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.0694759099833862,
|
| 167 |
+
"grad_norm": 1.1382918007143823,
|
| 168 |
+
"learning_rate": 9.598130841121496e-05,
|
| 169 |
+
"loss": 0.8047,
|
| 170 |
+
"step": 230
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.07249660172179428,
|
| 174 |
+
"grad_norm": 1.263038309691145,
|
| 175 |
+
"learning_rate": 9.566978193146418e-05,
|
| 176 |
+
"loss": 0.7401,
|
| 177 |
+
"step": 240
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.07551729346020239,
|
| 181 |
+
"grad_norm": 1.223191453286042,
|
| 182 |
+
"learning_rate": 9.53582554517134e-05,
|
| 183 |
+
"loss": 0.7566,
|
| 184 |
+
"step": 250
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.07853798519861048,
|
| 188 |
+
"grad_norm": 1.4311949668852009,
|
| 189 |
+
"learning_rate": 9.504672897196261e-05,
|
| 190 |
+
"loss": 0.8324,
|
| 191 |
+
"step": 260
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.08155867693701857,
|
| 195 |
+
"grad_norm": 1.1199798019887752,
|
| 196 |
+
"learning_rate": 9.473520249221183e-05,
|
| 197 |
+
"loss": 0.7918,
|
| 198 |
+
"step": 270
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.08457936867542667,
|
| 202 |
+
"grad_norm": 1.1353541939369336,
|
| 203 |
+
"learning_rate": 9.442367601246106e-05,
|
| 204 |
+
"loss": 0.7982,
|
| 205 |
+
"step": 280
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.08760006041383477,
|
| 209 |
+
"grad_norm": 1.0377240839273387,
|
| 210 |
+
"learning_rate": 9.411214953271028e-05,
|
| 211 |
+
"loss": 0.759,
|
| 212 |
+
"step": 290
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.09062075215224287,
|
| 216 |
+
"grad_norm": 1.0430291081517695,
|
| 217 |
+
"learning_rate": 9.38006230529595e-05,
|
| 218 |
+
"loss": 0.7489,
|
| 219 |
+
"step": 300
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.09364144389065096,
|
| 223 |
+
"grad_norm": 1.0556490523453508,
|
| 224 |
+
"learning_rate": 9.348909657320873e-05,
|
| 225 |
+
"loss": 0.7972,
|
| 226 |
+
"step": 310
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.09666213562905905,
|
| 230 |
+
"grad_norm": 0.8377680272528175,
|
| 231 |
+
"learning_rate": 9.317757009345795e-05,
|
| 232 |
+
"loss": 0.7883,
|
| 233 |
+
"step": 320
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.09968282736746716,
|
| 237 |
+
"grad_norm": 1.133534627320159,
|
| 238 |
+
"learning_rate": 9.286604361370717e-05,
|
| 239 |
+
"loss": 0.7963,
|
| 240 |
+
"step": 330
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.10270351910587525,
|
| 244 |
+
"grad_norm": 1.115336306572395,
|
| 245 |
+
"learning_rate": 9.255451713395639e-05,
|
| 246 |
+
"loss": 0.7115,
|
| 247 |
+
"step": 340
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.10572421084428334,
|
| 251 |
+
"grad_norm": 1.0771112865985937,
|
| 252 |
+
"learning_rate": 9.22429906542056e-05,
|
| 253 |
+
"loss": 0.6852,
|
| 254 |
+
"step": 350
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.10874490258269144,
|
| 258 |
+
"grad_norm": 1.1279818268954833,
|
| 259 |
+
"learning_rate": 9.193146417445482e-05,
|
| 260 |
+
"loss": 0.752,
|
| 261 |
+
"step": 360
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.11176559432109953,
|
| 265 |
+
"grad_norm": 1.1447909234403442,
|
| 266 |
+
"learning_rate": 9.161993769470405e-05,
|
| 267 |
+
"loss": 0.7813,
|
| 268 |
+
"step": 370
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.11478628605950762,
|
| 272 |
+
"grad_norm": 0.9732050999087476,
|
| 273 |
+
"learning_rate": 9.130841121495327e-05,
|
| 274 |
+
"loss": 0.7529,
|
| 275 |
+
"step": 380
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.11780697779791573,
|
| 279 |
+
"grad_norm": 0.9183505631954165,
|
| 280 |
+
"learning_rate": 9.099688473520249e-05,
|
| 281 |
+
"loss": 0.8184,
|
| 282 |
+
"step": 390
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.12082766953632382,
|
| 286 |
+
"grad_norm": 1.3406866026704565,
|
| 287 |
+
"learning_rate": 9.068535825545172e-05,
|
| 288 |
+
"loss": 0.7546,
|
| 289 |
+
"step": 400
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.12384836127473191,
|
| 293 |
+
"grad_norm": 1.080230905978024,
|
| 294 |
+
"learning_rate": 9.037383177570094e-05,
|
| 295 |
+
"loss": 0.7943,
|
| 296 |
+
"step": 410
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.12686905301314,
|
| 300 |
+
"grad_norm": 0.981708271024814,
|
| 301 |
+
"learning_rate": 9.006230529595016e-05,
|
| 302 |
+
"loss": 0.7748,
|
| 303 |
+
"step": 420
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.1298897447515481,
|
| 307 |
+
"grad_norm": 1.032944901370365,
|
| 308 |
+
"learning_rate": 8.975077881619937e-05,
|
| 309 |
+
"loss": 0.7367,
|
| 310 |
+
"step": 430
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.1329104364899562,
|
| 314 |
+
"grad_norm": 0.9828039027398626,
|
| 315 |
+
"learning_rate": 8.943925233644859e-05,
|
| 316 |
+
"loss": 0.7562,
|
| 317 |
+
"step": 440
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.13593112822836428,
|
| 321 |
+
"grad_norm": 1.025926244346921,
|
| 322 |
+
"learning_rate": 8.912772585669782e-05,
|
| 323 |
+
"loss": 0.7799,
|
| 324 |
+
"step": 450
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.1389518199667724,
|
| 328 |
+
"grad_norm": 1.0500959648810009,
|
| 329 |
+
"learning_rate": 8.881619937694704e-05,
|
| 330 |
+
"loss": 0.7523,
|
| 331 |
+
"step": 460
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.1419725117051805,
|
| 335 |
+
"grad_norm": 1.1701611318445844,
|
| 336 |
+
"learning_rate": 8.850467289719626e-05,
|
| 337 |
+
"loss": 0.7686,
|
| 338 |
+
"step": 470
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.14499320344358857,
|
| 342 |
+
"grad_norm": 0.8990869023938298,
|
| 343 |
+
"learning_rate": 8.819314641744548e-05,
|
| 344 |
+
"loss": 0.7289,
|
| 345 |
+
"step": 480
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.14801389518199667,
|
| 349 |
+
"grad_norm": 1.1095162409175816,
|
| 350 |
+
"learning_rate": 8.78816199376947e-05,
|
| 351 |
+
"loss": 0.7332,
|
| 352 |
+
"step": 490
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.15103458692040478,
|
| 356 |
+
"grad_norm": 1.0108816660931794,
|
| 357 |
+
"learning_rate": 8.757009345794393e-05,
|
| 358 |
+
"loss": 0.7125,
|
| 359 |
+
"step": 500
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.15405527865881286,
|
| 363 |
+
"grad_norm": 1.3324995099592942,
|
| 364 |
+
"learning_rate": 8.725856697819315e-05,
|
| 365 |
+
"loss": 0.7254,
|
| 366 |
+
"step": 510
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.15707597039722096,
|
| 370 |
+
"grad_norm": 1.0845745603450858,
|
| 371 |
+
"learning_rate": 8.694704049844236e-05,
|
| 372 |
+
"loss": 0.7115,
|
| 373 |
+
"step": 520
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.16009666213562906,
|
| 377 |
+
"grad_norm": 0.7967865715388287,
|
| 378 |
+
"learning_rate": 8.663551401869158e-05,
|
| 379 |
+
"loss": 0.7334,
|
| 380 |
+
"step": 530
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.16311735387403714,
|
| 384 |
+
"grad_norm": 0.9490631327452955,
|
| 385 |
+
"learning_rate": 8.63239875389408e-05,
|
| 386 |
+
"loss": 0.7399,
|
| 387 |
+
"step": 540
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.16613804561244525,
|
| 391 |
+
"grad_norm": 0.8962644592007071,
|
| 392 |
+
"learning_rate": 8.601246105919003e-05,
|
| 393 |
+
"loss": 0.7498,
|
| 394 |
+
"step": 550
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.16915873735085335,
|
| 398 |
+
"grad_norm": 1.1678557286758378,
|
| 399 |
+
"learning_rate": 8.570093457943925e-05,
|
| 400 |
+
"loss": 0.7067,
|
| 401 |
+
"step": 560
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.17217942908926145,
|
| 405 |
+
"grad_norm": 1.088949753627781,
|
| 406 |
+
"learning_rate": 8.538940809968847e-05,
|
| 407 |
+
"loss": 0.748,
|
| 408 |
+
"step": 570
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.17520012082766953,
|
| 412 |
+
"grad_norm": 1.0893408136265033,
|
| 413 |
+
"learning_rate": 8.50778816199377e-05,
|
| 414 |
+
"loss": 0.7623,
|
| 415 |
+
"step": 580
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.17822081256607764,
|
| 419 |
+
"grad_norm": 0.9840533089260339,
|
| 420 |
+
"learning_rate": 8.476635514018692e-05,
|
| 421 |
+
"loss": 0.6797,
|
| 422 |
+
"step": 590
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.18124150430448574,
|
| 426 |
+
"grad_norm": 1.0812493862448191,
|
| 427 |
+
"learning_rate": 8.445482866043614e-05,
|
| 428 |
+
"loss": 0.7275,
|
| 429 |
+
"step": 600
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.18426219604289382,
|
| 433 |
+
"grad_norm": 1.2195461231662528,
|
| 434 |
+
"learning_rate": 8.414330218068535e-05,
|
| 435 |
+
"loss": 0.6863,
|
| 436 |
+
"step": 610
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.18728288778130192,
|
| 440 |
+
"grad_norm": 0.8428943981031285,
|
| 441 |
+
"learning_rate": 8.383177570093457e-05,
|
| 442 |
+
"loss": 0.7412,
|
| 443 |
+
"step": 620
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.19030357951971003,
|
| 447 |
+
"grad_norm": 0.9484156439369784,
|
| 448 |
+
"learning_rate": 8.35202492211838e-05,
|
| 449 |
+
"loss": 0.7479,
|
| 450 |
+
"step": 630
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.1933242712581181,
|
| 454 |
+
"grad_norm": 1.5268349287880287,
|
| 455 |
+
"learning_rate": 8.320872274143302e-05,
|
| 456 |
+
"loss": 0.7421,
|
| 457 |
+
"step": 640
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.1963449629965262,
|
| 461 |
+
"grad_norm": 1.0210188261100828,
|
| 462 |
+
"learning_rate": 8.289719626168224e-05,
|
| 463 |
+
"loss": 0.735,
|
| 464 |
+
"step": 650
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.1993656547349343,
|
| 468 |
+
"grad_norm": 1.0507811901351063,
|
| 469 |
+
"learning_rate": 8.258566978193147e-05,
|
| 470 |
+
"loss": 0.7174,
|
| 471 |
+
"step": 660
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.2023863464733424,
|
| 475 |
+
"grad_norm": 0.977107559554224,
|
| 476 |
+
"learning_rate": 8.227414330218069e-05,
|
| 477 |
+
"loss": 0.6539,
|
| 478 |
+
"step": 670
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.2054070382117505,
|
| 482 |
+
"grad_norm": 1.0747746808663372,
|
| 483 |
+
"learning_rate": 8.196261682242991e-05,
|
| 484 |
+
"loss": 0.6529,
|
| 485 |
+
"step": 680
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.2084277299501586,
|
| 489 |
+
"grad_norm": 0.9082287831059276,
|
| 490 |
+
"learning_rate": 8.165109034267912e-05,
|
| 491 |
+
"loss": 0.732,
|
| 492 |
+
"step": 690
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 0.21144842168856667,
|
| 496 |
+
"grad_norm": 0.9212172953657185,
|
| 497 |
+
"learning_rate": 8.133956386292834e-05,
|
| 498 |
+
"loss": 0.6734,
|
| 499 |
+
"step": 700
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.21446911342697478,
|
| 503 |
+
"grad_norm": 0.7665370190006591,
|
| 504 |
+
"learning_rate": 8.102803738317756e-05,
|
| 505 |
+
"loss": 0.7048,
|
| 506 |
+
"step": 710
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.21748980516538288,
|
| 510 |
+
"grad_norm": 0.8983656963975071,
|
| 511 |
+
"learning_rate": 8.071651090342679e-05,
|
| 512 |
+
"loss": 0.7166,
|
| 513 |
+
"step": 720
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.22051049690379096,
|
| 517 |
+
"grad_norm": 0.996733266552071,
|
| 518 |
+
"learning_rate": 8.040498442367601e-05,
|
| 519 |
+
"loss": 0.7354,
|
| 520 |
+
"step": 730
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.22353118864219906,
|
| 524 |
+
"grad_norm": 0.8782696913983529,
|
| 525 |
+
"learning_rate": 8.009345794392523e-05,
|
| 526 |
+
"loss": 0.6911,
|
| 527 |
+
"step": 740
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.22655188038060717,
|
| 531 |
+
"grad_norm": 1.0464611268652604,
|
| 532 |
+
"learning_rate": 7.978193146417447e-05,
|
| 533 |
+
"loss": 0.6791,
|
| 534 |
+
"step": 750
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.22957257211901524,
|
| 538 |
+
"grad_norm": 1.429521202085466,
|
| 539 |
+
"learning_rate": 7.947040498442369e-05,
|
| 540 |
+
"loss": 0.7254,
|
| 541 |
+
"step": 760
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.23259326385742335,
|
| 545 |
+
"grad_norm": 0.8265481561359109,
|
| 546 |
+
"learning_rate": 7.91588785046729e-05,
|
| 547 |
+
"loss": 0.6933,
|
| 548 |
+
"step": 770
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.23561395559583145,
|
| 552 |
+
"grad_norm": 1.1395351111445386,
|
| 553 |
+
"learning_rate": 7.884735202492212e-05,
|
| 554 |
+
"loss": 0.6927,
|
| 555 |
+
"step": 780
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.23863464733423953,
|
| 559 |
+
"grad_norm": 0.8683132914875289,
|
| 560 |
+
"learning_rate": 7.853582554517135e-05,
|
| 561 |
+
"loss": 0.7115,
|
| 562 |
+
"step": 790
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 0.24165533907264763,
|
| 566 |
+
"grad_norm": 1.0048357199050497,
|
| 567 |
+
"learning_rate": 7.822429906542057e-05,
|
| 568 |
+
"loss": 0.7258,
|
| 569 |
+
"step": 800
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.24467603081105574,
|
| 573 |
+
"grad_norm": 0.9186551341926896,
|
| 574 |
+
"learning_rate": 7.791277258566979e-05,
|
| 575 |
+
"loss": 0.6937,
|
| 576 |
+
"step": 810
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 0.24769672254946382,
|
| 580 |
+
"grad_norm": 0.90949443461049,
|
| 581 |
+
"learning_rate": 7.760124610591901e-05,
|
| 582 |
+
"loss": 0.717,
|
| 583 |
+
"step": 820
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 0.2507174142878719,
|
| 587 |
+
"grad_norm": 1.2709292068990643,
|
| 588 |
+
"learning_rate": 7.728971962616824e-05,
|
| 589 |
+
"loss": 0.7313,
|
| 590 |
+
"step": 830
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.25373810602628,
|
| 594 |
+
"grad_norm": 1.039112358866248,
|
| 595 |
+
"learning_rate": 7.697819314641746e-05,
|
| 596 |
+
"loss": 0.6961,
|
| 597 |
+
"step": 840
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 0.25675879776468813,
|
| 601 |
+
"grad_norm": 0.9738088719115688,
|
| 602 |
+
"learning_rate": 7.666666666666667e-05,
|
| 603 |
+
"loss": 0.7418,
|
| 604 |
+
"step": 850
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 0.2597794895030962,
|
| 608 |
+
"grad_norm": 1.0293122474854857,
|
| 609 |
+
"learning_rate": 7.635514018691589e-05,
|
| 610 |
+
"loss": 0.7208,
|
| 611 |
+
"step": 860
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 0.2628001812415043,
|
| 615 |
+
"grad_norm": 0.9219605479548113,
|
| 616 |
+
"learning_rate": 7.604361370716511e-05,
|
| 617 |
+
"loss": 0.6573,
|
| 618 |
+
"step": 870
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 0.2658208729799124,
|
| 622 |
+
"grad_norm": 0.986441148226433,
|
| 623 |
+
"learning_rate": 7.573208722741434e-05,
|
| 624 |
+
"loss": 0.6882,
|
| 625 |
+
"step": 880
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.2688415647183205,
|
| 629 |
+
"grad_norm": 0.8721644047390303,
|
| 630 |
+
"learning_rate": 7.542056074766356e-05,
|
| 631 |
+
"loss": 0.699,
|
| 632 |
+
"step": 890
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.27186225645672857,
|
| 636 |
+
"grad_norm": 0.8397538474783728,
|
| 637 |
+
"learning_rate": 7.510903426791278e-05,
|
| 638 |
+
"loss": 0.7473,
|
| 639 |
+
"step": 900
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.2748829481951367,
|
| 643 |
+
"grad_norm": 0.9204492440191102,
|
| 644 |
+
"learning_rate": 7.4797507788162e-05,
|
| 645 |
+
"loss": 0.682,
|
| 646 |
+
"step": 910
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 0.2779036399335448,
|
| 650 |
+
"grad_norm": 0.9916189555392334,
|
| 651 |
+
"learning_rate": 7.448598130841123e-05,
|
| 652 |
+
"loss": 0.6714,
|
| 653 |
+
"step": 920
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 0.28092433167195285,
|
| 657 |
+
"grad_norm": 0.9959802073019116,
|
| 658 |
+
"learning_rate": 7.417445482866045e-05,
|
| 659 |
+
"loss": 0.7068,
|
| 660 |
+
"step": 930
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 0.283945023410361,
|
| 664 |
+
"grad_norm": 0.822413918128088,
|
| 665 |
+
"learning_rate": 7.386292834890966e-05,
|
| 666 |
+
"loss": 0.6993,
|
| 667 |
+
"step": 940
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 0.28696571514876906,
|
| 671 |
+
"grad_norm": 0.9857850966520738,
|
| 672 |
+
"learning_rate": 7.355140186915888e-05,
|
| 673 |
+
"loss": 0.708,
|
| 674 |
+
"step": 950
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 0.28998640688717714,
|
| 678 |
+
"grad_norm": 0.7935815329490274,
|
| 679 |
+
"learning_rate": 7.32398753894081e-05,
|
| 680 |
+
"loss": 0.7262,
|
| 681 |
+
"step": 960
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 0.29300709862558527,
|
| 685 |
+
"grad_norm": 0.9762914722861878,
|
| 686 |
+
"learning_rate": 7.292834890965733e-05,
|
| 687 |
+
"loss": 0.6663,
|
| 688 |
+
"step": 970
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 0.29602779036399335,
|
| 692 |
+
"grad_norm": 0.9841123533490521,
|
| 693 |
+
"learning_rate": 7.261682242990655e-05,
|
| 694 |
+
"loss": 0.693,
|
| 695 |
+
"step": 980
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 0.2990484821024014,
|
| 699 |
+
"grad_norm": 0.9136079490982437,
|
| 700 |
+
"learning_rate": 7.230529595015577e-05,
|
| 701 |
+
"loss": 0.7049,
|
| 702 |
+
"step": 990
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 0.30206917384080956,
|
| 706 |
+
"grad_norm": 0.8902281683899582,
|
| 707 |
+
"learning_rate": 7.1993769470405e-05,
|
| 708 |
+
"loss": 0.6956,
|
| 709 |
+
"step": 1000
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 0.30508986557921763,
|
| 713 |
+
"grad_norm": 0.9555339802943792,
|
| 714 |
+
"learning_rate": 7.168224299065422e-05,
|
| 715 |
+
"loss": 0.6923,
|
| 716 |
+
"step": 1010
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 0.3081105573176257,
|
| 720 |
+
"grad_norm": 0.9226515554395557,
|
| 721 |
+
"learning_rate": 7.137071651090344e-05,
|
| 722 |
+
"loss": 0.7047,
|
| 723 |
+
"step": 1020
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 0.31113124905603384,
|
| 727 |
+
"grad_norm": 1.0265101302708708,
|
| 728 |
+
"learning_rate": 7.105919003115265e-05,
|
| 729 |
+
"loss": 0.6342,
|
| 730 |
+
"step": 1030
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 0.3141519407944419,
|
| 734 |
+
"grad_norm": 0.7354809813158237,
|
| 735 |
+
"learning_rate": 7.074766355140187e-05,
|
| 736 |
+
"loss": 0.7276,
|
| 737 |
+
"step": 1040
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 0.31717263253285,
|
| 741 |
+
"grad_norm": 0.753423387786277,
|
| 742 |
+
"learning_rate": 7.04361370716511e-05,
|
| 743 |
+
"loss": 0.6743,
|
| 744 |
+
"step": 1050
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 0.32019332427125813,
|
| 748 |
+
"grad_norm": 1.057091313600942,
|
| 749 |
+
"learning_rate": 7.012461059190032e-05,
|
| 750 |
+
"loss": 0.6715,
|
| 751 |
+
"step": 1060
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 0.3232140160096662,
|
| 755 |
+
"grad_norm": 0.7838790412397071,
|
| 756 |
+
"learning_rate": 6.981308411214954e-05,
|
| 757 |
+
"loss": 0.6671,
|
| 758 |
+
"step": 1070
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 0.3262347077480743,
|
| 762 |
+
"grad_norm": 0.8980932417690051,
|
| 763 |
+
"learning_rate": 6.950155763239876e-05,
|
| 764 |
+
"loss": 0.6774,
|
| 765 |
+
"step": 1080
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 0.3292553994864824,
|
| 769 |
+
"grad_norm": 0.9220267366468554,
|
| 770 |
+
"learning_rate": 6.919003115264799e-05,
|
| 771 |
+
"loss": 0.6428,
|
| 772 |
+
"step": 1090
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 0.3322760912248905,
|
| 776 |
+
"grad_norm": 0.9214463298045865,
|
| 777 |
+
"learning_rate": 6.887850467289721e-05,
|
| 778 |
+
"loss": 0.7271,
|
| 779 |
+
"step": 1100
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 0.3352967829632986,
|
| 783 |
+
"grad_norm": 0.9321131041370911,
|
| 784 |
+
"learning_rate": 6.856697819314643e-05,
|
| 785 |
+
"loss": 0.7258,
|
| 786 |
+
"step": 1110
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 0.3383174747017067,
|
| 790 |
+
"grad_norm": 1.279210720148468,
|
| 791 |
+
"learning_rate": 6.825545171339564e-05,
|
| 792 |
+
"loss": 0.6632,
|
| 793 |
+
"step": 1120
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 0.3413381664401148,
|
| 797 |
+
"grad_norm": 1.0059923489692595,
|
| 798 |
+
"learning_rate": 6.794392523364486e-05,
|
| 799 |
+
"loss": 0.6364,
|
| 800 |
+
"step": 1130
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 0.3443588581785229,
|
| 804 |
+
"grad_norm": 1.047129656921137,
|
| 805 |
+
"learning_rate": 6.763239875389409e-05,
|
| 806 |
+
"loss": 0.7291,
|
| 807 |
+
"step": 1140
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 0.347379549916931,
|
| 811 |
+
"grad_norm": 0.9272803839751749,
|
| 812 |
+
"learning_rate": 6.732087227414331e-05,
|
| 813 |
+
"loss": 0.6395,
|
| 814 |
+
"step": 1150
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 0.35040024165533906,
|
| 818 |
+
"grad_norm": 1.1151649879002739,
|
| 819 |
+
"learning_rate": 6.700934579439253e-05,
|
| 820 |
+
"loss": 0.6657,
|
| 821 |
+
"step": 1160
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 0.3534209333937472,
|
| 825 |
+
"grad_norm": 0.9241781872831288,
|
| 826 |
+
"learning_rate": 6.669781931464175e-05,
|
| 827 |
+
"loss": 0.6653,
|
| 828 |
+
"step": 1170
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 0.35644162513215527,
|
| 832 |
+
"grad_norm": 0.8358913806558893,
|
| 833 |
+
"learning_rate": 6.638629283489098e-05,
|
| 834 |
+
"loss": 0.7389,
|
| 835 |
+
"step": 1180
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 0.35946231687056335,
|
| 839 |
+
"grad_norm": 1.0319211344473511,
|
| 840 |
+
"learning_rate": 6.60747663551402e-05,
|
| 841 |
+
"loss": 0.6954,
|
| 842 |
+
"step": 1190
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 0.3624830086089715,
|
| 846 |
+
"grad_norm": 1.1505728351225282,
|
| 847 |
+
"learning_rate": 6.576323987538941e-05,
|
| 848 |
+
"loss": 0.7043,
|
| 849 |
+
"step": 1200
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 0.36550370034737956,
|
| 853 |
+
"grad_norm": 0.7959153182677297,
|
| 854 |
+
"learning_rate": 6.545171339563863e-05,
|
| 855 |
+
"loss": 0.6695,
|
| 856 |
+
"step": 1210
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 0.36852439208578763,
|
| 860 |
+
"grad_norm": 0.9337856655171398,
|
| 861 |
+
"learning_rate": 6.514018691588785e-05,
|
| 862 |
+
"loss": 0.7429,
|
| 863 |
+
"step": 1220
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 0.37154508382419577,
|
| 867 |
+
"grad_norm": 0.8063590050110404,
|
| 868 |
+
"learning_rate": 6.482866043613708e-05,
|
| 869 |
+
"loss": 0.6899,
|
| 870 |
+
"step": 1230
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 0.37456577556260384,
|
| 874 |
+
"grad_norm": 1.0599327700787213,
|
| 875 |
+
"learning_rate": 6.45171339563863e-05,
|
| 876 |
+
"loss": 0.6542,
|
| 877 |
+
"step": 1240
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 0.3775864673010119,
|
| 881 |
+
"grad_norm": 0.9744322038123949,
|
| 882 |
+
"learning_rate": 6.420560747663552e-05,
|
| 883 |
+
"loss": 0.706,
|
| 884 |
+
"step": 1250
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 0.38060715903942005,
|
| 888 |
+
"grad_norm": 0.8586307372164459,
|
| 889 |
+
"learning_rate": 6.389408099688475e-05,
|
| 890 |
+
"loss": 0.655,
|
| 891 |
+
"step": 1260
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 0.3836278507778281,
|
| 895 |
+
"grad_norm": 0.9830477484209696,
|
| 896 |
+
"learning_rate": 6.358255451713397e-05,
|
| 897 |
+
"loss": 0.6742,
|
| 898 |
+
"step": 1270
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 0.3866485425162362,
|
| 902 |
+
"grad_norm": 0.9902981913412452,
|
| 903 |
+
"learning_rate": 6.327102803738319e-05,
|
| 904 |
+
"loss": 0.6978,
|
| 905 |
+
"step": 1280
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 0.38966923425464434,
|
| 909 |
+
"grad_norm": 1.1185182362842843,
|
| 910 |
+
"learning_rate": 6.29595015576324e-05,
|
| 911 |
+
"loss": 0.73,
|
| 912 |
+
"step": 1290
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 0.3926899259930524,
|
| 916 |
+
"grad_norm": 0.877292130995857,
|
| 917 |
+
"learning_rate": 6.264797507788162e-05,
|
| 918 |
+
"loss": 0.7157,
|
| 919 |
+
"step": 1300
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 0.3957106177314605,
|
| 923 |
+
"grad_norm": 0.9682140121699045,
|
| 924 |
+
"learning_rate": 6.233644859813084e-05,
|
| 925 |
+
"loss": 0.678,
|
| 926 |
+
"step": 1310
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 0.3987313094698686,
|
| 930 |
+
"grad_norm": 1.0751962176923822,
|
| 931 |
+
"learning_rate": 6.202492211838007e-05,
|
| 932 |
+
"loss": 0.6393,
|
| 933 |
+
"step": 1320
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 0.4017520012082767,
|
| 937 |
+
"grad_norm": 0.8236677739270527,
|
| 938 |
+
"learning_rate": 6.171339563862929e-05,
|
| 939 |
+
"loss": 0.6709,
|
| 940 |
+
"step": 1330
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 0.4047726929466848,
|
| 944 |
+
"grad_norm": 0.9406647365344084,
|
| 945 |
+
"learning_rate": 6.140186915887851e-05,
|
| 946 |
+
"loss": 0.6588,
|
| 947 |
+
"step": 1340
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 0.4077933846850929,
|
| 951 |
+
"grad_norm": 1.2046005814359064,
|
| 952 |
+
"learning_rate": 6.109034267912774e-05,
|
| 953 |
+
"loss": 0.6807,
|
| 954 |
+
"step": 1350
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 0.410814076423501,
|
| 958 |
+
"grad_norm": 0.9066360882428902,
|
| 959 |
+
"learning_rate": 6.077881619937695e-05,
|
| 960 |
+
"loss": 0.7168,
|
| 961 |
+
"step": 1360
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 0.41383476816190906,
|
| 965 |
+
"grad_norm": 0.9527150302995969,
|
| 966 |
+
"learning_rate": 6.0467289719626174e-05,
|
| 967 |
+
"loss": 0.6984,
|
| 968 |
+
"step": 1370
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 0.4168554599003172,
|
| 972 |
+
"grad_norm": 0.8541783908332572,
|
| 973 |
+
"learning_rate": 6.01557632398754e-05,
|
| 974 |
+
"loss": 0.6596,
|
| 975 |
+
"step": 1380
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 0.41987615163872527,
|
| 979 |
+
"grad_norm": 0.9929215209215169,
|
| 980 |
+
"learning_rate": 5.984423676012462e-05,
|
| 981 |
+
"loss": 0.6496,
|
| 982 |
+
"step": 1390
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 0.42289684337713335,
|
| 986 |
+
"grad_norm": 0.8614209306432289,
|
| 987 |
+
"learning_rate": 5.9532710280373836e-05,
|
| 988 |
+
"loss": 0.6639,
|
| 989 |
+
"step": 1400
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 0.4259175351155415,
|
| 993 |
+
"grad_norm": 0.9182204803735963,
|
| 994 |
+
"learning_rate": 5.922118380062306e-05,
|
| 995 |
+
"loss": 0.6599,
|
| 996 |
+
"step": 1410
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 0.42893822685394956,
|
| 1000 |
+
"grad_norm": 0.991639681929367,
|
| 1001 |
+
"learning_rate": 5.890965732087228e-05,
|
| 1002 |
+
"loss": 0.6997,
|
| 1003 |
+
"step": 1420
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 0.43195891859235763,
|
| 1007 |
+
"grad_norm": 0.9192897320069349,
|
| 1008 |
+
"learning_rate": 5.8598130841121504e-05,
|
| 1009 |
+
"loss": 0.6911,
|
| 1010 |
+
"step": 1430
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 0.43497961033076576,
|
| 1014 |
+
"grad_norm": 0.8791677727387652,
|
| 1015 |
+
"learning_rate": 5.828660436137072e-05,
|
| 1016 |
+
"loss": 0.6133,
|
| 1017 |
+
"step": 1440
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 0.43800030206917384,
|
| 1021 |
+
"grad_norm": 0.9146654670599264,
|
| 1022 |
+
"learning_rate": 5.797507788161994e-05,
|
| 1023 |
+
"loss": 0.6671,
|
| 1024 |
+
"step": 1450
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 0.4410209938075819,
|
| 1028 |
+
"grad_norm": 0.7904811312174689,
|
| 1029 |
+
"learning_rate": 5.7663551401869165e-05,
|
| 1030 |
+
"loss": 0.6664,
|
| 1031 |
+
"step": 1460
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 0.44404168554599005,
|
| 1035 |
+
"grad_norm": 0.7986284407143894,
|
| 1036 |
+
"learning_rate": 5.735202492211839e-05,
|
| 1037 |
+
"loss": 0.7036,
|
| 1038 |
+
"step": 1470
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 0.4470623772843981,
|
| 1042 |
+
"grad_norm": 1.1832322364502883,
|
| 1043 |
+
"learning_rate": 5.704049844236761e-05,
|
| 1044 |
+
"loss": 0.6819,
|
| 1045 |
+
"step": 1480
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 0.4500830690228062,
|
| 1049 |
+
"grad_norm": 1.416546083504211,
|
| 1050 |
+
"learning_rate": 5.6728971962616826e-05,
|
| 1051 |
+
"loss": 0.6478,
|
| 1052 |
+
"step": 1490
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 0.45310376076121434,
|
| 1056 |
+
"grad_norm": 0.9415627484625246,
|
| 1057 |
+
"learning_rate": 5.641744548286605e-05,
|
| 1058 |
+
"loss": 0.6146,
|
| 1059 |
+
"step": 1500
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 0.4561244524996224,
|
| 1063 |
+
"grad_norm": 0.9223288773940925,
|
| 1064 |
+
"learning_rate": 5.610591900311527e-05,
|
| 1065 |
+
"loss": 0.7053,
|
| 1066 |
+
"step": 1510
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 0.4591451442380305,
|
| 1070 |
+
"grad_norm": 0.970256765114157,
|
| 1071 |
+
"learning_rate": 5.5794392523364494e-05,
|
| 1072 |
+
"loss": 0.6443,
|
| 1073 |
+
"step": 1520
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 0.4621658359764386,
|
| 1077 |
+
"grad_norm": 0.7403345850081653,
|
| 1078 |
+
"learning_rate": 5.548286604361371e-05,
|
| 1079 |
+
"loss": 0.6207,
|
| 1080 |
+
"step": 1530
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 0.4651865277148467,
|
| 1084 |
+
"grad_norm": 1.1294910445688315,
|
| 1085 |
+
"learning_rate": 5.517133956386293e-05,
|
| 1086 |
+
"loss": 0.6837,
|
| 1087 |
+
"step": 1540
|
| 1088 |
+
},
|
| 1089 |
+
{
|
| 1090 |
+
"epoch": 0.4682072194532548,
|
| 1091 |
+
"grad_norm": 1.084685705504392,
|
| 1092 |
+
"learning_rate": 5.4859813084112156e-05,
|
| 1093 |
+
"loss": 0.6619,
|
| 1094 |
+
"step": 1550
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 0.4712279111916629,
|
| 1098 |
+
"grad_norm": 0.9613159354858272,
|
| 1099 |
+
"learning_rate": 5.454828660436138e-05,
|
| 1100 |
+
"loss": 0.6515,
|
| 1101 |
+
"step": 1560
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 0.474248602930071,
|
| 1105 |
+
"grad_norm": 0.9837692555022853,
|
| 1106 |
+
"learning_rate": 5.4236760124610594e-05,
|
| 1107 |
+
"loss": 0.5924,
|
| 1108 |
+
"step": 1570
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"epoch": 0.47726929466847906,
|
| 1112 |
+
"grad_norm": 0.9979615856918954,
|
| 1113 |
+
"learning_rate": 5.392523364485982e-05,
|
| 1114 |
+
"loss": 0.7001,
|
| 1115 |
+
"step": 1580
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 0.4802899864068872,
|
| 1119 |
+
"grad_norm": 0.9728439737052923,
|
| 1120 |
+
"learning_rate": 5.361370716510904e-05,
|
| 1121 |
+
"loss": 0.6458,
|
| 1122 |
+
"step": 1590
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 0.48331067814529527,
|
| 1126 |
+
"grad_norm": 1.0126438240737292,
|
| 1127 |
+
"learning_rate": 5.330218068535826e-05,
|
| 1128 |
+
"loss": 0.7399,
|
| 1129 |
+
"step": 1600
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 0.48633136988370335,
|
| 1133 |
+
"grad_norm": 1.0282465120937962,
|
| 1134 |
+
"learning_rate": 5.2990654205607485e-05,
|
| 1135 |
+
"loss": 0.6591,
|
| 1136 |
+
"step": 1610
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 0.4893520616221115,
|
| 1140 |
+
"grad_norm": 1.0608069664171718,
|
| 1141 |
+
"learning_rate": 5.26791277258567e-05,
|
| 1142 |
+
"loss": 0.6852,
|
| 1143 |
+
"step": 1620
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 0.49237275336051955,
|
| 1147 |
+
"grad_norm": 1.06193177949756,
|
| 1148 |
+
"learning_rate": 5.2367601246105924e-05,
|
| 1149 |
+
"loss": 0.6256,
|
| 1150 |
+
"step": 1630
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 0.49539344509892763,
|
| 1154 |
+
"grad_norm": 0.8994514739072711,
|
| 1155 |
+
"learning_rate": 5.2056074766355146e-05,
|
| 1156 |
+
"loss": 0.6157,
|
| 1157 |
+
"step": 1640
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 0.49841413683733576,
|
| 1161 |
+
"grad_norm": 0.8831776925916133,
|
| 1162 |
+
"learning_rate": 5.174454828660437e-05,
|
| 1163 |
+
"loss": 0.6585,
|
| 1164 |
+
"step": 1650
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 0.5014348285757438,
|
| 1168 |
+
"grad_norm": 0.9057465973326372,
|
| 1169 |
+
"learning_rate": 5.1433021806853585e-05,
|
| 1170 |
+
"loss": 0.6548,
|
| 1171 |
+
"step": 1660
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 0.504455520314152,
|
| 1175 |
+
"grad_norm": 0.8470568048981005,
|
| 1176 |
+
"learning_rate": 5.112149532710281e-05,
|
| 1177 |
+
"loss": 0.5995,
|
| 1178 |
+
"step": 1670
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 0.50747621205256,
|
| 1182 |
+
"grad_norm": 0.9274346159474457,
|
| 1183 |
+
"learning_rate": 5.080996884735203e-05,
|
| 1184 |
+
"loss": 0.6695,
|
| 1185 |
+
"step": 1680
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 0.5104969037909681,
|
| 1189 |
+
"grad_norm": 0.8266785564731086,
|
| 1190 |
+
"learning_rate": 5.049844236760125e-05,
|
| 1191 |
+
"loss": 0.6749,
|
| 1192 |
+
"step": 1690
|
| 1193 |
+
},
|
| 1194 |
+
{
|
| 1195 |
+
"epoch": 0.5135175955293763,
|
| 1196 |
+
"grad_norm": 1.0808475616587816,
|
| 1197 |
+
"learning_rate": 5.018691588785047e-05,
|
| 1198 |
+
"loss": 0.6696,
|
| 1199 |
+
"step": 1700
|
| 1200 |
+
},
|
| 1201 |
+
{
|
| 1202 |
+
"epoch": 0.5165382872677843,
|
| 1203 |
+
"grad_norm": 0.938631750728695,
|
| 1204 |
+
"learning_rate": 4.987538940809969e-05,
|
| 1205 |
+
"loss": 0.6953,
|
| 1206 |
+
"step": 1710
|
| 1207 |
+
},
|
| 1208 |
+
{
|
| 1209 |
+
"epoch": 0.5195589790061924,
|
| 1210 |
+
"grad_norm": 1.0041642156156723,
|
| 1211 |
+
"learning_rate": 4.9563862928348914e-05,
|
| 1212 |
+
"loss": 0.6873,
|
| 1213 |
+
"step": 1720
|
| 1214 |
+
},
|
| 1215 |
+
{
|
| 1216 |
+
"epoch": 0.5225796707446005,
|
| 1217 |
+
"grad_norm": 0.8610395612409752,
|
| 1218 |
+
"learning_rate": 4.925233644859814e-05,
|
| 1219 |
+
"loss": 0.7225,
|
| 1220 |
+
"step": 1730
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 0.5256003624830086,
|
| 1224 |
+
"grad_norm": 0.9496184351237903,
|
| 1225 |
+
"learning_rate": 4.894080996884736e-05,
|
| 1226 |
+
"loss": 0.6382,
|
| 1227 |
+
"step": 1740
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"epoch": 0.5286210542214167,
|
| 1231 |
+
"grad_norm": 1.0830488045922504,
|
| 1232 |
+
"learning_rate": 4.8629283489096575e-05,
|
| 1233 |
+
"loss": 0.6314,
|
| 1234 |
+
"step": 1750
|
| 1235 |
+
},
|
| 1236 |
+
{
|
| 1237 |
+
"epoch": 0.5316417459598248,
|
| 1238 |
+
"grad_norm": 1.3308812167096526,
|
| 1239 |
+
"learning_rate": 4.83177570093458e-05,
|
| 1240 |
+
"loss": 0.6279,
|
| 1241 |
+
"step": 1760
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 0.5346624376982329,
|
| 1245 |
+
"grad_norm": 1.0384804569396457,
|
| 1246 |
+
"learning_rate": 4.800623052959502e-05,
|
| 1247 |
+
"loss": 0.6661,
|
| 1248 |
+
"step": 1770
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 0.537683129436641,
|
| 1252 |
+
"grad_norm": 0.9268941184363317,
|
| 1253 |
+
"learning_rate": 4.7694704049844244e-05,
|
| 1254 |
+
"loss": 0.5982,
|
| 1255 |
+
"step": 1780
|
| 1256 |
+
},
|
| 1257 |
+
{
|
| 1258 |
+
"epoch": 0.5407038211750491,
|
| 1259 |
+
"grad_norm": 1.0041389636701843,
|
| 1260 |
+
"learning_rate": 4.738317757009346e-05,
|
| 1261 |
+
"loss": 0.5852,
|
| 1262 |
+
"step": 1790
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 0.5437245129134571,
|
| 1266 |
+
"grad_norm": 1.0812438556748192,
|
| 1267 |
+
"learning_rate": 4.707165109034268e-05,
|
| 1268 |
+
"loss": 0.6445,
|
| 1269 |
+
"step": 1800
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 0.5467452046518653,
|
| 1273 |
+
"grad_norm": 0.7998207518212354,
|
| 1274 |
+
"learning_rate": 4.6760124610591905e-05,
|
| 1275 |
+
"loss": 0.6212,
|
| 1276 |
+
"step": 1810
|
| 1277 |
+
},
|
| 1278 |
+
{
|
| 1279 |
+
"epoch": 0.5497658963902734,
|
| 1280 |
+
"grad_norm": 0.9860219965019006,
|
| 1281 |
+
"learning_rate": 4.644859813084113e-05,
|
| 1282 |
+
"loss": 0.6453,
|
| 1283 |
+
"step": 1820
|
| 1284 |
+
},
|
| 1285 |
+
{
|
| 1286 |
+
"epoch": 0.5527865881286814,
|
| 1287 |
+
"grad_norm": 1.0673930524321107,
|
| 1288 |
+
"learning_rate": 4.6137071651090343e-05,
|
| 1289 |
+
"loss": 0.5837,
|
| 1290 |
+
"step": 1830
|
| 1291 |
+
},
|
| 1292 |
+
{
|
| 1293 |
+
"epoch": 0.5558072798670896,
|
| 1294 |
+
"grad_norm": 1.143808636631285,
|
| 1295 |
+
"learning_rate": 4.5825545171339566e-05,
|
| 1296 |
+
"loss": 0.6996,
|
| 1297 |
+
"step": 1840
|
| 1298 |
+
},
|
| 1299 |
+
{
|
| 1300 |
+
"epoch": 0.5588279716054977,
|
| 1301 |
+
"grad_norm": 0.9805310678029436,
|
| 1302 |
+
"learning_rate": 4.551401869158879e-05,
|
| 1303 |
+
"loss": 0.6062,
|
| 1304 |
+
"step": 1850
|
| 1305 |
+
},
|
| 1306 |
+
{
|
| 1307 |
+
"epoch": 0.5618486633439057,
|
| 1308 |
+
"grad_norm": 1.0396445753595607,
|
| 1309 |
+
"learning_rate": 4.520249221183801e-05,
|
| 1310 |
+
"loss": 0.6249,
|
| 1311 |
+
"step": 1860
|
| 1312 |
+
},
|
| 1313 |
+
{
|
| 1314 |
+
"epoch": 0.5648693550823138,
|
| 1315 |
+
"grad_norm": 0.9626964932261893,
|
| 1316 |
+
"learning_rate": 4.4890965732087234e-05,
|
| 1317 |
+
"loss": 0.6571,
|
| 1318 |
+
"step": 1870
|
| 1319 |
+
},
|
| 1320 |
+
{
|
| 1321 |
+
"epoch": 0.567890046820722,
|
| 1322 |
+
"grad_norm": 0.9540429394126168,
|
| 1323 |
+
"learning_rate": 4.457943925233645e-05,
|
| 1324 |
+
"loss": 0.6324,
|
| 1325 |
+
"step": 1880
|
| 1326 |
+
},
|
| 1327 |
+
{
|
| 1328 |
+
"epoch": 0.57091073855913,
|
| 1329 |
+
"grad_norm": 0.9015174511995171,
|
| 1330 |
+
"learning_rate": 4.426791277258567e-05,
|
| 1331 |
+
"loss": 0.6863,
|
| 1332 |
+
"step": 1890
|
| 1333 |
+
},
|
| 1334 |
+
{
|
| 1335 |
+
"epoch": 0.5739314302975381,
|
| 1336 |
+
"grad_norm": 0.7989728677956948,
|
| 1337 |
+
"learning_rate": 4.3956386292834895e-05,
|
| 1338 |
+
"loss": 0.6107,
|
| 1339 |
+
"step": 1900
|
| 1340 |
+
},
|
| 1341 |
+
{
|
| 1342 |
+
"epoch": 0.5769521220359463,
|
| 1343 |
+
"grad_norm": 1.0283655646158905,
|
| 1344 |
+
"learning_rate": 4.364485981308412e-05,
|
| 1345 |
+
"loss": 0.5799,
|
| 1346 |
+
"step": 1910
|
| 1347 |
+
},
|
| 1348 |
+
{
|
| 1349 |
+
"epoch": 0.5799728137743543,
|
| 1350 |
+
"grad_norm": 0.8496110087014387,
|
| 1351 |
+
"learning_rate": 4.3333333333333334e-05,
|
| 1352 |
+
"loss": 0.6114,
|
| 1353 |
+
"step": 1920
|
| 1354 |
+
},
|
| 1355 |
+
{
|
| 1356 |
+
"epoch": 0.5829935055127624,
|
| 1357 |
+
"grad_norm": 0.9378297749359396,
|
| 1358 |
+
"learning_rate": 4.302180685358256e-05,
|
| 1359 |
+
"loss": 0.6238,
|
| 1360 |
+
"step": 1930
|
| 1361 |
+
},
|
| 1362 |
+
{
|
| 1363 |
+
"epoch": 0.5860141972511705,
|
| 1364 |
+
"grad_norm": 0.7724041203791813,
|
| 1365 |
+
"learning_rate": 4.271028037383178e-05,
|
| 1366 |
+
"loss": 0.6684,
|
| 1367 |
+
"step": 1940
|
| 1368 |
+
},
|
| 1369 |
+
{
|
| 1370 |
+
"epoch": 0.5890348889895786,
|
| 1371 |
+
"grad_norm": 0.919952085375962,
|
| 1372 |
+
"learning_rate": 4.2398753894081e-05,
|
| 1373 |
+
"loss": 0.658,
|
| 1374 |
+
"step": 1950
|
| 1375 |
+
},
|
| 1376 |
+
{
|
| 1377 |
+
"epoch": 0.5920555807279867,
|
| 1378 |
+
"grad_norm": 1.0984535902148247,
|
| 1379 |
+
"learning_rate": 4.2087227414330225e-05,
|
| 1380 |
+
"loss": 0.6271,
|
| 1381 |
+
"step": 1960
|
| 1382 |
+
},
|
| 1383 |
+
{
|
| 1384 |
+
"epoch": 0.5950762724663948,
|
| 1385 |
+
"grad_norm": 1.089793595134447,
|
| 1386 |
+
"learning_rate": 4.177570093457944e-05,
|
| 1387 |
+
"loss": 0.6401,
|
| 1388 |
+
"step": 1970
|
| 1389 |
+
},
|
| 1390 |
+
{
|
| 1391 |
+
"epoch": 0.5980969642048029,
|
| 1392 |
+
"grad_norm": 0.9470634176090514,
|
| 1393 |
+
"learning_rate": 4.1464174454828663e-05,
|
| 1394 |
+
"loss": 0.6269,
|
| 1395 |
+
"step": 1980
|
| 1396 |
+
},
|
| 1397 |
+
{
|
| 1398 |
+
"epoch": 0.601117655943211,
|
| 1399 |
+
"grad_norm": 0.9586690317848461,
|
| 1400 |
+
"learning_rate": 4.1152647975077886e-05,
|
| 1401 |
+
"loss": 0.6455,
|
| 1402 |
+
"step": 1990
|
| 1403 |
+
},
|
| 1404 |
+
{
|
| 1405 |
+
"epoch": 0.6041383476816191,
|
| 1406 |
+
"grad_norm": 0.9355637416127348,
|
| 1407 |
+
"learning_rate": 4.084112149532711e-05,
|
| 1408 |
+
"loss": 0.6505,
|
| 1409 |
+
"step": 2000
|
| 1410 |
+
},
|
| 1411 |
+
{
|
| 1412 |
+
"epoch": 0.6071590394200271,
|
| 1413 |
+
"grad_norm": 0.9422028995953846,
|
| 1414 |
+
"learning_rate": 4.0529595015576325e-05,
|
| 1415 |
+
"loss": 0.5881,
|
| 1416 |
+
"step": 2010
|
| 1417 |
+
},
|
| 1418 |
+
{
|
| 1419 |
+
"epoch": 0.6101797311584353,
|
| 1420 |
+
"grad_norm": 1.0662379927757937,
|
| 1421 |
+
"learning_rate": 4.021806853582555e-05,
|
| 1422 |
+
"loss": 0.6965,
|
| 1423 |
+
"step": 2020
|
| 1424 |
+
},
|
| 1425 |
+
{
|
| 1426 |
+
"epoch": 0.6132004228968434,
|
| 1427 |
+
"grad_norm": 1.112340047069814,
|
| 1428 |
+
"learning_rate": 3.990654205607477e-05,
|
| 1429 |
+
"loss": 0.6696,
|
| 1430 |
+
"step": 2030
|
| 1431 |
+
},
|
| 1432 |
+
{
|
| 1433 |
+
"epoch": 0.6162211146352514,
|
| 1434 |
+
"grad_norm": 0.9811828345166662,
|
| 1435 |
+
"learning_rate": 3.959501557632399e-05,
|
| 1436 |
+
"loss": 0.6774,
|
| 1437 |
+
"step": 2040
|
| 1438 |
+
},
|
| 1439 |
+
{
|
| 1440 |
+
"epoch": 0.6192418063736596,
|
| 1441 |
+
"grad_norm": 0.8797632809116769,
|
| 1442 |
+
"learning_rate": 3.928348909657321e-05,
|
| 1443 |
+
"loss": 0.6814,
|
| 1444 |
+
"step": 2050
|
| 1445 |
+
},
|
| 1446 |
+
{
|
| 1447 |
+
"epoch": 0.6222624981120677,
|
| 1448 |
+
"grad_norm": 1.0012864041487008,
|
| 1449 |
+
"learning_rate": 3.897196261682243e-05,
|
| 1450 |
+
"loss": 0.6789,
|
| 1451 |
+
"step": 2060
|
| 1452 |
+
},
|
| 1453 |
+
{
|
| 1454 |
+
"epoch": 0.6252831898504757,
|
| 1455 |
+
"grad_norm": 0.9363857271307736,
|
| 1456 |
+
"learning_rate": 3.8660436137071654e-05,
|
| 1457 |
+
"loss": 0.6286,
|
| 1458 |
+
"step": 2070
|
| 1459 |
+
},
|
| 1460 |
+
{
|
| 1461 |
+
"epoch": 0.6283038815888838,
|
| 1462 |
+
"grad_norm": 0.8994398345785907,
|
| 1463 |
+
"learning_rate": 3.834890965732088e-05,
|
| 1464 |
+
"loss": 0.5957,
|
| 1465 |
+
"step": 2080
|
| 1466 |
+
},
|
| 1467 |
+
{
|
| 1468 |
+
"epoch": 0.631324573327292,
|
| 1469 |
+
"grad_norm": 0.9307639618859869,
|
| 1470 |
+
"learning_rate": 3.80373831775701e-05,
|
| 1471 |
+
"loss": 0.6365,
|
| 1472 |
+
"step": 2090
|
| 1473 |
+
},
|
| 1474 |
+
{
|
| 1475 |
+
"epoch": 0.6343452650657,
|
| 1476 |
+
"grad_norm": 0.8458149477874153,
|
| 1477 |
+
"learning_rate": 3.7725856697819315e-05,
|
| 1478 |
+
"loss": 0.682,
|
| 1479 |
+
"step": 2100
|
| 1480 |
+
},
|
| 1481 |
+
{
|
| 1482 |
+
"epoch": 0.6373659568041081,
|
| 1483 |
+
"grad_norm": 0.782298940529486,
|
| 1484 |
+
"learning_rate": 3.741433021806854e-05,
|
| 1485 |
+
"loss": 0.68,
|
| 1486 |
+
"step": 2110
|
| 1487 |
+
},
|
| 1488 |
+
{
|
| 1489 |
+
"epoch": 0.6403866485425163,
|
| 1490 |
+
"grad_norm": 0.9771834449277986,
|
| 1491 |
+
"learning_rate": 3.710280373831776e-05,
|
| 1492 |
+
"loss": 0.6732,
|
| 1493 |
+
"step": 2120
|
| 1494 |
+
},
|
| 1495 |
+
{
|
| 1496 |
+
"epoch": 0.6434073402809243,
|
| 1497 |
+
"grad_norm": 0.9295914583308844,
|
| 1498 |
+
"learning_rate": 3.6791277258566983e-05,
|
| 1499 |
+
"loss": 0.6173,
|
| 1500 |
+
"step": 2130
|
| 1501 |
+
},
|
| 1502 |
+
{
|
| 1503 |
+
"epoch": 0.6464280320193324,
|
| 1504 |
+
"grad_norm": 1.179909806414257,
|
| 1505 |
+
"learning_rate": 3.64797507788162e-05,
|
| 1506 |
+
"loss": 0.6193,
|
| 1507 |
+
"step": 2140
|
| 1508 |
+
},
|
| 1509 |
+
{
|
| 1510 |
+
"epoch": 0.6494487237577405,
|
| 1511 |
+
"grad_norm": 0.99878344996002,
|
| 1512 |
+
"learning_rate": 3.616822429906542e-05,
|
| 1513 |
+
"loss": 0.6116,
|
| 1514 |
+
"step": 2150
|
| 1515 |
+
},
|
| 1516 |
+
{
|
| 1517 |
+
"epoch": 0.6524694154961486,
|
| 1518 |
+
"grad_norm": 1.1005346639943379,
|
| 1519 |
+
"learning_rate": 3.5856697819314645e-05,
|
| 1520 |
+
"loss": 0.6794,
|
| 1521 |
+
"step": 2160
|
| 1522 |
+
},
|
| 1523 |
+
{
|
| 1524 |
+
"epoch": 0.6554901072345567,
|
| 1525 |
+
"grad_norm": 0.8754349383421103,
|
| 1526 |
+
"learning_rate": 3.554517133956387e-05,
|
| 1527 |
+
"loss": 0.627,
|
| 1528 |
+
"step": 2170
|
| 1529 |
+
},
|
| 1530 |
+
{
|
| 1531 |
+
"epoch": 0.6585107989729648,
|
| 1532 |
+
"grad_norm": 0.8266136534894406,
|
| 1533 |
+
"learning_rate": 3.523364485981308e-05,
|
| 1534 |
+
"loss": 0.5961,
|
| 1535 |
+
"step": 2180
|
| 1536 |
+
},
|
| 1537 |
+
{
|
| 1538 |
+
"epoch": 0.6615314907113728,
|
| 1539 |
+
"grad_norm": 1.0546584904026568,
|
| 1540 |
+
"learning_rate": 3.4922118380062306e-05,
|
| 1541 |
+
"loss": 0.6374,
|
| 1542 |
+
"step": 2190
|
| 1543 |
+
},
|
| 1544 |
+
{
|
| 1545 |
+
"epoch": 0.664552182449781,
|
| 1546 |
+
"grad_norm": 0.9197595680333482,
|
| 1547 |
+
"learning_rate": 3.461059190031153e-05,
|
| 1548 |
+
"loss": 0.6493,
|
| 1549 |
+
"step": 2200
|
| 1550 |
+
},
|
| 1551 |
+
{
|
| 1552 |
+
"epoch": 0.6675728741881891,
|
| 1553 |
+
"grad_norm": 0.8832736598142242,
|
| 1554 |
+
"learning_rate": 3.429906542056075e-05,
|
| 1555 |
+
"loss": 0.6243,
|
| 1556 |
+
"step": 2210
|
| 1557 |
+
},
|
| 1558 |
+
{
|
| 1559 |
+
"epoch": 0.6705935659265972,
|
| 1560 |
+
"grad_norm": 0.7953765164555499,
|
| 1561 |
+
"learning_rate": 3.3987538940809974e-05,
|
| 1562 |
+
"loss": 0.6413,
|
| 1563 |
+
"step": 2220
|
| 1564 |
+
},
|
| 1565 |
+
{
|
| 1566 |
+
"epoch": 0.6736142576650053,
|
| 1567 |
+
"grad_norm": 0.8616203793917606,
|
| 1568 |
+
"learning_rate": 3.367601246105919e-05,
|
| 1569 |
+
"loss": 0.6466,
|
| 1570 |
+
"step": 2230
|
| 1571 |
+
},
|
| 1572 |
+
{
|
| 1573 |
+
"epoch": 0.6766349494034134,
|
| 1574 |
+
"grad_norm": 0.9546691619897619,
|
| 1575 |
+
"learning_rate": 3.336448598130841e-05,
|
| 1576 |
+
"loss": 0.61,
|
| 1577 |
+
"step": 2240
|
| 1578 |
+
},
|
| 1579 |
+
{
|
| 1580 |
+
"epoch": 0.6796556411418215,
|
| 1581 |
+
"grad_norm": 0.9931567171367084,
|
| 1582 |
+
"learning_rate": 3.3052959501557635e-05,
|
| 1583 |
+
"loss": 0.6541,
|
| 1584 |
+
"step": 2250
|
| 1585 |
+
},
|
| 1586 |
+
{
|
| 1587 |
+
"epoch": 0.6826763328802296,
|
| 1588 |
+
"grad_norm": 0.910445823032414,
|
| 1589 |
+
"learning_rate": 3.274143302180686e-05,
|
| 1590 |
+
"loss": 0.6512,
|
| 1591 |
+
"step": 2260
|
| 1592 |
+
},
|
| 1593 |
+
{
|
| 1594 |
+
"epoch": 0.6856970246186377,
|
| 1595 |
+
"grad_norm": 0.9925620053257106,
|
| 1596 |
+
"learning_rate": 3.2429906542056074e-05,
|
| 1597 |
+
"loss": 0.6731,
|
| 1598 |
+
"step": 2270
|
| 1599 |
+
},
|
| 1600 |
+
{
|
| 1601 |
+
"epoch": 0.6887177163570458,
|
| 1602 |
+
"grad_norm": 0.9314964027183025,
|
| 1603 |
+
"learning_rate": 3.21183800623053e-05,
|
| 1604 |
+
"loss": 0.6349,
|
| 1605 |
+
"step": 2280
|
| 1606 |
+
},
|
| 1607 |
+
{
|
| 1608 |
+
"epoch": 0.6917384080954538,
|
| 1609 |
+
"grad_norm": 0.9924403199233388,
|
| 1610 |
+
"learning_rate": 3.180685358255452e-05,
|
| 1611 |
+
"loss": 0.6119,
|
| 1612 |
+
"step": 2290
|
| 1613 |
+
},
|
| 1614 |
+
{
|
| 1615 |
+
"epoch": 0.694759099833862,
|
| 1616 |
+
"grad_norm": 1.016617404918067,
|
| 1617 |
+
"learning_rate": 3.149532710280374e-05,
|
| 1618 |
+
"loss": 0.6364,
|
| 1619 |
+
"step": 2300
|
| 1620 |
+
},
|
| 1621 |
+
{
|
| 1622 |
+
"epoch": 0.6977797915722701,
|
| 1623 |
+
"grad_norm": 1.2100326550184481,
|
| 1624 |
+
"learning_rate": 3.118380062305296e-05,
|
| 1625 |
+
"loss": 0.6502,
|
| 1626 |
+
"step": 2310
|
| 1627 |
+
},
|
| 1628 |
+
{
|
| 1629 |
+
"epoch": 0.7008004833106781,
|
| 1630 |
+
"grad_norm": 0.8640417568500796,
|
| 1631 |
+
"learning_rate": 3.087227414330218e-05,
|
| 1632 |
+
"loss": 0.6129,
|
| 1633 |
+
"step": 2320
|
| 1634 |
+
},
|
| 1635 |
+
{
|
| 1636 |
+
"epoch": 0.7038211750490863,
|
| 1637 |
+
"grad_norm": 1.0875619618857508,
|
| 1638 |
+
"learning_rate": 3.05607476635514e-05,
|
| 1639 |
+
"loss": 0.6615,
|
| 1640 |
+
"step": 2330
|
| 1641 |
+
},
|
| 1642 |
+
{
|
| 1643 |
+
"epoch": 0.7068418667874944,
|
| 1644 |
+
"grad_norm": 0.8939128045261261,
|
| 1645 |
+
"learning_rate": 3.0249221183800626e-05,
|
| 1646 |
+
"loss": 0.6081,
|
| 1647 |
+
"step": 2340
|
| 1648 |
+
},
|
| 1649 |
+
{
|
| 1650 |
+
"epoch": 0.7098625585259024,
|
| 1651 |
+
"grad_norm": 0.8987589575526329,
|
| 1652 |
+
"learning_rate": 2.9937694704049845e-05,
|
| 1653 |
+
"loss": 0.6811,
|
| 1654 |
+
"step": 2350
|
| 1655 |
+
},
|
| 1656 |
+
{
|
| 1657 |
+
"epoch": 0.7128832502643105,
|
| 1658 |
+
"grad_norm": 0.9024007443688661,
|
| 1659 |
+
"learning_rate": 2.9626168224299068e-05,
|
| 1660 |
+
"loss": 0.6009,
|
| 1661 |
+
"step": 2360
|
| 1662 |
+
},
|
| 1663 |
+
{
|
| 1664 |
+
"epoch": 0.7159039420027187,
|
| 1665 |
+
"grad_norm": 0.963849254096249,
|
| 1666 |
+
"learning_rate": 2.9314641744548287e-05,
|
| 1667 |
+
"loss": 0.6509,
|
| 1668 |
+
"step": 2370
|
| 1669 |
+
},
|
| 1670 |
+
{
|
| 1671 |
+
"epoch": 0.7189246337411267,
|
| 1672 |
+
"grad_norm": 1.0800126230646785,
|
| 1673 |
+
"learning_rate": 2.900311526479751e-05,
|
| 1674 |
+
"loss": 0.6791,
|
| 1675 |
+
"step": 2380
|
| 1676 |
+
},
|
| 1677 |
+
{
|
| 1678 |
+
"epoch": 0.7219453254795348,
|
| 1679 |
+
"grad_norm": 0.8294159891121566,
|
| 1680 |
+
"learning_rate": 2.869158878504673e-05,
|
| 1681 |
+
"loss": 0.6334,
|
| 1682 |
+
"step": 2390
|
| 1683 |
+
},
|
| 1684 |
+
{
|
| 1685 |
+
"epoch": 0.724966017217943,
|
| 1686 |
+
"grad_norm": 1.0308203810536511,
|
| 1687 |
+
"learning_rate": 2.8380062305295952e-05,
|
| 1688 |
+
"loss": 0.627,
|
| 1689 |
+
"step": 2400
|
| 1690 |
+
},
|
| 1691 |
+
{
|
| 1692 |
+
"epoch": 0.727986708956351,
|
| 1693 |
+
"grad_norm": 0.9512924723408093,
|
| 1694 |
+
"learning_rate": 2.806853582554517e-05,
|
| 1695 |
+
"loss": 0.5754,
|
| 1696 |
+
"step": 2410
|
| 1697 |
+
},
|
| 1698 |
+
{
|
| 1699 |
+
"epoch": 0.7310074006947591,
|
| 1700 |
+
"grad_norm": 1.010781394411619,
|
| 1701 |
+
"learning_rate": 2.7757009345794394e-05,
|
| 1702 |
+
"loss": 0.6416,
|
| 1703 |
+
"step": 2420
|
| 1704 |
+
},
|
| 1705 |
+
{
|
| 1706 |
+
"epoch": 0.7340280924331672,
|
| 1707 |
+
"grad_norm": 1.0951399649684406,
|
| 1708 |
+
"learning_rate": 2.7445482866043613e-05,
|
| 1709 |
+
"loss": 0.6088,
|
| 1710 |
+
"step": 2430
|
| 1711 |
+
},
|
| 1712 |
+
{
|
| 1713 |
+
"epoch": 0.7370487841715753,
|
| 1714 |
+
"grad_norm": 1.0040444715395704,
|
| 1715 |
+
"learning_rate": 2.7133956386292836e-05,
|
| 1716 |
+
"loss": 0.6473,
|
| 1717 |
+
"step": 2440
|
| 1718 |
+
},
|
| 1719 |
+
{
|
| 1720 |
+
"epoch": 0.7400694759099834,
|
| 1721 |
+
"grad_norm": 0.9086640535011864,
|
| 1722 |
+
"learning_rate": 2.682242990654206e-05,
|
| 1723 |
+
"loss": 0.5991,
|
| 1724 |
+
"step": 2450
|
| 1725 |
+
},
|
| 1726 |
+
{
|
| 1727 |
+
"epoch": 0.7430901676483915,
|
| 1728 |
+
"grad_norm": 0.9979590888744413,
|
| 1729 |
+
"learning_rate": 2.6510903426791278e-05,
|
| 1730 |
+
"loss": 0.6554,
|
| 1731 |
+
"step": 2460
|
| 1732 |
+
},
|
| 1733 |
+
{
|
| 1734 |
+
"epoch": 0.7461108593867996,
|
| 1735 |
+
"grad_norm": 0.996844139395838,
|
| 1736 |
+
"learning_rate": 2.61993769470405e-05,
|
| 1737 |
+
"loss": 0.611,
|
| 1738 |
+
"step": 2470
|
| 1739 |
+
},
|
| 1740 |
+
{
|
| 1741 |
+
"epoch": 0.7491315511252077,
|
| 1742 |
+
"grad_norm": 1.0523632147482274,
|
| 1743 |
+
"learning_rate": 2.588785046728972e-05,
|
| 1744 |
+
"loss": 0.6327,
|
| 1745 |
+
"step": 2480
|
| 1746 |
+
},
|
| 1747 |
+
{
|
| 1748 |
+
"epoch": 0.7521522428636158,
|
| 1749 |
+
"grad_norm": 1.1077614436376035,
|
| 1750 |
+
"learning_rate": 2.5576323987538943e-05,
|
| 1751 |
+
"loss": 0.6846,
|
| 1752 |
+
"step": 2490
|
| 1753 |
+
},
|
| 1754 |
+
{
|
| 1755 |
+
"epoch": 0.7551729346020238,
|
| 1756 |
+
"grad_norm": 1.0062846906925333,
|
| 1757 |
+
"learning_rate": 2.5264797507788162e-05,
|
| 1758 |
+
"loss": 0.632,
|
| 1759 |
+
"step": 2500
|
| 1760 |
+
},
|
| 1761 |
+
{
|
| 1762 |
+
"epoch": 0.758193626340432,
|
| 1763 |
+
"grad_norm": 0.8044057354860985,
|
| 1764 |
+
"learning_rate": 2.4953271028037385e-05,
|
| 1765 |
+
"loss": 0.6423,
|
| 1766 |
+
"step": 2510
|
| 1767 |
+
},
|
| 1768 |
+
{
|
| 1769 |
+
"epoch": 0.7612143180788401,
|
| 1770 |
+
"grad_norm": 1.0776815216184104,
|
| 1771 |
+
"learning_rate": 2.4641744548286604e-05,
|
| 1772 |
+
"loss": 0.5959,
|
| 1773 |
+
"step": 2520
|
| 1774 |
+
},
|
| 1775 |
+
{
|
| 1776 |
+
"epoch": 0.7642350098172481,
|
| 1777 |
+
"grad_norm": 1.3229091185592443,
|
| 1778 |
+
"learning_rate": 2.4330218068535827e-05,
|
| 1779 |
+
"loss": 0.6308,
|
| 1780 |
+
"step": 2530
|
| 1781 |
+
},
|
| 1782 |
+
{
|
| 1783 |
+
"epoch": 0.7672557015556563,
|
| 1784 |
+
"grad_norm": 1.0134005011914144,
|
| 1785 |
+
"learning_rate": 2.4018691588785046e-05,
|
| 1786 |
+
"loss": 0.6082,
|
| 1787 |
+
"step": 2540
|
| 1788 |
+
},
|
| 1789 |
+
{
|
| 1790 |
+
"epoch": 0.7702763932940644,
|
| 1791 |
+
"grad_norm": 1.0710215536966488,
|
| 1792 |
+
"learning_rate": 2.370716510903427e-05,
|
| 1793 |
+
"loss": 0.5964,
|
| 1794 |
+
"step": 2550
|
| 1795 |
+
},
|
| 1796 |
+
{
|
| 1797 |
+
"epoch": 0.7732970850324724,
|
| 1798 |
+
"grad_norm": 1.0857761269794204,
|
| 1799 |
+
"learning_rate": 2.3395638629283488e-05,
|
| 1800 |
+
"loss": 0.5891,
|
| 1801 |
+
"step": 2560
|
| 1802 |
+
},
|
| 1803 |
+
{
|
| 1804 |
+
"epoch": 0.7763177767708805,
|
| 1805 |
+
"grad_norm": 0.9369701363776228,
|
| 1806 |
+
"learning_rate": 2.308411214953271e-05,
|
| 1807 |
+
"loss": 0.6104,
|
| 1808 |
+
"step": 2570
|
| 1809 |
+
},
|
| 1810 |
+
{
|
| 1811 |
+
"epoch": 0.7793384685092887,
|
| 1812 |
+
"grad_norm": 0.8529120199979476,
|
| 1813 |
+
"learning_rate": 2.2772585669781933e-05,
|
| 1814 |
+
"loss": 0.6451,
|
| 1815 |
+
"step": 2580
|
| 1816 |
+
},
|
| 1817 |
+
{
|
| 1818 |
+
"epoch": 0.7823591602476967,
|
| 1819 |
+
"grad_norm": 0.9061359038046752,
|
| 1820 |
+
"learning_rate": 2.2461059190031153e-05,
|
| 1821 |
+
"loss": 0.6287,
|
| 1822 |
+
"step": 2590
|
| 1823 |
+
},
|
| 1824 |
+
{
|
| 1825 |
+
"epoch": 0.7853798519861048,
|
| 1826 |
+
"grad_norm": 1.022842846924079,
|
| 1827 |
+
"learning_rate": 2.2149532710280375e-05,
|
| 1828 |
+
"loss": 0.6424,
|
| 1829 |
+
"step": 2600
|
| 1830 |
+
},
|
| 1831 |
+
{
|
| 1832 |
+
"epoch": 0.788400543724513,
|
| 1833 |
+
"grad_norm": 1.1644483297932762,
|
| 1834 |
+
"learning_rate": 2.1838006230529595e-05,
|
| 1835 |
+
"loss": 0.6165,
|
| 1836 |
+
"step": 2610
|
| 1837 |
+
},
|
| 1838 |
+
{
|
| 1839 |
+
"epoch": 0.791421235462921,
|
| 1840 |
+
"grad_norm": 0.901213276967422,
|
| 1841 |
+
"learning_rate": 2.1526479750778817e-05,
|
| 1842 |
+
"loss": 0.6631,
|
| 1843 |
+
"step": 2620
|
| 1844 |
+
},
|
| 1845 |
+
{
|
| 1846 |
+
"epoch": 0.7944419272013291,
|
| 1847 |
+
"grad_norm": 0.939968751090103,
|
| 1848 |
+
"learning_rate": 2.1214953271028037e-05,
|
| 1849 |
+
"loss": 0.6562,
|
| 1850 |
+
"step": 2630
|
| 1851 |
+
},
|
| 1852 |
+
{
|
| 1853 |
+
"epoch": 0.7974626189397372,
|
| 1854 |
+
"grad_norm": 0.9433346580971289,
|
| 1855 |
+
"learning_rate": 2.090342679127726e-05,
|
| 1856 |
+
"loss": 0.6181,
|
| 1857 |
+
"step": 2640
|
| 1858 |
+
},
|
| 1859 |
+
{
|
| 1860 |
+
"epoch": 0.8004833106781453,
|
| 1861 |
+
"grad_norm": 1.0194887755025552,
|
| 1862 |
+
"learning_rate": 2.059190031152648e-05,
|
| 1863 |
+
"loss": 0.6399,
|
| 1864 |
+
"step": 2650
|
| 1865 |
+
},
|
| 1866 |
+
{
|
| 1867 |
+
"epoch": 0.8035040024165534,
|
| 1868 |
+
"grad_norm": 1.1638844007159146,
|
| 1869 |
+
"learning_rate": 2.02803738317757e-05,
|
| 1870 |
+
"loss": 0.6491,
|
| 1871 |
+
"step": 2660
|
| 1872 |
+
},
|
| 1873 |
+
{
|
| 1874 |
+
"epoch": 0.8065246941549615,
|
| 1875 |
+
"grad_norm": 1.0276398667542848,
|
| 1876 |
+
"learning_rate": 1.996884735202492e-05,
|
| 1877 |
+
"loss": 0.6145,
|
| 1878 |
+
"step": 2670
|
| 1879 |
+
},
|
| 1880 |
+
{
|
| 1881 |
+
"epoch": 0.8095453858933696,
|
| 1882 |
+
"grad_norm": 0.8210616041641313,
|
| 1883 |
+
"learning_rate": 1.9657320872274143e-05,
|
| 1884 |
+
"loss": 0.6466,
|
| 1885 |
+
"step": 2680
|
| 1886 |
+
},
|
| 1887 |
+
{
|
| 1888 |
+
"epoch": 0.8125660776317777,
|
| 1889 |
+
"grad_norm": 1.093871978484254,
|
| 1890 |
+
"learning_rate": 1.9345794392523366e-05,
|
| 1891 |
+
"loss": 0.6194,
|
| 1892 |
+
"step": 2690
|
| 1893 |
+
},
|
| 1894 |
+
{
|
| 1895 |
+
"epoch": 0.8155867693701858,
|
| 1896 |
+
"grad_norm": 1.2620240032817829,
|
| 1897 |
+
"learning_rate": 1.9034267912772585e-05,
|
| 1898 |
+
"loss": 0.6183,
|
| 1899 |
+
"step": 2700
|
| 1900 |
+
},
|
| 1901 |
+
{
|
| 1902 |
+
"epoch": 0.8186074611085938,
|
| 1903 |
+
"grad_norm": 0.8779223909750274,
|
| 1904 |
+
"learning_rate": 1.8722741433021808e-05,
|
| 1905 |
+
"loss": 0.6586,
|
| 1906 |
+
"step": 2710
|
| 1907 |
+
},
|
| 1908 |
+
{
|
| 1909 |
+
"epoch": 0.821628152847002,
|
| 1910 |
+
"grad_norm": 0.7699492596332658,
|
| 1911 |
+
"learning_rate": 1.8411214953271027e-05,
|
| 1912 |
+
"loss": 0.5942,
|
| 1913 |
+
"step": 2720
|
| 1914 |
+
},
|
| 1915 |
+
{
|
| 1916 |
+
"epoch": 0.8246488445854101,
|
| 1917 |
+
"grad_norm": 1.1685873990863083,
|
| 1918 |
+
"learning_rate": 1.809968847352025e-05,
|
| 1919 |
+
"loss": 0.6443,
|
| 1920 |
+
"step": 2730
|
| 1921 |
+
},
|
| 1922 |
+
{
|
| 1923 |
+
"epoch": 0.8276695363238181,
|
| 1924 |
+
"grad_norm": 1.083337396850288,
|
| 1925 |
+
"learning_rate": 1.778816199376947e-05,
|
| 1926 |
+
"loss": 0.6482,
|
| 1927 |
+
"step": 2740
|
| 1928 |
+
},
|
| 1929 |
+
{
|
| 1930 |
+
"epoch": 0.8306902280622263,
|
| 1931 |
+
"grad_norm": 0.8063684272668464,
|
| 1932 |
+
"learning_rate": 1.7476635514018695e-05,
|
| 1933 |
+
"loss": 0.606,
|
| 1934 |
+
"step": 2750
|
| 1935 |
+
},
|
| 1936 |
+
{
|
| 1937 |
+
"epoch": 0.8337109198006344,
|
| 1938 |
+
"grad_norm": 0.824663714107508,
|
| 1939 |
+
"learning_rate": 1.7165109034267915e-05,
|
| 1940 |
+
"loss": 0.586,
|
| 1941 |
+
"step": 2760
|
| 1942 |
+
},
|
| 1943 |
+
{
|
| 1944 |
+
"epoch": 0.8367316115390424,
|
| 1945 |
+
"grad_norm": 0.9760183523742957,
|
| 1946 |
+
"learning_rate": 1.6853582554517137e-05,
|
| 1947 |
+
"loss": 0.5791,
|
| 1948 |
+
"step": 2770
|
| 1949 |
+
},
|
| 1950 |
+
{
|
| 1951 |
+
"epoch": 0.8397523032774505,
|
| 1952 |
+
"grad_norm": 1.0330407004022142,
|
| 1953 |
+
"learning_rate": 1.6542056074766357e-05,
|
| 1954 |
+
"loss": 0.6037,
|
| 1955 |
+
"step": 2780
|
| 1956 |
+
},
|
| 1957 |
+
{
|
| 1958 |
+
"epoch": 0.8427729950158587,
|
| 1959 |
+
"grad_norm": 0.9234529390674108,
|
| 1960 |
+
"learning_rate": 1.623052959501558e-05,
|
| 1961 |
+
"loss": 0.61,
|
| 1962 |
+
"step": 2790
|
| 1963 |
+
},
|
| 1964 |
+
{
|
| 1965 |
+
"epoch": 0.8457936867542667,
|
| 1966 |
+
"grad_norm": 1.0821057854566793,
|
| 1967 |
+
"learning_rate": 1.59190031152648e-05,
|
| 1968 |
+
"loss": 0.6463,
|
| 1969 |
+
"step": 2800
|
| 1970 |
+
},
|
| 1971 |
+
{
|
| 1972 |
+
"epoch": 0.8488143784926748,
|
| 1973 |
+
"grad_norm": 1.0890558497862515,
|
| 1974 |
+
"learning_rate": 1.560747663551402e-05,
|
| 1975 |
+
"loss": 0.6438,
|
| 1976 |
+
"step": 2810
|
| 1977 |
+
},
|
| 1978 |
+
{
|
| 1979 |
+
"epoch": 0.851835070231083,
|
| 1980 |
+
"grad_norm": 0.7374750576790412,
|
| 1981 |
+
"learning_rate": 1.529595015576324e-05,
|
| 1982 |
+
"loss": 0.6205,
|
| 1983 |
+
"step": 2820
|
| 1984 |
+
},
|
| 1985 |
+
{
|
| 1986 |
+
"epoch": 0.854855761969491,
|
| 1987 |
+
"grad_norm": 0.8883192786185565,
|
| 1988 |
+
"learning_rate": 1.4984423676012463e-05,
|
| 1989 |
+
"loss": 0.6249,
|
| 1990 |
+
"step": 2830
|
| 1991 |
+
},
|
| 1992 |
+
{
|
| 1993 |
+
"epoch": 0.8578764537078991,
|
| 1994 |
+
"grad_norm": 1.1008000896235184,
|
| 1995 |
+
"learning_rate": 1.4672897196261684e-05,
|
| 1996 |
+
"loss": 0.5964,
|
| 1997 |
+
"step": 2840
|
| 1998 |
+
},
|
| 1999 |
+
{
|
| 2000 |
+
"epoch": 0.8608971454463072,
|
| 2001 |
+
"grad_norm": 0.9588924680911977,
|
| 2002 |
+
"learning_rate": 1.4361370716510905e-05,
|
| 2003 |
+
"loss": 0.6246,
|
| 2004 |
+
"step": 2850
|
| 2005 |
+
},
|
| 2006 |
+
{
|
| 2007 |
+
"epoch": 0.8639178371847153,
|
| 2008 |
+
"grad_norm": 1.224312320705181,
|
| 2009 |
+
"learning_rate": 1.4049844236760126e-05,
|
| 2010 |
+
"loss": 0.6571,
|
| 2011 |
+
"step": 2860
|
| 2012 |
+
},
|
| 2013 |
+
{
|
| 2014 |
+
"epoch": 0.8669385289231234,
|
| 2015 |
+
"grad_norm": 0.9905124714238223,
|
| 2016 |
+
"learning_rate": 1.3738317757009347e-05,
|
| 2017 |
+
"loss": 0.6687,
|
| 2018 |
+
"step": 2870
|
| 2019 |
+
},
|
| 2020 |
+
{
|
| 2021 |
+
"epoch": 0.8699592206615315,
|
| 2022 |
+
"grad_norm": 0.8353778159477359,
|
| 2023 |
+
"learning_rate": 1.3426791277258568e-05,
|
| 2024 |
+
"loss": 0.6,
|
| 2025 |
+
"step": 2880
|
| 2026 |
+
},
|
| 2027 |
+
{
|
| 2028 |
+
"epoch": 0.8729799123999396,
|
| 2029 |
+
"grad_norm": 0.905312921332729,
|
| 2030 |
+
"learning_rate": 1.311526479750779e-05,
|
| 2031 |
+
"loss": 0.606,
|
| 2032 |
+
"step": 2890
|
| 2033 |
+
},
|
| 2034 |
+
{
|
| 2035 |
+
"epoch": 0.8760006041383477,
|
| 2036 |
+
"grad_norm": 0.8649231871193698,
|
| 2037 |
+
"learning_rate": 1.280373831775701e-05,
|
| 2038 |
+
"loss": 0.5743,
|
| 2039 |
+
"step": 2900
|
| 2040 |
+
},
|
| 2041 |
+
{
|
| 2042 |
+
"epoch": 0.8790212958767558,
|
| 2043 |
+
"grad_norm": 1.0254699594493124,
|
| 2044 |
+
"learning_rate": 1.2492211838006231e-05,
|
| 2045 |
+
"loss": 0.6204,
|
| 2046 |
+
"step": 2910
|
| 2047 |
+
},
|
| 2048 |
+
{
|
| 2049 |
+
"epoch": 0.8820419876151638,
|
| 2050 |
+
"grad_norm": 1.0278406308415082,
|
| 2051 |
+
"learning_rate": 1.2180685358255452e-05,
|
| 2052 |
+
"loss": 0.5853,
|
| 2053 |
+
"step": 2920
|
| 2054 |
+
},
|
| 2055 |
+
{
|
| 2056 |
+
"epoch": 0.885062679353572,
|
| 2057 |
+
"grad_norm": 0.9990676256188065,
|
| 2058 |
+
"learning_rate": 1.1869158878504673e-05,
|
| 2059 |
+
"loss": 0.6224,
|
| 2060 |
+
"step": 2930
|
| 2061 |
+
},
|
| 2062 |
+
{
|
| 2063 |
+
"epoch": 0.8880833710919801,
|
| 2064 |
+
"grad_norm": 1.0949270827311683,
|
| 2065 |
+
"learning_rate": 1.1557632398753894e-05,
|
| 2066 |
+
"loss": 0.6125,
|
| 2067 |
+
"step": 2940
|
| 2068 |
+
},
|
| 2069 |
+
{
|
| 2070 |
+
"epoch": 0.8911040628303881,
|
| 2071 |
+
"grad_norm": 1.2006202540115667,
|
| 2072 |
+
"learning_rate": 1.1246105919003117e-05,
|
| 2073 |
+
"loss": 0.6392,
|
| 2074 |
+
"step": 2950
|
| 2075 |
+
},
|
| 2076 |
+
{
|
| 2077 |
+
"epoch": 0.8941247545687963,
|
| 2078 |
+
"grad_norm": 1.1187121516645784,
|
| 2079 |
+
"learning_rate": 1.0934579439252338e-05,
|
| 2080 |
+
"loss": 0.6357,
|
| 2081 |
+
"step": 2960
|
| 2082 |
+
},
|
| 2083 |
+
{
|
| 2084 |
+
"epoch": 0.8971454463072044,
|
| 2085 |
+
"grad_norm": 0.8638598049799623,
|
| 2086 |
+
"learning_rate": 1.0623052959501559e-05,
|
| 2087 |
+
"loss": 0.5989,
|
| 2088 |
+
"step": 2970
|
| 2089 |
+
},
|
| 2090 |
+
{
|
| 2091 |
+
"epoch": 0.9001661380456124,
|
| 2092 |
+
"grad_norm": 0.9297895550418135,
|
| 2093 |
+
"learning_rate": 1.031152647975078e-05,
|
| 2094 |
+
"loss": 0.6528,
|
| 2095 |
+
"step": 2980
|
| 2096 |
+
},
|
| 2097 |
+
{
|
| 2098 |
+
"epoch": 0.9031868297840205,
|
| 2099 |
+
"grad_norm": 0.9615985258637098,
|
| 2100 |
+
"learning_rate": 1e-05,
|
| 2101 |
+
"loss": 0.6123,
|
| 2102 |
+
"step": 2990
|
| 2103 |
+
},
|
| 2104 |
+
{
|
| 2105 |
+
"epoch": 0.9062075215224287,
|
| 2106 |
+
"grad_norm": 0.8309736923021925,
|
| 2107 |
+
"learning_rate": 9.688473520249222e-06,
|
| 2108 |
+
"loss": 0.6459,
|
| 2109 |
+
"step": 3000
|
| 2110 |
+
},
|
| 2111 |
+
{
|
| 2112 |
+
"epoch": 0.9092282132608367,
|
| 2113 |
+
"grad_norm": 0.9815142795709194,
|
| 2114 |
+
"learning_rate": 9.376947040498443e-06,
|
| 2115 |
+
"loss": 0.571,
|
| 2116 |
+
"step": 3010
|
| 2117 |
+
},
|
| 2118 |
+
{
|
| 2119 |
+
"epoch": 0.9122489049992448,
|
| 2120 |
+
"grad_norm": 0.8008160564268554,
|
| 2121 |
+
"learning_rate": 9.065420560747664e-06,
|
| 2122 |
+
"loss": 0.6037,
|
| 2123 |
+
"step": 3020
|
| 2124 |
+
},
|
| 2125 |
+
{
|
| 2126 |
+
"epoch": 0.915269596737653,
|
| 2127 |
+
"grad_norm": 1.0819593992572683,
|
| 2128 |
+
"learning_rate": 8.753894080996885e-06,
|
| 2129 |
+
"loss": 0.5869,
|
| 2130 |
+
"step": 3030
|
| 2131 |
+
},
|
| 2132 |
+
{
|
| 2133 |
+
"epoch": 0.918290288476061,
|
| 2134 |
+
"grad_norm": 0.8665948995270025,
|
| 2135 |
+
"learning_rate": 8.442367601246107e-06,
|
| 2136 |
+
"loss": 0.611,
|
| 2137 |
+
"step": 3040
|
| 2138 |
+
},
|
| 2139 |
+
{
|
| 2140 |
+
"epoch": 0.9213109802144691,
|
| 2141 |
+
"grad_norm": 0.8583950697207722,
|
| 2142 |
+
"learning_rate": 8.130841121495328e-06,
|
| 2143 |
+
"loss": 0.579,
|
| 2144 |
+
"step": 3050
|
| 2145 |
+
},
|
| 2146 |
+
{
|
| 2147 |
+
"epoch": 0.9243316719528772,
|
| 2148 |
+
"grad_norm": 0.921012636719109,
|
| 2149 |
+
"learning_rate": 7.81931464174455e-06,
|
| 2150 |
+
"loss": 0.6101,
|
| 2151 |
+
"step": 3060
|
| 2152 |
+
},
|
| 2153 |
+
{
|
| 2154 |
+
"epoch": 0.9273523636912853,
|
| 2155 |
+
"grad_norm": 0.8604105683200605,
|
| 2156 |
+
"learning_rate": 7.5077881619937705e-06,
|
| 2157 |
+
"loss": 0.585,
|
| 2158 |
+
"step": 3070
|
| 2159 |
+
},
|
| 2160 |
+
{
|
| 2161 |
+
"epoch": 0.9303730554296934,
|
| 2162 |
+
"grad_norm": 0.9924409498073824,
|
| 2163 |
+
"learning_rate": 7.1962616822429915e-06,
|
| 2164 |
+
"loss": 0.5966,
|
| 2165 |
+
"step": 3080
|
| 2166 |
+
},
|
| 2167 |
+
{
|
| 2168 |
+
"epoch": 0.9333937471681015,
|
| 2169 |
+
"grad_norm": 0.8554914123213949,
|
| 2170 |
+
"learning_rate": 6.8847352024922125e-06,
|
| 2171 |
+
"loss": 0.6001,
|
| 2172 |
+
"step": 3090
|
| 2173 |
+
},
|
| 2174 |
+
{
|
| 2175 |
+
"epoch": 0.9364144389065095,
|
| 2176 |
+
"grad_norm": 1.0189344509024982,
|
| 2177 |
+
"learning_rate": 6.5732087227414335e-06,
|
| 2178 |
+
"loss": 0.6072,
|
| 2179 |
+
"step": 3100
|
| 2180 |
+
},
|
| 2181 |
+
{
|
| 2182 |
+
"epoch": 0.9394351306449177,
|
| 2183 |
+
"grad_norm": 0.7551552875967538,
|
| 2184 |
+
"learning_rate": 6.2616822429906544e-06,
|
| 2185 |
+
"loss": 0.6195,
|
| 2186 |
+
"step": 3110
|
| 2187 |
+
},
|
| 2188 |
+
{
|
| 2189 |
+
"epoch": 0.9424558223833258,
|
| 2190 |
+
"grad_norm": 0.7876647280151428,
|
| 2191 |
+
"learning_rate": 5.9501557632398754e-06,
|
| 2192 |
+
"loss": 0.5982,
|
| 2193 |
+
"step": 3120
|
| 2194 |
+
},
|
| 2195 |
+
{
|
| 2196 |
+
"epoch": 0.9454765141217338,
|
| 2197 |
+
"grad_norm": 1.0198335362611242,
|
| 2198 |
+
"learning_rate": 5.638629283489097e-06,
|
| 2199 |
+
"loss": 0.6261,
|
| 2200 |
+
"step": 3130
|
| 2201 |
+
},
|
| 2202 |
+
{
|
| 2203 |
+
"epoch": 0.948497205860142,
|
| 2204 |
+
"grad_norm": 1.0604464411076657,
|
| 2205 |
+
"learning_rate": 5.327102803738318e-06,
|
| 2206 |
+
"loss": 0.6302,
|
| 2207 |
+
"step": 3140
|
| 2208 |
+
},
|
| 2209 |
+
{
|
| 2210 |
+
"epoch": 0.9515178975985501,
|
| 2211 |
+
"grad_norm": 0.951744438195523,
|
| 2212 |
+
"learning_rate": 5.015576323987539e-06,
|
| 2213 |
+
"loss": 0.6093,
|
| 2214 |
+
"step": 3150
|
| 2215 |
+
},
|
| 2216 |
+
{
|
| 2217 |
+
"epoch": 0.9545385893369581,
|
| 2218 |
+
"grad_norm": 0.8133396952083934,
|
| 2219 |
+
"learning_rate": 4.70404984423676e-06,
|
| 2220 |
+
"loss": 0.5976,
|
| 2221 |
+
"step": 3160
|
| 2222 |
+
},
|
| 2223 |
+
{
|
| 2224 |
+
"epoch": 0.9575592810753663,
|
| 2225 |
+
"grad_norm": 0.9048179249325315,
|
| 2226 |
+
"learning_rate": 4.392523364485981e-06,
|
| 2227 |
+
"loss": 0.5917,
|
| 2228 |
+
"step": 3170
|
| 2229 |
+
},
|
| 2230 |
+
{
|
| 2231 |
+
"epoch": 0.9605799728137744,
|
| 2232 |
+
"grad_norm": 0.9211845494853066,
|
| 2233 |
+
"learning_rate": 4.080996884735202e-06,
|
| 2234 |
+
"loss": 0.6598,
|
| 2235 |
+
"step": 3180
|
| 2236 |
+
},
|
| 2237 |
+
{
|
| 2238 |
+
"epoch": 0.9636006645521824,
|
| 2239 |
+
"grad_norm": 1.1245992131314346,
|
| 2240 |
+
"learning_rate": 3.7694704049844237e-06,
|
| 2241 |
+
"loss": 0.6366,
|
| 2242 |
+
"step": 3190
|
| 2243 |
+
},
|
| 2244 |
+
{
|
| 2245 |
+
"epoch": 0.9666213562905905,
|
| 2246 |
+
"grad_norm": 0.8622643092177033,
|
| 2247 |
+
"learning_rate": 3.4579439252336447e-06,
|
| 2248 |
+
"loss": 0.6032,
|
| 2249 |
+
"step": 3200
|
| 2250 |
+
},
|
| 2251 |
+
{
|
| 2252 |
+
"epoch": 0.9696420480289987,
|
| 2253 |
+
"grad_norm": 0.9757245422407422,
|
| 2254 |
+
"learning_rate": 3.146417445482866e-06,
|
| 2255 |
+
"loss": 0.6109,
|
| 2256 |
+
"step": 3210
|
| 2257 |
+
},
|
| 2258 |
+
{
|
| 2259 |
+
"epoch": 0.9726627397674067,
|
| 2260 |
+
"grad_norm": 0.9177066421934855,
|
| 2261 |
+
"learning_rate": 2.8348909657320875e-06,
|
| 2262 |
+
"loss": 0.5852,
|
| 2263 |
+
"step": 3220
|
| 2264 |
+
},
|
| 2265 |
+
{
|
| 2266 |
+
"epoch": 0.9756834315058148,
|
| 2267 |
+
"grad_norm": 1.0589749886784343,
|
| 2268 |
+
"learning_rate": 2.5233644859813085e-06,
|
| 2269 |
+
"loss": 0.6162,
|
| 2270 |
+
"step": 3230
|
| 2271 |
+
},
|
| 2272 |
+
{
|
| 2273 |
+
"epoch": 0.978704123244223,
|
| 2274 |
+
"grad_norm": 0.8439326188110577,
|
| 2275 |
+
"learning_rate": 2.21183800623053e-06,
|
| 2276 |
+
"loss": 0.5679,
|
| 2277 |
+
"step": 3240
|
| 2278 |
+
},
|
| 2279 |
+
{
|
| 2280 |
+
"epoch": 0.981724814982631,
|
| 2281 |
+
"grad_norm": 0.8843365815693212,
|
| 2282 |
+
"learning_rate": 1.900311526479751e-06,
|
| 2283 |
+
"loss": 0.5856,
|
| 2284 |
+
"step": 3250
|
| 2285 |
+
},
|
| 2286 |
+
{
|
| 2287 |
+
"epoch": 0.9847455067210391,
|
| 2288 |
+
"grad_norm": 0.91811081864421,
|
| 2289 |
+
"learning_rate": 1.588785046728972e-06,
|
| 2290 |
+
"loss": 0.5229,
|
| 2291 |
+
"step": 3260
|
| 2292 |
+
},
|
| 2293 |
+
{
|
| 2294 |
+
"epoch": 0.9877661984594472,
|
| 2295 |
+
"grad_norm": 0.8575199450142503,
|
| 2296 |
+
"learning_rate": 1.2772585669781931e-06,
|
| 2297 |
+
"loss": 0.6195,
|
| 2298 |
+
"step": 3270
|
| 2299 |
+
},
|
| 2300 |
+
{
|
| 2301 |
+
"epoch": 0.9907868901978553,
|
| 2302 |
+
"grad_norm": 0.9360300893347252,
|
| 2303 |
+
"learning_rate": 9.657320872274145e-07,
|
| 2304 |
+
"loss": 0.6176,
|
| 2305 |
+
"step": 3280
|
| 2306 |
+
},
|
| 2307 |
+
{
|
| 2308 |
+
"epoch": 0.9938075819362634,
|
| 2309 |
+
"grad_norm": 0.9000581153467687,
|
| 2310 |
+
"learning_rate": 6.542056074766356e-07,
|
| 2311 |
+
"loss": 0.5858,
|
| 2312 |
+
"step": 3290
|
| 2313 |
+
},
|
| 2314 |
+
{
|
| 2315 |
+
"epoch": 0.9968282736746715,
|
| 2316 |
+
"grad_norm": 0.9999683040058044,
|
| 2317 |
+
"learning_rate": 3.4267912772585674e-07,
|
| 2318 |
+
"loss": 0.5731,
|
| 2319 |
+
"step": 3300
|
| 2320 |
+
},
|
| 2321 |
+
{
|
| 2322 |
+
"epoch": 0.9998489654130795,
|
| 2323 |
+
"grad_norm": 0.9739409873164033,
|
| 2324 |
+
"learning_rate": 3.1152647975077883e-08,
|
| 2325 |
+
"loss": 0.6497,
|
| 2326 |
+
"step": 3310
|
| 2327 |
+
}
|
| 2328 |
+
],
|
| 2329 |
+
"logging_steps": 10,
|
| 2330 |
+
"max_steps": 3310,
|
| 2331 |
+
"num_input_tokens_seen": 0,
|
| 2332 |
+
"num_train_epochs": 1,
|
| 2333 |
+
"save_steps": 100,
|
| 2334 |
+
"stateful_callbacks": {
|
| 2335 |
+
"TrainerControl": {
|
| 2336 |
+
"args": {
|
| 2337 |
+
"should_epoch_stop": false,
|
| 2338 |
+
"should_evaluate": false,
|
| 2339 |
+
"should_log": false,
|
| 2340 |
+
"should_save": true,
|
| 2341 |
+
"should_training_stop": true
|
| 2342 |
+
},
|
| 2343 |
+
"attributes": {}
|
| 2344 |
+
}
|
| 2345 |
+
},
|
| 2346 |
+
"total_flos": 0.0,
|
| 2347 |
+
"train_batch_size": 32,
|
| 2348 |
+
"trial_name": null,
|
| 2349 |
+
"trial_params": null
|
| 2350 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c7eb5e6738373140f9ceaea10b91147f6aead460fed27c539494474ad32d8c3f
|
| 3 |
+
size 7683
|
zero_to_fp32.py
ADDED
|
@@ -0,0 +1,760 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 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:
|
| 14 |
+
# python zero_to_fp32.py . output_dir/
|
| 15 |
+
# or
|
| 16 |
+
# python zero_to_fp32.py . output_dir/ --safe_serialization
|
| 17 |
+
|
| 18 |
+
import argparse
|
| 19 |
+
import torch
|
| 20 |
+
import glob
|
| 21 |
+
import math
|
| 22 |
+
import os
|
| 23 |
+
import re
|
| 24 |
+
import gc
|
| 25 |
+
import json
|
| 26 |
+
import numpy as np
|
| 27 |
+
from tqdm import tqdm
|
| 28 |
+
from collections import OrderedDict
|
| 29 |
+
from dataclasses import dataclass
|
| 30 |
+
|
| 31 |
+
# while this script doesn't use deepspeed to recover data, since the checkpoints are pickled with
|
| 32 |
+
# DeepSpeed data structures it has to be available in the current python environment.
|
| 33 |
+
from deepspeed.utils import logger
|
| 34 |
+
from deepspeed.checkpoint.constants import (DS_VERSION, OPTIMIZER_STATE_DICT, SINGLE_PARTITION_OF_FP32_GROUPS,
|
| 35 |
+
FP32_FLAT_GROUPS, ZERO_STAGE, PARTITION_COUNT, PARAM_SHAPES, BUFFER_NAMES,
|
| 36 |
+
FROZEN_PARAM_SHAPES, FROZEN_PARAM_FRAGMENTS)
|
| 37 |
+
|
| 38 |
+
|
| 39 |
+
@dataclass
|
| 40 |
+
class zero_model_state:
|
| 41 |
+
buffers: dict()
|
| 42 |
+
param_shapes: dict()
|
| 43 |
+
shared_params: list
|
| 44 |
+
ds_version: int
|
| 45 |
+
frozen_param_shapes: dict()
|
| 46 |
+
frozen_param_fragments: dict()
|
| 47 |
+
|
| 48 |
+
|
| 49 |
+
debug = 0
|
| 50 |
+
|
| 51 |
+
# load to cpu
|
| 52 |
+
device = torch.device('cpu')
|
| 53 |
+
|
| 54 |
+
|
| 55 |
+
def atoi(text):
|
| 56 |
+
return int(text) if text.isdigit() else text
|
| 57 |
+
|
| 58 |
+
|
| 59 |
+
def natural_keys(text):
|
| 60 |
+
'''
|
| 61 |
+
alist.sort(key=natural_keys) sorts in human order
|
| 62 |
+
http://nedbatchelder.com/blog/200712/human_sorting.html
|
| 63 |
+
(See Toothy's implementation in the comments)
|
| 64 |
+
'''
|
| 65 |
+
return [atoi(c) for c in re.split(r'(\d+)', text)]
|
| 66 |
+
|
| 67 |
+
|
| 68 |
+
def get_model_state_file(checkpoint_dir, zero_stage):
|
| 69 |
+
if not os.path.isdir(checkpoint_dir):
|
| 70 |
+
raise FileNotFoundError(f"Directory '{checkpoint_dir}' doesn't exist")
|
| 71 |
+
|
| 72 |
+
# there should be only one file
|
| 73 |
+
if zero_stage <= 2:
|
| 74 |
+
file = os.path.join(checkpoint_dir, "mp_rank_00_model_states.pt")
|
| 75 |
+
elif zero_stage == 3:
|
| 76 |
+
file = os.path.join(checkpoint_dir, "zero_pp_rank_0_mp_rank_00_model_states.pt")
|
| 77 |
+
|
| 78 |
+
if not os.path.exists(file):
|
| 79 |
+
raise FileNotFoundError(f"can't find model states file at '{file}'")
|
| 80 |
+
|
| 81 |
+
return file
|
| 82 |
+
|
| 83 |
+
|
| 84 |
+
def get_checkpoint_files(checkpoint_dir, glob_pattern):
|
| 85 |
+
# XXX: need to test that this simple glob rule works for multi-node setup too
|
| 86 |
+
ckpt_files = sorted(glob.glob(os.path.join(checkpoint_dir, glob_pattern)), key=natural_keys)
|
| 87 |
+
|
| 88 |
+
if len(ckpt_files) == 0:
|
| 89 |
+
raise FileNotFoundError(f"can't find {glob_pattern} files in directory '{checkpoint_dir}'")
|
| 90 |
+
|
| 91 |
+
return ckpt_files
|
| 92 |
+
|
| 93 |
+
|
| 94 |
+
def get_optim_files(checkpoint_dir):
|
| 95 |
+
return get_checkpoint_files(checkpoint_dir, "*_optim_states.pt")
|
| 96 |
+
|
| 97 |
+
|
| 98 |
+
def get_model_state_files(checkpoint_dir):
|
| 99 |
+
return get_checkpoint_files(checkpoint_dir, "*_model_states.pt")
|
| 100 |
+
|
| 101 |
+
|
| 102 |
+
def parse_model_states(files):
|
| 103 |
+
zero_model_states = []
|
| 104 |
+
for file in files:
|
| 105 |
+
state_dict = torch.load(file, map_location=device, weights_only=False)
|
| 106 |
+
|
| 107 |
+
if BUFFER_NAMES not in state_dict:
|
| 108 |
+
raise ValueError(f"{file} is not a model state checkpoint")
|
| 109 |
+
buffer_names = state_dict[BUFFER_NAMES]
|
| 110 |
+
if debug:
|
| 111 |
+
print("Found buffers:", buffer_names)
|
| 112 |
+
|
| 113 |
+
# recover just the buffers while restoring them to fp32 if they were saved in fp16
|
| 114 |
+
buffers = {k: v.float() for k, v in state_dict["module"].items() if k in buffer_names}
|
| 115 |
+
param_shapes = state_dict[PARAM_SHAPES]
|
| 116 |
+
|
| 117 |
+
# collect parameters that are included in param_shapes
|
| 118 |
+
param_names = []
|
| 119 |
+
for s in param_shapes:
|
| 120 |
+
for name in s.keys():
|
| 121 |
+
param_names.append(name)
|
| 122 |
+
|
| 123 |
+
# update with frozen parameters
|
| 124 |
+
frozen_param_shapes = state_dict.get(FROZEN_PARAM_SHAPES, None)
|
| 125 |
+
if frozen_param_shapes is not None:
|
| 126 |
+
if debug:
|
| 127 |
+
print(f"Found frozen_param_shapes: {frozen_param_shapes}")
|
| 128 |
+
param_names += list(frozen_param_shapes.keys())
|
| 129 |
+
|
| 130 |
+
# handle shared params
|
| 131 |
+
shared_params = [[k, v] for k, v in state_dict["shared_params"].items()]
|
| 132 |
+
|
| 133 |
+
ds_version = state_dict.get(DS_VERSION, None)
|
| 134 |
+
|
| 135 |
+
frozen_param_fragments = state_dict.get(FROZEN_PARAM_FRAGMENTS, None)
|
| 136 |
+
|
| 137 |
+
z_model_state = zero_model_state(buffers=buffers,
|
| 138 |
+
param_shapes=param_shapes,
|
| 139 |
+
shared_params=shared_params,
|
| 140 |
+
ds_version=ds_version,
|
| 141 |
+
frozen_param_shapes=frozen_param_shapes,
|
| 142 |
+
frozen_param_fragments=frozen_param_fragments)
|
| 143 |
+
zero_model_states.append(z_model_state)
|
| 144 |
+
|
| 145 |
+
return zero_model_states
|
| 146 |
+
|
| 147 |
+
|
| 148 |
+
def parse_optim_states(files, ds_checkpoint_dir):
|
| 149 |
+
total_files = len(files)
|
| 150 |
+
state_dicts = []
|
| 151 |
+
for f in tqdm(files, desc='Loading checkpoint shards'):
|
| 152 |
+
state_dict = torch.load(f, map_location=device, mmap=True, weights_only=False)
|
| 153 |
+
# immediately discard the potentially huge 2 optimizer states as we only care for fp32 master weights
|
| 154 |
+
# and also handle the case where it was already removed by another helper script
|
| 155 |
+
state_dict["optimizer_state_dict"].pop("optimizer_state_dict", None)
|
| 156 |
+
state_dicts.append(state_dict)
|
| 157 |
+
|
| 158 |
+
if not ZERO_STAGE in state_dicts[0][OPTIMIZER_STATE_DICT]:
|
| 159 |
+
raise ValueError(f"{files[0]} is not a zero checkpoint")
|
| 160 |
+
zero_stage = state_dicts[0][OPTIMIZER_STATE_DICT][ZERO_STAGE]
|
| 161 |
+
world_size = state_dicts[0][OPTIMIZER_STATE_DICT][PARTITION_COUNT]
|
| 162 |
+
|
| 163 |
+
# For ZeRO-2 each param group can have different partition_count as data parallelism for expert
|
| 164 |
+
# parameters can be different from data parallelism for non-expert parameters. So we can just
|
| 165 |
+
# use the max of the partition_count to get the dp world_size.
|
| 166 |
+
|
| 167 |
+
if type(world_size) is list:
|
| 168 |
+
world_size = max(world_size)
|
| 169 |
+
|
| 170 |
+
if world_size != total_files:
|
| 171 |
+
raise ValueError(
|
| 172 |
+
f"Expected {world_size} of '*_optim_states.pt' under '{ds_checkpoint_dir}' but found {total_files} files. "
|
| 173 |
+
"Possibly due to an overwrite of an old checkpoint, or a checkpoint didn't get saved by one or more processes."
|
| 174 |
+
)
|
| 175 |
+
|
| 176 |
+
# the groups are named differently in each stage
|
| 177 |
+
if zero_stage <= 2:
|
| 178 |
+
fp32_groups_key = SINGLE_PARTITION_OF_FP32_GROUPS
|
| 179 |
+
elif zero_stage == 3:
|
| 180 |
+
fp32_groups_key = FP32_FLAT_GROUPS
|
| 181 |
+
else:
|
| 182 |
+
raise ValueError(f"unknown zero stage {zero_stage}")
|
| 183 |
+
|
| 184 |
+
fp32_flat_groups = [state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key] for i in range(len(state_dicts))]
|
| 185 |
+
return zero_stage, world_size, fp32_flat_groups
|
| 186 |
+
|
| 187 |
+
|
| 188 |
+
def _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters):
|
| 189 |
+
"""
|
| 190 |
+
Returns fp32 state_dict reconstructed from ds checkpoint
|
| 191 |
+
|
| 192 |
+
Args:
|
| 193 |
+
- ``ds_checkpoint_dir``: path to the deepspeed checkpoint folder (where the optimizer files are)
|
| 194 |
+
|
| 195 |
+
"""
|
| 196 |
+
print(f"Processing zero checkpoint '{ds_checkpoint_dir}'")
|
| 197 |
+
|
| 198 |
+
optim_files = get_optim_files(ds_checkpoint_dir)
|
| 199 |
+
zero_stage, world_size, fp32_flat_groups = parse_optim_states(optim_files, ds_checkpoint_dir)
|
| 200 |
+
print(f"Detected checkpoint of type zero stage {zero_stage}, world_size: {world_size}")
|
| 201 |
+
|
| 202 |
+
model_files = get_model_state_files(ds_checkpoint_dir)
|
| 203 |
+
|
| 204 |
+
zero_model_states = parse_model_states(model_files)
|
| 205 |
+
print(f'Parsing checkpoint created by deepspeed=={zero_model_states[0].ds_version}')
|
| 206 |
+
|
| 207 |
+
if zero_stage <= 2:
|
| 208 |
+
return _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states,
|
| 209 |
+
exclude_frozen_parameters)
|
| 210 |
+
elif zero_stage == 3:
|
| 211 |
+
return _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states,
|
| 212 |
+
exclude_frozen_parameters)
|
| 213 |
+
|
| 214 |
+
|
| 215 |
+
def _zero2_merge_frozen_params(state_dict, zero_model_states):
|
| 216 |
+
if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0:
|
| 217 |
+
return
|
| 218 |
+
|
| 219 |
+
frozen_param_shapes = zero_model_states[0].frozen_param_shapes
|
| 220 |
+
frozen_param_fragments = zero_model_states[0].frozen_param_fragments
|
| 221 |
+
|
| 222 |
+
if debug:
|
| 223 |
+
num_elem = sum(s.numel() for s in frozen_param_shapes.values())
|
| 224 |
+
print(f'rank 0: {FROZEN_PARAM_SHAPES}.numel = {num_elem}')
|
| 225 |
+
|
| 226 |
+
wanted_params = len(frozen_param_shapes)
|
| 227 |
+
wanted_numel = sum(s.numel() for s in frozen_param_shapes.values())
|
| 228 |
+
avail_numel = sum([p.numel() for p in frozen_param_fragments.values()])
|
| 229 |
+
print(f'Frozen params: Have {avail_numel} numels to process.')
|
| 230 |
+
print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params')
|
| 231 |
+
|
| 232 |
+
total_params = 0
|
| 233 |
+
total_numel = 0
|
| 234 |
+
for name, shape in frozen_param_shapes.items():
|
| 235 |
+
total_params += 1
|
| 236 |
+
unpartitioned_numel = shape.numel()
|
| 237 |
+
total_numel += unpartitioned_numel
|
| 238 |
+
|
| 239 |
+
state_dict[name] = frozen_param_fragments[name]
|
| 240 |
+
|
| 241 |
+
if debug:
|
| 242 |
+
print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ")
|
| 243 |
+
|
| 244 |
+
print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements")
|
| 245 |
+
|
| 246 |
+
|
| 247 |
+
def _has_callable(obj, fn):
|
| 248 |
+
attr = getattr(obj, fn, None)
|
| 249 |
+
return callable(attr)
|
| 250 |
+
|
| 251 |
+
|
| 252 |
+
def _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states):
|
| 253 |
+
param_shapes = zero_model_states[0].param_shapes
|
| 254 |
+
|
| 255 |
+
# Reconstruction protocol:
|
| 256 |
+
#
|
| 257 |
+
# XXX: document this
|
| 258 |
+
|
| 259 |
+
if debug:
|
| 260 |
+
for i in range(world_size):
|
| 261 |
+
for j in range(len(fp32_flat_groups[0])):
|
| 262 |
+
print(f"{FP32_FLAT_GROUPS}[{i}][{j}].shape={fp32_flat_groups[i][j].shape}")
|
| 263 |
+
|
| 264 |
+
# XXX: memory usage doubles here (zero2)
|
| 265 |
+
num_param_groups = len(fp32_flat_groups[0])
|
| 266 |
+
merged_single_partition_of_fp32_groups = []
|
| 267 |
+
for i in range(num_param_groups):
|
| 268 |
+
merged_partitions = [sd[i] for sd in fp32_flat_groups]
|
| 269 |
+
full_single_fp32_vector = torch.cat(merged_partitions, 0)
|
| 270 |
+
merged_single_partition_of_fp32_groups.append(full_single_fp32_vector)
|
| 271 |
+
avail_numel = sum(
|
| 272 |
+
[full_single_fp32_vector.numel() for full_single_fp32_vector in merged_single_partition_of_fp32_groups])
|
| 273 |
+
|
| 274 |
+
if debug:
|
| 275 |
+
wanted_params = sum([len(shapes) for shapes in param_shapes])
|
| 276 |
+
wanted_numel = sum([sum(shape.numel() for shape in shapes.values()) for shapes in param_shapes])
|
| 277 |
+
# not asserting if there is a mismatch due to possible padding
|
| 278 |
+
print(f"Have {avail_numel} numels to process.")
|
| 279 |
+
print(f"Need {wanted_numel} numels in {wanted_params} params.")
|
| 280 |
+
|
| 281 |
+
# params
|
| 282 |
+
# XXX: for huge models that can't fit into the host's RAM we will have to recode this to support
|
| 283 |
+
# out-of-core computing solution
|
| 284 |
+
total_numel = 0
|
| 285 |
+
total_params = 0
|
| 286 |
+
for shapes, full_single_fp32_vector in zip(param_shapes, merged_single_partition_of_fp32_groups):
|
| 287 |
+
offset = 0
|
| 288 |
+
avail_numel = full_single_fp32_vector.numel()
|
| 289 |
+
for name, shape in shapes.items():
|
| 290 |
+
|
| 291 |
+
unpartitioned_numel = shape.numel() if _has_callable(shape, 'numel') else math.prod(shape)
|
| 292 |
+
total_numel += unpartitioned_numel
|
| 293 |
+
total_params += 1
|
| 294 |
+
|
| 295 |
+
if debug:
|
| 296 |
+
print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ")
|
| 297 |
+
state_dict[name] = full_single_fp32_vector.narrow(0, offset, unpartitioned_numel).view(shape)
|
| 298 |
+
offset += unpartitioned_numel
|
| 299 |
+
|
| 300 |
+
# Z2 started to align to 2*world_size to improve nccl performance. Therefore both offset and
|
| 301 |
+
# avail_numel can differ by anywhere between 0..2*world_size. Due to two unrelated complex
|
| 302 |
+
# paddings performed in the code it's almost impossible to predict the exact numbers w/o the
|
| 303 |
+
# live optimizer object, so we are checking that the numbers are within the right range
|
| 304 |
+
align_to = 2 * world_size
|
| 305 |
+
|
| 306 |
+
def zero2_align(x):
|
| 307 |
+
return align_to * math.ceil(x / align_to)
|
| 308 |
+
|
| 309 |
+
if debug:
|
| 310 |
+
print(f"original offset={offset}, avail_numel={avail_numel}")
|
| 311 |
+
|
| 312 |
+
offset = zero2_align(offset)
|
| 313 |
+
avail_numel = zero2_align(avail_numel)
|
| 314 |
+
|
| 315 |
+
if debug:
|
| 316 |
+
print(f"aligned offset={offset}, avail_numel={avail_numel}")
|
| 317 |
+
|
| 318 |
+
# Sanity check
|
| 319 |
+
if offset != avail_numel:
|
| 320 |
+
raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong")
|
| 321 |
+
|
| 322 |
+
print(f"Reconstructed fp32 state dict with {total_params} params {total_numel} elements")
|
| 323 |
+
|
| 324 |
+
|
| 325 |
+
def _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states,
|
| 326 |
+
exclude_frozen_parameters):
|
| 327 |
+
state_dict = OrderedDict()
|
| 328 |
+
|
| 329 |
+
# buffers
|
| 330 |
+
buffers = zero_model_states[0].buffers
|
| 331 |
+
state_dict.update(buffers)
|
| 332 |
+
if debug:
|
| 333 |
+
print(f"added {len(buffers)} buffers")
|
| 334 |
+
|
| 335 |
+
if not exclude_frozen_parameters:
|
| 336 |
+
_zero2_merge_frozen_params(state_dict, zero_model_states)
|
| 337 |
+
|
| 338 |
+
_zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states)
|
| 339 |
+
|
| 340 |
+
# recover shared parameters
|
| 341 |
+
for pair in zero_model_states[0].shared_params:
|
| 342 |
+
if pair[1] in state_dict:
|
| 343 |
+
state_dict[pair[0]] = state_dict[pair[1]]
|
| 344 |
+
|
| 345 |
+
return state_dict
|
| 346 |
+
|
| 347 |
+
|
| 348 |
+
def zero3_partitioned_param_info(unpartitioned_numel, world_size):
|
| 349 |
+
remainder = unpartitioned_numel % world_size
|
| 350 |
+
padding_numel = (world_size - remainder) if remainder else 0
|
| 351 |
+
partitioned_numel = math.ceil(unpartitioned_numel / world_size)
|
| 352 |
+
return partitioned_numel, padding_numel
|
| 353 |
+
|
| 354 |
+
|
| 355 |
+
def _zero3_merge_frozen_params(state_dict, world_size, zero_model_states):
|
| 356 |
+
if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0:
|
| 357 |
+
return
|
| 358 |
+
|
| 359 |
+
if debug:
|
| 360 |
+
for i in range(world_size):
|
| 361 |
+
num_elem = sum(s.numel() for s in zero_model_states[i].frozen_param_fragments.values())
|
| 362 |
+
print(f'rank {i}: {FROZEN_PARAM_SHAPES}.numel = {num_elem}')
|
| 363 |
+
|
| 364 |
+
frozen_param_shapes = zero_model_states[0].frozen_param_shapes
|
| 365 |
+
wanted_params = len(frozen_param_shapes)
|
| 366 |
+
wanted_numel = sum(s.numel() for s in frozen_param_shapes.values())
|
| 367 |
+
avail_numel = sum([p.numel() for p in zero_model_states[0].frozen_param_fragments.values()]) * world_size
|
| 368 |
+
print(f'Frozen params: Have {avail_numel} numels to process.')
|
| 369 |
+
print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params')
|
| 370 |
+
|
| 371 |
+
total_params = 0
|
| 372 |
+
total_numel = 0
|
| 373 |
+
for name, shape in zero_model_states[0].frozen_param_shapes.items():
|
| 374 |
+
total_params += 1
|
| 375 |
+
unpartitioned_numel = shape.numel()
|
| 376 |
+
total_numel += unpartitioned_numel
|
| 377 |
+
|
| 378 |
+
param_frags = tuple(model_state.frozen_param_fragments[name] for model_state in zero_model_states)
|
| 379 |
+
state_dict[name] = torch.cat(param_frags, 0).narrow(0, 0, unpartitioned_numel).view(shape)
|
| 380 |
+
|
| 381 |
+
partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size)
|
| 382 |
+
|
| 383 |
+
if debug:
|
| 384 |
+
print(
|
| 385 |
+
f"Frozen params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}"
|
| 386 |
+
)
|
| 387 |
+
|
| 388 |
+
print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements")
|
| 389 |
+
|
| 390 |
+
|
| 391 |
+
class GatheredTensor:
|
| 392 |
+
"""
|
| 393 |
+
A pseudo tensor that collects partitioned weights.
|
| 394 |
+
It is more memory efficient when there are multiple groups.
|
| 395 |
+
"""
|
| 396 |
+
|
| 397 |
+
def __init__(self, flat_groups, flat_groups_offset, offset, partitioned_numel, shape):
|
| 398 |
+
self.flat_groups = flat_groups
|
| 399 |
+
self.flat_groups_offset = flat_groups_offset
|
| 400 |
+
self.offset = offset
|
| 401 |
+
self.partitioned_numel = partitioned_numel
|
| 402 |
+
self.shape = shape
|
| 403 |
+
self.dtype = self.flat_groups[0][0].dtype
|
| 404 |
+
|
| 405 |
+
def contiguous(self):
|
| 406 |
+
"""
|
| 407 |
+
Merge partitioned weights from flat_groups into a single tensor.
|
| 408 |
+
"""
|
| 409 |
+
end_idx = self.offset + self.partitioned_numel
|
| 410 |
+
world_size = len(self.flat_groups)
|
| 411 |
+
pad_flat_param_chunks = []
|
| 412 |
+
|
| 413 |
+
for rank_i in range(world_size):
|
| 414 |
+
# for each rank, we need to collect weights from related group/groups
|
| 415 |
+
flat_groups_at_rank_i = self.flat_groups[rank_i]
|
| 416 |
+
start_group_id = None
|
| 417 |
+
end_group_id = None
|
| 418 |
+
for group_id in range(len(self.flat_groups_offset)):
|
| 419 |
+
if self.flat_groups_offset[group_id] <= self.offset < self.flat_groups_offset[group_id + 1]:
|
| 420 |
+
start_group_id = group_id
|
| 421 |
+
if self.flat_groups_offset[group_id] < end_idx <= self.flat_groups_offset[group_id + 1]:
|
| 422 |
+
end_group_id = group_id
|
| 423 |
+
break
|
| 424 |
+
# collect weights from related group/groups
|
| 425 |
+
for group_id in range(start_group_id, end_group_id + 1):
|
| 426 |
+
flat_tensor = flat_groups_at_rank_i[group_id]
|
| 427 |
+
start_offset = self.offset - self.flat_groups_offset[group_id]
|
| 428 |
+
end_offset = min(end_idx, self.flat_groups_offset[group_id + 1]) - self.flat_groups_offset[group_id]
|
| 429 |
+
pad_flat_param_chunks.append(flat_tensor[start_offset:end_offset])
|
| 430 |
+
|
| 431 |
+
# collect weights from all ranks
|
| 432 |
+
pad_flat_param = torch.cat(pad_flat_param_chunks, dim=0)
|
| 433 |
+
param = pad_flat_param[:self.shape.numel()].view(self.shape).contiguous()
|
| 434 |
+
return param
|
| 435 |
+
|
| 436 |
+
|
| 437 |
+
def _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states):
|
| 438 |
+
param_shapes = zero_model_states[0].param_shapes
|
| 439 |
+
avail_numel = sum([flat_group.numel() for flat_group in fp32_flat_groups[0]]) * world_size
|
| 440 |
+
|
| 441 |
+
# Reconstruction protocol: For zero3 we need to zip the partitions together at boundary of each
|
| 442 |
+
# param, re-consolidating each param, while dealing with padding if any
|
| 443 |
+
|
| 444 |
+
# merge list of dicts, preserving order
|
| 445 |
+
param_shapes = {k: v for d in param_shapes for k, v in d.items()}
|
| 446 |
+
|
| 447 |
+
if debug:
|
| 448 |
+
for i in range(world_size):
|
| 449 |
+
print(f"{FP32_FLAT_GROUPS}[{i}].shape={fp32_flat_groups[i].shape}")
|
| 450 |
+
|
| 451 |
+
wanted_params = len(param_shapes)
|
| 452 |
+
wanted_numel = sum(shape.numel() for shape in param_shapes.values())
|
| 453 |
+
# not asserting if there is a mismatch due to possible padding
|
| 454 |
+
avail_numel = fp32_flat_groups[0].numel() * world_size
|
| 455 |
+
print(f"Trainable params: Have {avail_numel} numels to process.")
|
| 456 |
+
print(f"Trainable params: Need {wanted_numel} numels in {wanted_params} params.")
|
| 457 |
+
|
| 458 |
+
# params
|
| 459 |
+
# XXX: for huge models that can't fit into the host's RAM we will have to recode this to support
|
| 460 |
+
# out-of-core computing solution
|
| 461 |
+
offset = 0
|
| 462 |
+
total_numel = 0
|
| 463 |
+
total_params = 0
|
| 464 |
+
flat_groups_offset = [0] + list(np.cumsum([flat_tensor.numel() for flat_tensor in fp32_flat_groups[0]]))
|
| 465 |
+
for name, shape in tqdm(param_shapes.items(), desc='Gathering sharded weights'):
|
| 466 |
+
unpartitioned_numel = shape.numel()
|
| 467 |
+
total_numel += unpartitioned_numel
|
| 468 |
+
total_params += 1
|
| 469 |
+
partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size)
|
| 470 |
+
|
| 471 |
+
if debug:
|
| 472 |
+
print(
|
| 473 |
+
f"Trainable params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}"
|
| 474 |
+
)
|
| 475 |
+
|
| 476 |
+
# memory efficient tensor
|
| 477 |
+
tensor = GatheredTensor(fp32_flat_groups, flat_groups_offset, offset, partitioned_numel, shape)
|
| 478 |
+
state_dict[name] = tensor
|
| 479 |
+
offset += partitioned_numel
|
| 480 |
+
|
| 481 |
+
offset *= world_size
|
| 482 |
+
|
| 483 |
+
# Sanity check
|
| 484 |
+
if offset != avail_numel:
|
| 485 |
+
raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong")
|
| 486 |
+
|
| 487 |
+
print(f"Reconstructed Trainable fp32 state dict with {total_params} params {total_numel} elements")
|
| 488 |
+
|
| 489 |
+
|
| 490 |
+
def _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states,
|
| 491 |
+
exclude_frozen_parameters):
|
| 492 |
+
state_dict = OrderedDict()
|
| 493 |
+
|
| 494 |
+
# buffers
|
| 495 |
+
buffers = zero_model_states[0].buffers
|
| 496 |
+
state_dict.update(buffers)
|
| 497 |
+
if debug:
|
| 498 |
+
print(f"added {len(buffers)} buffers")
|
| 499 |
+
|
| 500 |
+
if not exclude_frozen_parameters:
|
| 501 |
+
_zero3_merge_frozen_params(state_dict, world_size, zero_model_states)
|
| 502 |
+
|
| 503 |
+
_zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states)
|
| 504 |
+
|
| 505 |
+
# recover shared parameters
|
| 506 |
+
for pair in zero_model_states[0].shared_params:
|
| 507 |
+
if pair[1] in state_dict:
|
| 508 |
+
state_dict[pair[0]] = state_dict[pair[1]]
|
| 509 |
+
|
| 510 |
+
return state_dict
|
| 511 |
+
|
| 512 |
+
|
| 513 |
+
def to_torch_tensor(state_dict, return_empty_tensor=False):
|
| 514 |
+
"""
|
| 515 |
+
Convert state_dict of GatheredTensor to torch tensor
|
| 516 |
+
"""
|
| 517 |
+
torch_state_dict = {}
|
| 518 |
+
converted_tensors = {}
|
| 519 |
+
for name, tensor in state_dict.items():
|
| 520 |
+
tensor_id = id(tensor)
|
| 521 |
+
if tensor_id in converted_tensors: # shared tensors
|
| 522 |
+
shared_tensor = torch_state_dict[converted_tensors[tensor_id]]
|
| 523 |
+
torch_state_dict[name] = shared_tensor
|
| 524 |
+
else:
|
| 525 |
+
converted_tensors[tensor_id] = name
|
| 526 |
+
if return_empty_tensor:
|
| 527 |
+
torch_state_dict[name] = torch.empty(tensor.shape, dtype=tensor.dtype)
|
| 528 |
+
else:
|
| 529 |
+
torch_state_dict[name] = tensor.contiguous()
|
| 530 |
+
return torch_state_dict
|
| 531 |
+
|
| 532 |
+
|
| 533 |
+
def get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir,
|
| 534 |
+
tag=None,
|
| 535 |
+
exclude_frozen_parameters=False,
|
| 536 |
+
lazy_mode=False):
|
| 537 |
+
"""
|
| 538 |
+
Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated state_dict that can be loaded with
|
| 539 |
+
``load_state_dict()`` and used for training without DeepSpeed or shared with others, for example
|
| 540 |
+
via a model hub.
|
| 541 |
+
|
| 542 |
+
Args:
|
| 543 |
+
- ``checkpoint_dir``: path to the desired checkpoint folder
|
| 544 |
+
- ``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``
|
| 545 |
+
- ``exclude_frozen_parameters``: exclude frozen parameters
|
| 546 |
+
- ``lazy_mode``: get state_dict in lazy mode. It returns a dict of pesduo tensor instead of torch tensor, which is more memory efficient.
|
| 547 |
+
Convert the pesduo tensor to torch tensor by ``.contiguous()``
|
| 548 |
+
|
| 549 |
+
Returns:
|
| 550 |
+
- pytorch ``state_dict``
|
| 551 |
+
|
| 552 |
+
A typical usage might be ::
|
| 553 |
+
|
| 554 |
+
from deepspeed.utils.zero_to_fp32 import get_fp32_state_dict_from_zero_checkpoint
|
| 555 |
+
# do the training and checkpoint saving
|
| 556 |
+
state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir) # already on cpu
|
| 557 |
+
model = model.cpu() # move to cpu
|
| 558 |
+
model.load_state_dict(state_dict)
|
| 559 |
+
# submit to model hub or save the model to share with others
|
| 560 |
+
|
| 561 |
+
In this example the ``model`` will no longer be usable in the deepspeed context of the same
|
| 562 |
+
application. i.e. you will need to re-initialize the deepspeed engine, since
|
| 563 |
+
``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it.
|
| 564 |
+
|
| 565 |
+
If you want it all done for you, use ``load_state_dict_from_zero_checkpoint`` instead.
|
| 566 |
+
|
| 567 |
+
Note: the above usage may not work if your application doesn't have sufficient free CPU memory.
|
| 568 |
+
You may need to use the offline approach using the ``zero_to_fp32.py`` script that is saved with
|
| 569 |
+
the checkpoint. Or you can load state_dict in lazy mode ::
|
| 570 |
+
|
| 571 |
+
from deepspeed.utils.zero_to_fp32 import get_fp32_state_dict_from_zero_checkpoint
|
| 572 |
+
state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, lazy_mode=True) # not on cpu
|
| 573 |
+
for name, lazy_tensor in state_dict.item():
|
| 574 |
+
tensor = lazy_tensor.contiguous() # to cpu
|
| 575 |
+
print(name, tensor)
|
| 576 |
+
# del tensor to release memory if it no longer in use
|
| 577 |
+
"""
|
| 578 |
+
if tag is None:
|
| 579 |
+
latest_path = os.path.join(checkpoint_dir, 'latest')
|
| 580 |
+
if os.path.isfile(latest_path):
|
| 581 |
+
with open(latest_path, 'r') as fd:
|
| 582 |
+
tag = fd.read().strip()
|
| 583 |
+
else:
|
| 584 |
+
raise ValueError(f"Unable to find 'latest' file at {latest_path}")
|
| 585 |
+
|
| 586 |
+
ds_checkpoint_dir = os.path.join(checkpoint_dir, tag)
|
| 587 |
+
|
| 588 |
+
if not os.path.isdir(ds_checkpoint_dir):
|
| 589 |
+
raise FileNotFoundError(f"Directory '{ds_checkpoint_dir}' doesn't exist")
|
| 590 |
+
|
| 591 |
+
state_dict = _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters)
|
| 592 |
+
if lazy_mode:
|
| 593 |
+
return state_dict
|
| 594 |
+
else:
|
| 595 |
+
return to_torch_tensor(state_dict)
|
| 596 |
+
|
| 597 |
+
|
| 598 |
+
def convert_zero_checkpoint_to_fp32_state_dict(checkpoint_dir,
|
| 599 |
+
output_dir,
|
| 600 |
+
max_shard_size="5GB",
|
| 601 |
+
safe_serialization=False,
|
| 602 |
+
tag=None,
|
| 603 |
+
exclude_frozen_parameters=False):
|
| 604 |
+
"""
|
| 605 |
+
Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict`` file that can be
|
| 606 |
+
loaded with ``torch.load(file)`` + ``load_state_dict()`` and used for training without DeepSpeed.
|
| 607 |
+
|
| 608 |
+
Args:
|
| 609 |
+
- ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``)
|
| 610 |
+
- ``output_dir``: directory to the pytorch fp32 state_dict output files
|
| 611 |
+
- ``max_shard_size``: the maximum size for a checkpoint before being sharded, default value is 5GB
|
| 612 |
+
- ``safe_serialization``: whether to save the model using `safetensors` or the traditional PyTorch way (that uses `pickle`).
|
| 613 |
+
- ``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``
|
| 614 |
+
- ``exclude_frozen_parameters``: exclude frozen parameters
|
| 615 |
+
"""
|
| 616 |
+
|
| 617 |
+
# Dependency pre-check
|
| 618 |
+
if safe_serialization:
|
| 619 |
+
try:
|
| 620 |
+
from safetensors.torch import save_file
|
| 621 |
+
except ImportError:
|
| 622 |
+
print('If you want to use `safe_serialization`, please `pip install safetensors`')
|
| 623 |
+
raise
|
| 624 |
+
if max_shard_size is not None:
|
| 625 |
+
try:
|
| 626 |
+
from huggingface_hub import split_torch_state_dict_into_shards
|
| 627 |
+
except ImportError:
|
| 628 |
+
print('If you want to use `max_shard_size`, please `pip install huggingface_hub`')
|
| 629 |
+
raise
|
| 630 |
+
|
| 631 |
+
# Convert zero checkpoint to state_dict
|
| 632 |
+
state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir,
|
| 633 |
+
tag,
|
| 634 |
+
exclude_frozen_parameters,
|
| 635 |
+
lazy_mode=True)
|
| 636 |
+
|
| 637 |
+
# Shard the model if it is too big.
|
| 638 |
+
weights_name = "model.safetensors" if safe_serialization else "pytorch_model.bin"
|
| 639 |
+
if max_shard_size is not None:
|
| 640 |
+
filename_pattern = weights_name.replace(".bin", "{suffix}.bin").replace(".safetensors", "{suffix}.safetensors")
|
| 641 |
+
# an memory-efficient approach for sharding
|
| 642 |
+
empty_state_dict = to_torch_tensor(state_dict, return_empty_tensor=True)
|
| 643 |
+
state_dict_split = split_torch_state_dict_into_shards(empty_state_dict,
|
| 644 |
+
filename_pattern=filename_pattern,
|
| 645 |
+
max_shard_size=max_shard_size)
|
| 646 |
+
else:
|
| 647 |
+
from collections import namedtuple
|
| 648 |
+
StateDictSplit = namedtuple("StateDictSplit", ["is_sharded", "filename_to_tensors"])
|
| 649 |
+
state_dict_split = StateDictSplit(is_sharded=False,
|
| 650 |
+
filename_to_tensors={weights_name: list(state_dict.keys())})
|
| 651 |
+
|
| 652 |
+
# Save the model by shard
|
| 653 |
+
os.makedirs(output_dir, exist_ok=True)
|
| 654 |
+
filename_to_tensors = state_dict_split.filename_to_tensors.items()
|
| 655 |
+
for shard_file, tensors in tqdm(filename_to_tensors, desc="Saving checkpoint shards"):
|
| 656 |
+
shard_state_dict = {tensor_name: state_dict[tensor_name] for tensor_name in tensors}
|
| 657 |
+
shard_state_dict = to_torch_tensor(shard_state_dict)
|
| 658 |
+
output_path = os.path.join(output_dir, shard_file)
|
| 659 |
+
if safe_serialization:
|
| 660 |
+
save_file(shard_state_dict, output_path, metadata={"format": "pt"})
|
| 661 |
+
else:
|
| 662 |
+
torch.save(shard_state_dict, output_path)
|
| 663 |
+
# release the memory of current shard
|
| 664 |
+
for tensor_name in list(shard_state_dict.keys()):
|
| 665 |
+
del state_dict[tensor_name]
|
| 666 |
+
del shard_state_dict[tensor_name]
|
| 667 |
+
del shard_state_dict
|
| 668 |
+
gc.collect()
|
| 669 |
+
|
| 670 |
+
# Save index if sharded
|
| 671 |
+
if state_dict_split.is_sharded:
|
| 672 |
+
index = {
|
| 673 |
+
"metadata": state_dict_split.metadata,
|
| 674 |
+
"weight_map": state_dict_split.tensor_to_filename,
|
| 675 |
+
}
|
| 676 |
+
save_index_file = "model.safetensors.index.json" if safe_serialization else "pytorch_model.bin.index.json"
|
| 677 |
+
save_index_file = os.path.join(output_dir, save_index_file)
|
| 678 |
+
with open(save_index_file, "w", encoding="utf-8") as f:
|
| 679 |
+
content = json.dumps(index, indent=2, sort_keys=True) + "\n"
|
| 680 |
+
f.write(content)
|
| 681 |
+
|
| 682 |
+
|
| 683 |
+
def load_state_dict_from_zero_checkpoint(model, checkpoint_dir, tag=None):
|
| 684 |
+
"""
|
| 685 |
+
1. Put the provided model to cpu
|
| 686 |
+
2. Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict``
|
| 687 |
+
3. Load it into the provided model
|
| 688 |
+
|
| 689 |
+
Args:
|
| 690 |
+
- ``model``: the model object to update
|
| 691 |
+
- ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``)
|
| 692 |
+
- ``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``
|
| 693 |
+
|
| 694 |
+
Returns:
|
| 695 |
+
- ``model`: modified model
|
| 696 |
+
|
| 697 |
+
Make sure you have plenty of CPU memory available before you call this function. If you don't
|
| 698 |
+
have enough use the ``zero_to_fp32.py`` utility to do the conversion. You will find it
|
| 699 |
+
conveniently placed for you in the checkpoint folder.
|
| 700 |
+
|
| 701 |
+
A typical usage might be ::
|
| 702 |
+
|
| 703 |
+
from deepspeed.utils.zero_to_fp32 import load_state_dict_from_zero_checkpoint
|
| 704 |
+
model = load_state_dict_from_zero_checkpoint(trainer.model, checkpoint_dir)
|
| 705 |
+
# submit to model hub or save the model to share with others
|
| 706 |
+
|
| 707 |
+
Note, that once this was run, the ``model`` will no longer be usable in the deepspeed context
|
| 708 |
+
of the same application. i.e. you will need to re-initialize the deepspeed engine, since
|
| 709 |
+
``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it.
|
| 710 |
+
|
| 711 |
+
"""
|
| 712 |
+
logger.info(f"Extracting fp32 weights")
|
| 713 |
+
state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag)
|
| 714 |
+
|
| 715 |
+
logger.info(f"Overwriting model with fp32 weights")
|
| 716 |
+
model = model.cpu()
|
| 717 |
+
model.load_state_dict(state_dict, strict=False)
|
| 718 |
+
|
| 719 |
+
return model
|
| 720 |
+
|
| 721 |
+
|
| 722 |
+
if __name__ == "__main__":
|
| 723 |
+
parser = argparse.ArgumentParser()
|
| 724 |
+
parser.add_argument("checkpoint_dir",
|
| 725 |
+
type=str,
|
| 726 |
+
help="path to the desired checkpoint folder, e.g., path/checkpoint-12")
|
| 727 |
+
parser.add_argument("output_dir",
|
| 728 |
+
type=str,
|
| 729 |
+
help="directory to the pytorch fp32 state_dict output files"
|
| 730 |
+
"(e.g. path/checkpoint-12-output/)")
|
| 731 |
+
parser.add_argument(
|
| 732 |
+
"--max_shard_size",
|
| 733 |
+
type=str,
|
| 734 |
+
default="5GB",
|
| 735 |
+
help="The maximum size for a checkpoint before being sharded. Checkpoints shard will then be each of size"
|
| 736 |
+
"lower than this size. If expressed as a string, needs to be digits followed by a unit (like `5MB`"
|
| 737 |
+
"We default it to 5GB in order for models to be able to run easily on free-tier google colab instances"
|
| 738 |
+
"without CPU OOM issues.")
|
| 739 |
+
parser.add_argument(
|
| 740 |
+
"--safe_serialization",
|
| 741 |
+
default=False,
|
| 742 |
+
action='store_true',
|
| 743 |
+
help="Whether to save the model using `safetensors` or the traditional PyTorch way (that uses `pickle`).")
|
| 744 |
+
parser.add_argument("-t",
|
| 745 |
+
"--tag",
|
| 746 |
+
type=str,
|
| 747 |
+
default=None,
|
| 748 |
+
help="checkpoint tag used as a unique identifier for checkpoint. e.g., global_step1")
|
| 749 |
+
parser.add_argument("--exclude_frozen_parameters", action='store_true', help="exclude frozen parameters")
|
| 750 |
+
parser.add_argument("-d", "--debug", action='store_true', help="enable debug")
|
| 751 |
+
args = parser.parse_args()
|
| 752 |
+
|
| 753 |
+
debug = args.debug
|
| 754 |
+
|
| 755 |
+
convert_zero_checkpoint_to_fp32_state_dict(args.checkpoint_dir,
|
| 756 |
+
args.output_dir,
|
| 757 |
+
max_shard_size=args.max_shard_size,
|
| 758 |
+
safe_serialization=args.safe_serialization,
|
| 759 |
+
tag=args.tag,
|
| 760 |
+
exclude_frozen_parameters=args.exclude_frozen_parameters)
|