Model save
Browse files- README.md +13 -12
- all_results.json +11 -11
- config.json +3 -3
- eval_results.json +6 -6
- model-00001-of-00003.safetensors +1 -1
- model-00002-of-00003.safetensors +1 -1
- model-00003-of-00003.safetensors +1 -1
- tokenizer_config.json +1 -1
- train_results.json +6 -6
- trainer_state.json +404 -196
- training_args.bin +1 -1
README.md
CHANGED
|
@@ -1,6 +1,6 @@
|
|
| 1 |
---
|
| 2 |
license: apache-2.0
|
| 3 |
-
base_model:
|
| 4 |
tags:
|
| 5 |
- generated_from_trainer
|
| 6 |
model-index:
|
|
@@ -13,9 +13,9 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 13 |
|
| 14 |
# mistral_channel_targeter
|
| 15 |
|
| 16 |
-
This model is a fine-tuned version of [
|
| 17 |
It achieves the following results on the evaluation set:
|
| 18 |
-
- Loss: 0.
|
| 19 |
|
| 20 |
## Model description
|
| 21 |
|
|
@@ -35,27 +35,28 @@ More information needed
|
|
| 35 |
|
| 36 |
The following hyperparameters were used during training:
|
| 37 |
- learning_rate: 5e-07
|
| 38 |
-
- train_batch_size:
|
| 39 |
-
- eval_batch_size:
|
| 40 |
- seed: 42
|
| 41 |
- distributed_type: multi-GPU
|
| 42 |
- num_devices: 3
|
| 43 |
-
- gradient_accumulation_steps: 2
|
| 44 |
- total_train_batch_size: 24
|
| 45 |
-
- total_eval_batch_size:
|
| 46 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 47 |
- lr_scheduler_type: cosine
|
| 48 |
- lr_scheduler_warmup_ratio: 0.1
|
| 49 |
-
- num_epochs:
|
| 50 |
|
| 51 |
### Training results
|
| 52 |
|
| 53 |
| Training Loss | Epoch | Step | Validation Loss |
|
| 54 |
|:-------------:|:-----:|:----:|:---------------:|
|
| 55 |
-
| 0.
|
| 56 |
-
| 0.
|
| 57 |
-
| 0.
|
| 58 |
-
| 0.
|
|
|
|
|
|
|
| 59 |
|
| 60 |
|
| 61 |
### Framework versions
|
|
|
|
| 1 |
---
|
| 2 |
license: apache-2.0
|
| 3 |
+
base_model: BioMistral/BioMistral-7B
|
| 4 |
tags:
|
| 5 |
- generated_from_trainer
|
| 6 |
model-index:
|
|
|
|
| 13 |
|
| 14 |
# mistral_channel_targeter
|
| 15 |
|
| 16 |
+
This model is a fine-tuned version of [BioMistral/BioMistral-7B](https://huggingface.co/BioMistral/BioMistral-7B) on an unknown dataset.
|
| 17 |
It achieves the following results on the evaluation set:
|
| 18 |
+
- Loss: 0.0708
|
| 19 |
|
| 20 |
## Model description
|
| 21 |
|
|
|
|
| 35 |
|
| 36 |
The following hyperparameters were used during training:
|
| 37 |
- learning_rate: 5e-07
|
| 38 |
+
- train_batch_size: 8
|
| 39 |
+
- eval_batch_size: 8
|
| 40 |
- seed: 42
|
| 41 |
- distributed_type: multi-GPU
|
| 42 |
- num_devices: 3
|
|
|
|
| 43 |
- total_train_batch_size: 24
|
| 44 |
+
- total_eval_batch_size: 24
|
| 45 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 46 |
- lr_scheduler_type: cosine
|
| 47 |
- lr_scheduler_warmup_ratio: 0.1
|
| 48 |
+
- num_epochs: 6
|
| 49 |
|
| 50 |
### Training results
|
| 51 |
|
| 52 |
| Training Loss | Epoch | Step | Validation Loss |
|
| 53 |
|:-------------:|:-----:|:----:|:---------------:|
|
| 54 |
+
| 0.0526 | 1.0 | 159 | 0.0547 |
|
| 55 |
+
| 0.0421 | 2.0 | 318 | 0.0447 |
|
| 56 |
+
| 0.0285 | 3.0 | 477 | 0.0385 |
|
| 57 |
+
| 0.0165 | 4.0 | 636 | 0.0465 |
|
| 58 |
+
| 0.0021 | 5.0 | 795 | 0.0659 |
|
| 59 |
+
| 0.0008 | 6.0 | 954 | 0.0708 |
|
| 60 |
|
| 61 |
|
| 62 |
### Framework versions
|
all_results.json
CHANGED
|
@@ -1,13 +1,13 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_loss": 0.
|
| 4 |
-
"eval_runtime":
|
| 5 |
-
"eval_samples":
|
| 6 |
-
"eval_samples_per_second": 9.
|
| 7 |
-
"eval_steps_per_second": 0.
|
| 8 |
-
"train_loss": 0.
|
| 9 |
-
"train_runtime":
|
| 10 |
-
"train_samples":
|
| 11 |
-
"train_samples_per_second": 2.
|
| 12 |
-
"train_steps_per_second": 0.
|
| 13 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 6.0,
|
| 3 |
+
"eval_loss": 0.07076797634363174,
|
| 4 |
+
"eval_runtime": 22.043,
|
| 5 |
+
"eval_samples": 206,
|
| 6 |
+
"eval_samples_per_second": 9.345,
|
| 7 |
+
"eval_steps_per_second": 0.408,
|
| 8 |
+
"train_loss": 0.18272512886366676,
|
| 9 |
+
"train_runtime": 9319.8502,
|
| 10 |
+
"train_samples": 3795,
|
| 11 |
+
"train_samples_per_second": 2.443,
|
| 12 |
+
"train_steps_per_second": 0.102
|
| 13 |
}
|
config.json
CHANGED
|
@@ -1,5 +1,5 @@
|
|
| 1 |
{
|
| 2 |
-
"_name_or_path": "
|
| 3 |
"architectures": [
|
| 4 |
"MistralForCausalLM"
|
| 5 |
],
|
|
@@ -16,8 +16,8 @@
|
|
| 16 |
"num_hidden_layers": 32,
|
| 17 |
"num_key_value_heads": 8,
|
| 18 |
"rms_norm_eps": 1e-05,
|
| 19 |
-
"rope_theta":
|
| 20 |
-
"sliding_window":
|
| 21 |
"tie_word_embeddings": false,
|
| 22 |
"torch_dtype": "bfloat16",
|
| 23 |
"transformers_version": "4.36.0",
|
|
|
|
| 1 |
{
|
| 2 |
+
"_name_or_path": "BioMistral/BioMistral-7B",
|
| 3 |
"architectures": [
|
| 4 |
"MistralForCausalLM"
|
| 5 |
],
|
|
|
|
| 16 |
"num_hidden_layers": 32,
|
| 17 |
"num_key_value_heads": 8,
|
| 18 |
"rms_norm_eps": 1e-05,
|
| 19 |
+
"rope_theta": 10000.0,
|
| 20 |
+
"sliding_window": 4096,
|
| 21 |
"tie_word_embeddings": false,
|
| 22 |
"torch_dtype": "bfloat16",
|
| 23 |
"transformers_version": "4.36.0",
|
eval_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_loss": 0.
|
| 4 |
-
"eval_runtime":
|
| 5 |
-
"eval_samples":
|
| 6 |
-
"eval_samples_per_second": 9.
|
| 7 |
-
"eval_steps_per_second": 0.
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 6.0,
|
| 3 |
+
"eval_loss": 0.07076797634363174,
|
| 4 |
+
"eval_runtime": 22.043,
|
| 5 |
+
"eval_samples": 206,
|
| 6 |
+
"eval_samples_per_second": 9.345,
|
| 7 |
+
"eval_steps_per_second": 0.408
|
| 8 |
}
|
model-00001-of-00003.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 4943162336
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:92afcd4e1fb769e701a48bf65d7750b51820f9f5a06061b2c6899dcc4a6f685f
|
| 3 |
size 4943162336
|
model-00002-of-00003.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 4999819336
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:41cf70dcb5493928483654d17938b754ba42fc66b87167bce580acd30dddb91a
|
| 3 |
size 4999819336
|
model-00003-of-00003.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 4540516344
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:70519e3fb132eb05253592faf31ad4116681d32d6b102810c5cb276ac15ce06c
|
| 3 |
size 4540516344
|
tokenizer_config.json
CHANGED
|
@@ -29,7 +29,7 @@
|
|
| 29 |
},
|
| 30 |
"additional_special_tokens": [],
|
| 31 |
"bos_token": "<s>",
|
| 32 |
-
"chat_template": "{% for message in messages %}\n{% if message['role'] == 'user' %}\n{{ message['content'] }}\n{% elif message['role'] == 'system' %}\n{{ message['content'] }}\n{% elif message['role'] == 'assistant' %}\n{{ '\n###Response :\n' + message['content'] }}\n{% endif %}\n{% if loop.last and add_generation_prompt %}\n{{ '###Response :\n' }}\n{% endif %}\n{% endfor %}",
|
| 33 |
"clean_up_tokenization_spaces": false,
|
| 34 |
"eos_token": "</s>",
|
| 35 |
"legacy": true,
|
|
|
|
| 29 |
},
|
| 30 |
"additional_special_tokens": [],
|
| 31 |
"bos_token": "<s>",
|
| 32 |
+
"chat_template": "{% for message in messages %}\n{% if message['role'] == 'user' %}\n{{ '### Input:\n\n' + message['content'] }}\n{% elif message['role'] == 'system' %}\n{{ '### Instruction:\n\n' + message['content'] }}\n{% elif message['role'] == 'assistant' %}\n{{ '\n###Response :\n\n' + message['content'] }}\n{% endif %}\n{% if loop.last and add_generation_prompt %}\n{{ '###Response :\n\n' }}\n{% endif %}\n{% endfor %}",
|
| 33 |
"clean_up_tokenization_spaces": false,
|
| 34 |
"eos_token": "</s>",
|
| 35 |
"legacy": true,
|
train_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"train_loss": 0.
|
| 4 |
-
"train_runtime":
|
| 5 |
-
"train_samples":
|
| 6 |
-
"train_samples_per_second": 2.
|
| 7 |
-
"train_steps_per_second": 0.
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 6.0,
|
| 3 |
+
"train_loss": 0.18272512886366676,
|
| 4 |
+
"train_runtime": 9319.8502,
|
| 5 |
+
"train_samples": 3795,
|
| 6 |
+
"train_samples_per_second": 2.443,
|
| 7 |
+
"train_steps_per_second": 0.102
|
| 8 |
}
|
trainer_state.json
CHANGED
|
@@ -1,440 +1,648 @@
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
-
"epoch":
|
| 5 |
"eval_steps": 500,
|
| 6 |
-
"global_step":
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
"epoch": 0.06,
|
| 13 |
-
"learning_rate":
|
| 14 |
-
"loss":
|
| 15 |
"step": 10
|
| 16 |
},
|
| 17 |
{
|
| 18 |
"epoch": 0.13,
|
| 19 |
-
"learning_rate": 1.
|
| 20 |
-
"loss":
|
| 21 |
"step": 20
|
| 22 |
},
|
| 23 |
{
|
| 24 |
"epoch": 0.19,
|
| 25 |
-
"learning_rate":
|
| 26 |
-
"loss":
|
| 27 |
"step": 30
|
| 28 |
},
|
| 29 |
{
|
| 30 |
"epoch": 0.25,
|
| 31 |
-
"learning_rate":
|
| 32 |
-
"loss":
|
| 33 |
"step": 40
|
| 34 |
},
|
| 35 |
{
|
| 36 |
-
"epoch": 0.
|
| 37 |
-
"learning_rate":
|
| 38 |
-
"loss": 0.
|
| 39 |
"step": 50
|
| 40 |
},
|
| 41 |
{
|
| 42 |
"epoch": 0.38,
|
| 43 |
-
"learning_rate":
|
| 44 |
-
"loss": 0.
|
| 45 |
"step": 60
|
| 46 |
},
|
| 47 |
{
|
| 48 |
"epoch": 0.44,
|
| 49 |
-
"learning_rate":
|
| 50 |
-
"loss": 0.
|
| 51 |
"step": 70
|
| 52 |
},
|
| 53 |
{
|
| 54 |
"epoch": 0.5,
|
| 55 |
-
"learning_rate": 4.
|
| 56 |
-
"loss": 0.
|
| 57 |
"step": 80
|
| 58 |
},
|
| 59 |
{
|
| 60 |
"epoch": 0.57,
|
| 61 |
-
"learning_rate": 4.
|
| 62 |
-
"loss": 0.
|
| 63 |
"step": 90
|
| 64 |
},
|
| 65 |
{
|
| 66 |
"epoch": 0.63,
|
| 67 |
-
"learning_rate": 4.
|
| 68 |
-
"loss": 0.
|
| 69 |
"step": 100
|
| 70 |
},
|
| 71 |
{
|
| 72 |
"epoch": 0.69,
|
| 73 |
-
"learning_rate": 4.
|
| 74 |
-
"loss": 0.
|
| 75 |
"step": 110
|
| 76 |
},
|
| 77 |
{
|
| 78 |
-
"epoch": 0.
|
| 79 |
-
"learning_rate": 4.
|
| 80 |
-
"loss": 0.
|
| 81 |
"step": 120
|
| 82 |
},
|
| 83 |
{
|
| 84 |
"epoch": 0.82,
|
| 85 |
-
"learning_rate": 4.
|
| 86 |
-
"loss": 0.
|
| 87 |
"step": 130
|
| 88 |
},
|
| 89 |
{
|
| 90 |
"epoch": 0.88,
|
| 91 |
-
"learning_rate": 4.
|
| 92 |
-
"loss": 0.
|
| 93 |
"step": 140
|
| 94 |
},
|
| 95 |
{
|
| 96 |
-
"epoch": 0.
|
| 97 |
-
"learning_rate": 4.
|
| 98 |
-
"loss": 0.
|
| 99 |
"step": 150
|
| 100 |
},
|
| 101 |
{
|
| 102 |
"epoch": 1.0,
|
| 103 |
-
"eval_loss": 0.
|
| 104 |
-
"eval_runtime": 21.
|
| 105 |
-
"eval_samples_per_second": 9.
|
| 106 |
-
"eval_steps_per_second": 0.
|
| 107 |
-
"step":
|
| 108 |
},
|
| 109 |
{
|
| 110 |
"epoch": 1.01,
|
| 111 |
-
"learning_rate": 4.
|
| 112 |
-
"loss": 0.
|
| 113 |
"step": 160
|
| 114 |
},
|
| 115 |
{
|
| 116 |
"epoch": 1.07,
|
| 117 |
-
"learning_rate": 4.
|
| 118 |
-
"loss": 0.
|
| 119 |
"step": 170
|
| 120 |
},
|
| 121 |
{
|
| 122 |
-
"epoch": 1.
|
| 123 |
-
"learning_rate": 4.
|
| 124 |
-
"loss": 0.
|
| 125 |
"step": 180
|
| 126 |
},
|
| 127 |
{
|
| 128 |
-
"epoch": 1.
|
| 129 |
-
"learning_rate": 4.
|
| 130 |
-
"loss": 0.
|
| 131 |
"step": 190
|
| 132 |
},
|
| 133 |
{
|
| 134 |
"epoch": 1.26,
|
| 135 |
-
"learning_rate": 4.
|
| 136 |
-
"loss": 0.
|
| 137 |
"step": 200
|
| 138 |
},
|
| 139 |
{
|
| 140 |
"epoch": 1.32,
|
| 141 |
-
"learning_rate": 4.
|
| 142 |
-
"loss": 0.
|
| 143 |
"step": 210
|
| 144 |
},
|
| 145 |
{
|
| 146 |
-
"epoch": 1.
|
| 147 |
-
"learning_rate": 4.
|
| 148 |
-
"loss": 0.
|
| 149 |
"step": 220
|
| 150 |
},
|
| 151 |
{
|
| 152 |
"epoch": 1.45,
|
| 153 |
-
"learning_rate": 4.
|
| 154 |
-
"loss": 0.
|
| 155 |
"step": 230
|
| 156 |
},
|
| 157 |
{
|
| 158 |
"epoch": 1.51,
|
| 159 |
-
"learning_rate":
|
| 160 |
-
"loss": 0.
|
| 161 |
"step": 240
|
| 162 |
},
|
| 163 |
{
|
| 164 |
-
"epoch": 1.
|
| 165 |
-
"learning_rate":
|
| 166 |
-
"loss": 0.
|
| 167 |
"step": 250
|
| 168 |
},
|
| 169 |
{
|
| 170 |
"epoch": 1.64,
|
| 171 |
-
"learning_rate":
|
| 172 |
-
"loss": 0.
|
| 173 |
"step": 260
|
| 174 |
},
|
| 175 |
{
|
| 176 |
"epoch": 1.7,
|
| 177 |
-
"learning_rate":
|
| 178 |
-
"loss": 0.
|
| 179 |
"step": 270
|
| 180 |
},
|
| 181 |
{
|
| 182 |
-
"epoch": 1.
|
| 183 |
-
"learning_rate":
|
| 184 |
-
"loss": 0.
|
| 185 |
"step": 280
|
| 186 |
},
|
| 187 |
{
|
| 188 |
-
"epoch": 1.
|
| 189 |
-
"learning_rate":
|
| 190 |
-
"loss": 0.
|
| 191 |
"step": 290
|
| 192 |
},
|
| 193 |
{
|
| 194 |
"epoch": 1.89,
|
| 195 |
-
"learning_rate":
|
| 196 |
-
"loss": 0.
|
| 197 |
"step": 300
|
| 198 |
},
|
| 199 |
{
|
| 200 |
-
"epoch": 1.
|
| 201 |
-
"learning_rate":
|
| 202 |
-
"loss": 0.
|
| 203 |
"step": 310
|
| 204 |
},
|
| 205 |
{
|
| 206 |
"epoch": 2.0,
|
| 207 |
-
"eval_loss": 0.
|
| 208 |
-
"eval_runtime":
|
| 209 |
-
"eval_samples_per_second": 9.
|
| 210 |
-
"eval_steps_per_second": 0.
|
| 211 |
-
"step":
|
| 212 |
},
|
| 213 |
{
|
| 214 |
-
"epoch": 2.
|
| 215 |
-
"learning_rate":
|
| 216 |
-
"loss": 0.
|
| 217 |
"step": 320
|
| 218 |
},
|
| 219 |
{
|
| 220 |
"epoch": 2.08,
|
| 221 |
-
"learning_rate":
|
| 222 |
-
"loss": 0.
|
| 223 |
"step": 330
|
| 224 |
},
|
| 225 |
{
|
| 226 |
-
"epoch": 2.
|
| 227 |
-
"learning_rate":
|
| 228 |
-
"loss": 0.
|
| 229 |
"step": 340
|
| 230 |
},
|
| 231 |
{
|
| 232 |
-
"epoch": 2.
|
| 233 |
-
"learning_rate":
|
| 234 |
"loss": 0.0304,
|
| 235 |
"step": 350
|
| 236 |
},
|
| 237 |
{
|
| 238 |
-
"epoch": 2.
|
| 239 |
-
"learning_rate":
|
| 240 |
-
"loss": 0.
|
| 241 |
"step": 360
|
| 242 |
},
|
| 243 |
{
|
| 244 |
"epoch": 2.33,
|
| 245 |
-
"learning_rate":
|
| 246 |
-
"loss": 0.
|
| 247 |
"step": 370
|
| 248 |
},
|
| 249 |
{
|
| 250 |
-
"epoch": 2.
|
| 251 |
-
"learning_rate":
|
| 252 |
-
"loss": 0.
|
| 253 |
"step": 380
|
| 254 |
},
|
| 255 |
{
|
| 256 |
-
"epoch": 2.
|
| 257 |
-
"learning_rate":
|
| 258 |
"loss": 0.0295,
|
| 259 |
"step": 390
|
| 260 |
},
|
| 261 |
{
|
| 262 |
"epoch": 2.52,
|
| 263 |
-
"learning_rate":
|
| 264 |
-
"loss": 0.
|
| 265 |
"step": 400
|
| 266 |
},
|
| 267 |
{
|
| 268 |
-
"epoch": 2.
|
| 269 |
-
"learning_rate":
|
| 270 |
-
"loss": 0.
|
| 271 |
"step": 410
|
| 272 |
},
|
| 273 |
{
|
| 274 |
-
"epoch": 2.
|
| 275 |
-
"learning_rate":
|
| 276 |
-
"loss": 0.
|
| 277 |
"step": 420
|
| 278 |
},
|
| 279 |
{
|
| 280 |
-
"epoch": 2.
|
| 281 |
-
"learning_rate":
|
| 282 |
-
"loss": 0.
|
| 283 |
"step": 430
|
| 284 |
},
|
| 285 |
{
|
| 286 |
-
"epoch": 2.
|
| 287 |
-
"learning_rate":
|
| 288 |
"loss": 0.0314,
|
| 289 |
"step": 440
|
| 290 |
},
|
| 291 |
{
|
| 292 |
-
"epoch": 2.
|
| 293 |
-
"learning_rate":
|
| 294 |
-
"loss": 0.
|
| 295 |
"step": 450
|
| 296 |
},
|
| 297 |
{
|
| 298 |
-
"epoch": 2.
|
| 299 |
-
"learning_rate":
|
| 300 |
-
"loss": 0.
|
| 301 |
"step": 460
|
| 302 |
},
|
| 303 |
{
|
| 304 |
-
"epoch": 2.
|
| 305 |
-
"learning_rate":
|
| 306 |
-
"loss": 0.
|
| 307 |
"step": 470
|
| 308 |
},
|
| 309 |
{
|
| 310 |
"epoch": 3.0,
|
| 311 |
-
"eval_loss": 0.
|
| 312 |
-
"eval_runtime":
|
| 313 |
-
"eval_samples_per_second": 9.
|
| 314 |
-
"eval_steps_per_second": 0.
|
| 315 |
-
"step":
|
| 316 |
},
|
| 317 |
{
|
| 318 |
-
"epoch": 3.
|
| 319 |
-
"learning_rate":
|
| 320 |
-
"loss": 0.
|
| 321 |
"step": 480
|
| 322 |
},
|
| 323 |
{
|
| 324 |
-
"epoch": 3.
|
| 325 |
-
"learning_rate":
|
| 326 |
-
"loss": 0.
|
| 327 |
"step": 490
|
| 328 |
},
|
| 329 |
{
|
| 330 |
-
"epoch": 3.
|
| 331 |
-
"learning_rate":
|
| 332 |
-
"loss": 0.
|
| 333 |
"step": 500
|
| 334 |
},
|
| 335 |
{
|
| 336 |
-
"epoch": 3.
|
| 337 |
-
"learning_rate":
|
| 338 |
-
"loss": 0.
|
| 339 |
"step": 510
|
| 340 |
},
|
| 341 |
{
|
| 342 |
-
"epoch": 3.
|
| 343 |
-
"learning_rate":
|
| 344 |
-
"loss": 0.
|
| 345 |
"step": 520
|
| 346 |
},
|
| 347 |
{
|
| 348 |
-
"epoch": 3.
|
| 349 |
-
"learning_rate":
|
| 350 |
-
"loss": 0.
|
| 351 |
"step": 530
|
| 352 |
},
|
| 353 |
{
|
| 354 |
-
"epoch": 3.
|
| 355 |
-
"learning_rate":
|
| 356 |
-
"loss": 0.
|
| 357 |
"step": 540
|
| 358 |
},
|
| 359 |
{
|
| 360 |
-
"epoch": 3.
|
| 361 |
-
"learning_rate": 2.
|
| 362 |
-
"loss": 0.
|
| 363 |
"step": 550
|
| 364 |
},
|
| 365 |
{
|
| 366 |
-
"epoch": 3.
|
| 367 |
-
"learning_rate":
|
| 368 |
-
"loss": 0.
|
| 369 |
"step": 560
|
| 370 |
},
|
| 371 |
{
|
| 372 |
-
"epoch": 3.
|
| 373 |
-
"learning_rate":
|
| 374 |
-
"loss": 0.
|
| 375 |
"step": 570
|
| 376 |
},
|
| 377 |
{
|
| 378 |
-
"epoch": 3.
|
| 379 |
-
"learning_rate": 1.
|
| 380 |
-
"loss": 0.
|
| 381 |
"step": 580
|
| 382 |
},
|
| 383 |
{
|
| 384 |
-
"epoch": 3.
|
| 385 |
-
"learning_rate":
|
| 386 |
-
"loss": 0.
|
| 387 |
"step": 590
|
| 388 |
},
|
| 389 |
{
|
| 390 |
-
"epoch": 3.
|
| 391 |
-
"learning_rate":
|
| 392 |
-
"loss": 0.
|
| 393 |
"step": 600
|
| 394 |
},
|
| 395 |
{
|
| 396 |
-
"epoch": 3.
|
| 397 |
-
"learning_rate": 1.
|
| 398 |
-
"loss": 0.
|
| 399 |
"step": 610
|
| 400 |
},
|
| 401 |
{
|
| 402 |
-
"epoch": 3.
|
| 403 |
-
"learning_rate":
|
| 404 |
-
"loss": 0.
|
| 405 |
"step": 620
|
| 406 |
},
|
| 407 |
{
|
| 408 |
-
"epoch": 3.
|
| 409 |
-
"learning_rate": 1.
|
| 410 |
-
"loss": 0.
|
| 411 |
"step": 630
|
| 412 |
},
|
| 413 |
{
|
| 414 |
-
"epoch":
|
| 415 |
-
"eval_loss": 0.
|
| 416 |
-
"eval_runtime":
|
| 417 |
-
"eval_samples_per_second": 9.
|
| 418 |
-
"eval_steps_per_second": 0.
|
| 419 |
-
"step":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 420 |
},
|
| 421 |
{
|
| 422 |
-
"epoch":
|
| 423 |
-
"step":
|
| 424 |
-
"total_flos":
|
| 425 |
-
"train_loss": 0.
|
| 426 |
-
"train_runtime":
|
| 427 |
-
"train_samples_per_second": 2.
|
| 428 |
-
"train_steps_per_second": 0.
|
| 429 |
}
|
| 430 |
],
|
| 431 |
"logging_steps": 10,
|
| 432 |
-
"max_steps":
|
| 433 |
"num_input_tokens_seen": 0,
|
| 434 |
-
"num_train_epochs":
|
| 435 |
"save_steps": 500,
|
| 436 |
-
"total_flos":
|
| 437 |
-
"train_batch_size":
|
| 438 |
"trial_name": null,
|
| 439 |
"trial_params": null
|
| 440 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 6.0,
|
| 5 |
"eval_steps": 500,
|
| 6 |
+
"global_step": 954,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
"epoch": 0.06,
|
| 13 |
+
"learning_rate": 5.208333333333333e-08,
|
| 14 |
+
"loss": 5.1406,
|
| 15 |
"step": 10
|
| 16 |
},
|
| 17 |
{
|
| 18 |
"epoch": 0.13,
|
| 19 |
+
"learning_rate": 1.0416666666666667e-07,
|
| 20 |
+
"loss": 4.8324,
|
| 21 |
"step": 20
|
| 22 |
},
|
| 23 |
{
|
| 24 |
"epoch": 0.19,
|
| 25 |
+
"learning_rate": 1.5624999999999999e-07,
|
| 26 |
+
"loss": 3.282,
|
| 27 |
"step": 30
|
| 28 |
},
|
| 29 |
{
|
| 30 |
"epoch": 0.25,
|
| 31 |
+
"learning_rate": 2.0833333333333333e-07,
|
| 32 |
+
"loss": 1.166,
|
| 33 |
"step": 40
|
| 34 |
},
|
| 35 |
{
|
| 36 |
+
"epoch": 0.31,
|
| 37 |
+
"learning_rate": 2.604166666666667e-07,
|
| 38 |
+
"loss": 0.3744,
|
| 39 |
"step": 50
|
| 40 |
},
|
| 41 |
{
|
| 42 |
"epoch": 0.38,
|
| 43 |
+
"learning_rate": 3.1249999999999997e-07,
|
| 44 |
+
"loss": 0.2785,
|
| 45 |
"step": 60
|
| 46 |
},
|
| 47 |
{
|
| 48 |
"epoch": 0.44,
|
| 49 |
+
"learning_rate": 3.645833333333333e-07,
|
| 50 |
+
"loss": 0.2278,
|
| 51 |
"step": 70
|
| 52 |
},
|
| 53 |
{
|
| 54 |
"epoch": 0.5,
|
| 55 |
+
"learning_rate": 4.1666666666666667e-07,
|
| 56 |
+
"loss": 0.1459,
|
| 57 |
"step": 80
|
| 58 |
},
|
| 59 |
{
|
| 60 |
"epoch": 0.57,
|
| 61 |
+
"learning_rate": 4.6874999999999996e-07,
|
| 62 |
+
"loss": 0.0876,
|
| 63 |
"step": 90
|
| 64 |
},
|
| 65 |
{
|
| 66 |
"epoch": 0.63,
|
| 67 |
+
"learning_rate": 4.999731868769026e-07,
|
| 68 |
+
"loss": 0.071,
|
| 69 |
"step": 100
|
| 70 |
},
|
| 71 |
{
|
| 72 |
"epoch": 0.69,
|
| 73 |
+
"learning_rate": 4.996716052911017e-07,
|
| 74 |
+
"loss": 0.0626,
|
| 75 |
"step": 110
|
| 76 |
},
|
| 77 |
{
|
| 78 |
+
"epoch": 0.75,
|
| 79 |
+
"learning_rate": 4.990353313429303e-07,
|
| 80 |
+
"loss": 0.0586,
|
| 81 |
"step": 120
|
| 82 |
},
|
| 83 |
{
|
| 84 |
"epoch": 0.82,
|
| 85 |
+
"learning_rate": 4.980652179769217e-07,
|
| 86 |
+
"loss": 0.055,
|
| 87 |
"step": 130
|
| 88 |
},
|
| 89 |
{
|
| 90 |
"epoch": 0.88,
|
| 91 |
+
"learning_rate": 4.967625656594781e-07,
|
| 92 |
+
"loss": 0.0547,
|
| 93 |
"step": 140
|
| 94 |
},
|
| 95 |
{
|
| 96 |
+
"epoch": 0.94,
|
| 97 |
+
"learning_rate": 4.951291206355559e-07,
|
| 98 |
+
"loss": 0.0526,
|
| 99 |
"step": 150
|
| 100 |
},
|
| 101 |
{
|
| 102 |
"epoch": 1.0,
|
| 103 |
+
"eval_loss": 0.054711490869522095,
|
| 104 |
+
"eval_runtime": 21.9924,
|
| 105 |
+
"eval_samples_per_second": 9.367,
|
| 106 |
+
"eval_steps_per_second": 0.409,
|
| 107 |
+
"step": 159
|
| 108 |
},
|
| 109 |
{
|
| 110 |
"epoch": 1.01,
|
| 111 |
+
"learning_rate": 4.93167072587771e-07,
|
| 112 |
+
"loss": 0.0468,
|
| 113 |
"step": 160
|
| 114 |
},
|
| 115 |
{
|
| 116 |
"epoch": 1.07,
|
| 117 |
+
"learning_rate": 4.908790517010636e-07,
|
| 118 |
+
"loss": 0.0447,
|
| 119 |
"step": 170
|
| 120 |
},
|
| 121 |
{
|
| 122 |
+
"epoch": 1.13,
|
| 123 |
+
"learning_rate": 4.882681251368548e-07,
|
| 124 |
+
"loss": 0.0447,
|
| 125 |
"step": 180
|
| 126 |
},
|
| 127 |
{
|
| 128 |
+
"epoch": 1.19,
|
| 129 |
+
"learning_rate": 4.853377929214243e-07,
|
| 130 |
+
"loss": 0.0498,
|
| 131 |
"step": 190
|
| 132 |
},
|
| 133 |
{
|
| 134 |
"epoch": 1.26,
|
| 135 |
+
"learning_rate": 4.820919832540181e-07,
|
| 136 |
+
"loss": 0.0468,
|
| 137 |
"step": 200
|
| 138 |
},
|
| 139 |
{
|
| 140 |
"epoch": 1.32,
|
| 141 |
+
"learning_rate": 4.785350472409791e-07,
|
| 142 |
+
"loss": 0.0458,
|
| 143 |
"step": 210
|
| 144 |
},
|
| 145 |
{
|
| 146 |
+
"epoch": 1.38,
|
| 147 |
+
"learning_rate": 4.7467175306295647e-07,
|
| 148 |
+
"loss": 0.046,
|
| 149 |
"step": 220
|
| 150 |
},
|
| 151 |
{
|
| 152 |
"epoch": 1.45,
|
| 153 |
+
"learning_rate": 4.70507279583015e-07,
|
| 154 |
+
"loss": 0.043,
|
| 155 |
"step": 230
|
| 156 |
},
|
| 157 |
{
|
| 158 |
"epoch": 1.51,
|
| 159 |
+
"learning_rate": 4.6604720940421207e-07,
|
| 160 |
+
"loss": 0.0431,
|
| 161 |
"step": 240
|
| 162 |
},
|
| 163 |
{
|
| 164 |
+
"epoch": 1.57,
|
| 165 |
+
"learning_rate": 4.612975213859487e-07,
|
| 166 |
+
"loss": 0.0428,
|
| 167 |
"step": 250
|
| 168 |
},
|
| 169 |
{
|
| 170 |
"epoch": 1.64,
|
| 171 |
+
"learning_rate": 4.5626458262912735e-07,
|
| 172 |
+
"loss": 0.0468,
|
| 173 |
"step": 260
|
| 174 |
},
|
| 175 |
{
|
| 176 |
"epoch": 1.7,
|
| 177 |
+
"learning_rate": 4.5095513994085974e-07,
|
| 178 |
+
"loss": 0.0418,
|
| 179 |
"step": 270
|
| 180 |
},
|
| 181 |
{
|
| 182 |
+
"epoch": 1.76,
|
| 183 |
+
"learning_rate": 4.453763107901675e-07,
|
| 184 |
+
"loss": 0.0437,
|
| 185 |
"step": 280
|
| 186 |
},
|
| 187 |
{
|
| 188 |
+
"epoch": 1.82,
|
| 189 |
+
"learning_rate": 4.395355737667985e-07,
|
| 190 |
+
"loss": 0.0429,
|
| 191 |
"step": 290
|
| 192 |
},
|
| 193 |
{
|
| 194 |
"epoch": 1.89,
|
| 195 |
+
"learning_rate": 4.3344075855595097e-07,
|
| 196 |
+
"loss": 0.0408,
|
| 197 |
"step": 300
|
| 198 |
},
|
| 199 |
{
|
| 200 |
+
"epoch": 1.95,
|
| 201 |
+
"learning_rate": 4.271000354423425e-07,
|
| 202 |
+
"loss": 0.0421,
|
| 203 |
"step": 310
|
| 204 |
},
|
| 205 |
{
|
| 206 |
"epoch": 2.0,
|
| 207 |
+
"eval_loss": 0.04473445564508438,
|
| 208 |
+
"eval_runtime": 22.0105,
|
| 209 |
+
"eval_samples_per_second": 9.359,
|
| 210 |
+
"eval_steps_per_second": 0.409,
|
| 211 |
+
"step": 318
|
| 212 |
},
|
| 213 |
{
|
| 214 |
+
"epoch": 2.01,
|
| 215 |
+
"learning_rate": 4.2052190435769554e-07,
|
| 216 |
+
"loss": 0.0355,
|
| 217 |
"step": 320
|
| 218 |
},
|
| 219 |
{
|
| 220 |
"epoch": 2.08,
|
| 221 |
+
"learning_rate": 4.137151834863213e-07,
|
| 222 |
+
"loss": 0.0348,
|
| 223 |
"step": 330
|
| 224 |
},
|
| 225 |
{
|
| 226 |
+
"epoch": 2.14,
|
| 227 |
+
"learning_rate": 4.0668899744407567e-07,
|
| 228 |
+
"loss": 0.0326,
|
| 229 |
"step": 340
|
| 230 |
},
|
| 231 |
{
|
| 232 |
+
"epoch": 2.2,
|
| 233 |
+
"learning_rate": 3.994527650465352e-07,
|
| 234 |
"loss": 0.0304,
|
| 235 |
"step": 350
|
| 236 |
},
|
| 237 |
{
|
| 238 |
+
"epoch": 2.26,
|
| 239 |
+
"learning_rate": 3.920161866827889e-07,
|
| 240 |
+
"loss": 0.0258,
|
| 241 |
"step": 360
|
| 242 |
},
|
| 243 |
{
|
| 244 |
"epoch": 2.33,
|
| 245 |
+
"learning_rate": 3.8438923131177237e-07,
|
| 246 |
+
"loss": 0.0317,
|
| 247 |
"step": 370
|
| 248 |
},
|
| 249 |
{
|
| 250 |
+
"epoch": 2.39,
|
| 251 |
+
"learning_rate": 3.765821230985757e-07,
|
| 252 |
+
"loss": 0.0326,
|
| 253 |
"step": 380
|
| 254 |
},
|
| 255 |
{
|
| 256 |
+
"epoch": 2.45,
|
| 257 |
+
"learning_rate": 3.6860532770864005e-07,
|
| 258 |
"loss": 0.0295,
|
| 259 |
"step": 390
|
| 260 |
},
|
| 261 |
{
|
| 262 |
"epoch": 2.52,
|
| 263 |
+
"learning_rate": 3.604695382782159e-07,
|
| 264 |
+
"loss": 0.0275,
|
| 265 |
"step": 400
|
| 266 |
},
|
| 267 |
{
|
| 268 |
+
"epoch": 2.58,
|
| 269 |
+
"learning_rate": 3.5218566107988867e-07,
|
| 270 |
+
"loss": 0.0344,
|
| 271 |
"step": 410
|
| 272 |
},
|
| 273 |
{
|
| 274 |
+
"epoch": 2.64,
|
| 275 |
+
"learning_rate": 3.4376480090239047e-07,
|
| 276 |
+
"loss": 0.0341,
|
| 277 |
"step": 420
|
| 278 |
},
|
| 279 |
{
|
| 280 |
+
"epoch": 2.7,
|
| 281 |
+
"learning_rate": 3.3521824616429284e-07,
|
| 282 |
+
"loss": 0.0273,
|
| 283 |
"step": 430
|
| 284 |
},
|
| 285 |
{
|
| 286 |
+
"epoch": 2.77,
|
| 287 |
+
"learning_rate": 3.265574537815398e-07,
|
| 288 |
"loss": 0.0314,
|
| 289 |
"step": 440
|
| 290 |
},
|
| 291 |
{
|
| 292 |
+
"epoch": 2.83,
|
| 293 |
+
"learning_rate": 3.1779403380910425e-07,
|
| 294 |
+
"loss": 0.0354,
|
| 295 |
"step": 450
|
| 296 |
},
|
| 297 |
{
|
| 298 |
+
"epoch": 2.89,
|
| 299 |
+
"learning_rate": 3.0893973387735683e-07,
|
| 300 |
+
"loss": 0.0336,
|
| 301 |
"step": 460
|
| 302 |
},
|
| 303 |
{
|
| 304 |
+
"epoch": 2.96,
|
| 305 |
+
"learning_rate": 3.000064234440111e-07,
|
| 306 |
+
"loss": 0.0285,
|
| 307 |
"step": 470
|
| 308 |
},
|
| 309 |
{
|
| 310 |
"epoch": 3.0,
|
| 311 |
+
"eval_loss": 0.03846818581223488,
|
| 312 |
+
"eval_runtime": 22.5077,
|
| 313 |
+
"eval_samples_per_second": 9.152,
|
| 314 |
+
"eval_steps_per_second": 0.4,
|
| 315 |
+
"step": 477
|
| 316 |
},
|
| 317 |
{
|
| 318 |
+
"epoch": 3.02,
|
| 319 |
+
"learning_rate": 2.910060778827554e-07,
|
| 320 |
+
"loss": 0.0259,
|
| 321 |
"step": 480
|
| 322 |
},
|
| 323 |
{
|
| 324 |
+
"epoch": 3.08,
|
| 325 |
+
"learning_rate": 2.8195076242990116e-07,
|
| 326 |
+
"loss": 0.0098,
|
| 327 |
"step": 490
|
| 328 |
},
|
| 329 |
{
|
| 330 |
+
"epoch": 3.14,
|
| 331 |
+
"learning_rate": 2.7285261601056697e-07,
|
| 332 |
+
"loss": 0.0153,
|
| 333 |
"step": 500
|
| 334 |
},
|
| 335 |
{
|
| 336 |
+
"epoch": 3.21,
|
| 337 |
+
"learning_rate": 2.6372383496608186e-07,
|
| 338 |
+
"loss": 0.0162,
|
| 339 |
"step": 510
|
| 340 |
},
|
| 341 |
{
|
| 342 |
+
"epoch": 3.27,
|
| 343 |
+
"learning_rate": 2.5457665670441937e-07,
|
| 344 |
+
"loss": 0.0155,
|
| 345 |
"step": 520
|
| 346 |
},
|
| 347 |
{
|
| 348 |
+
"epoch": 3.33,
|
| 349 |
+
"learning_rate": 2.454233432955807e-07,
|
| 350 |
+
"loss": 0.0185,
|
| 351 |
"step": 530
|
| 352 |
},
|
| 353 |
{
|
| 354 |
+
"epoch": 3.4,
|
| 355 |
+
"learning_rate": 2.3627616503391812e-07,
|
| 356 |
+
"loss": 0.0138,
|
| 357 |
"step": 540
|
| 358 |
},
|
| 359 |
{
|
| 360 |
+
"epoch": 3.46,
|
| 361 |
+
"learning_rate": 2.2714738398943308e-07,
|
| 362 |
+
"loss": 0.0136,
|
| 363 |
"step": 550
|
| 364 |
},
|
| 365 |
{
|
| 366 |
+
"epoch": 3.52,
|
| 367 |
+
"learning_rate": 2.1804923757009882e-07,
|
| 368 |
+
"loss": 0.016,
|
| 369 |
"step": 560
|
| 370 |
},
|
| 371 |
{
|
| 372 |
+
"epoch": 3.58,
|
| 373 |
+
"learning_rate": 2.089939221172446e-07,
|
| 374 |
+
"loss": 0.0155,
|
| 375 |
"step": 570
|
| 376 |
},
|
| 377 |
{
|
| 378 |
+
"epoch": 3.65,
|
| 379 |
+
"learning_rate": 1.9999357655598891e-07,
|
| 380 |
+
"loss": 0.009,
|
| 381 |
"step": 580
|
| 382 |
},
|
| 383 |
{
|
| 384 |
+
"epoch": 3.71,
|
| 385 |
+
"learning_rate": 1.9106026612264315e-07,
|
| 386 |
+
"loss": 0.0186,
|
| 387 |
"step": 590
|
| 388 |
},
|
| 389 |
{
|
| 390 |
+
"epoch": 3.77,
|
| 391 |
+
"learning_rate": 1.8220596619089573e-07,
|
| 392 |
+
"loss": 0.0109,
|
| 393 |
"step": 600
|
| 394 |
},
|
| 395 |
{
|
| 396 |
+
"epoch": 3.84,
|
| 397 |
+
"learning_rate": 1.7344254621846017e-07,
|
| 398 |
+
"loss": 0.016,
|
| 399 |
"step": 610
|
| 400 |
},
|
| 401 |
{
|
| 402 |
+
"epoch": 3.9,
|
| 403 |
+
"learning_rate": 1.647817538357072e-07,
|
| 404 |
+
"loss": 0.0111,
|
| 405 |
"step": 620
|
| 406 |
},
|
| 407 |
{
|
| 408 |
+
"epoch": 3.96,
|
| 409 |
+
"learning_rate": 1.562351990976095e-07,
|
| 410 |
+
"loss": 0.0165,
|
| 411 |
"step": 630
|
| 412 |
},
|
| 413 |
{
|
| 414 |
+
"epoch": 4.0,
|
| 415 |
+
"eval_loss": 0.04646956920623779,
|
| 416 |
+
"eval_runtime": 22.0005,
|
| 417 |
+
"eval_samples_per_second": 9.363,
|
| 418 |
+
"eval_steps_per_second": 0.409,
|
| 419 |
+
"step": 636
|
| 420 |
+
},
|
| 421 |
+
{
|
| 422 |
+
"epoch": 4.03,
|
| 423 |
+
"learning_rate": 1.478143389201113e-07,
|
| 424 |
+
"loss": 0.0104,
|
| 425 |
+
"step": 640
|
| 426 |
+
},
|
| 427 |
+
{
|
| 428 |
+
"epoch": 4.09,
|
| 429 |
+
"learning_rate": 1.3953046172178413e-07,
|
| 430 |
+
"loss": 0.0041,
|
| 431 |
+
"step": 650
|
| 432 |
+
},
|
| 433 |
+
{
|
| 434 |
+
"epoch": 4.15,
|
| 435 |
+
"learning_rate": 1.3139467229135998e-07,
|
| 436 |
+
"loss": 0.0041,
|
| 437 |
+
"step": 660
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 4.21,
|
| 441 |
+
"learning_rate": 1.2341787690142435e-07,
|
| 442 |
+
"loss": 0.0011,
|
| 443 |
+
"step": 670
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 4.28,
|
| 447 |
+
"learning_rate": 1.1561076868822755e-07,
|
| 448 |
+
"loss": 0.0028,
|
| 449 |
+
"step": 680
|
| 450 |
+
},
|
| 451 |
+
{
|
| 452 |
+
"epoch": 4.34,
|
| 453 |
+
"learning_rate": 1.0798381331721107e-07,
|
| 454 |
+
"loss": 0.0024,
|
| 455 |
+
"step": 690
|
| 456 |
+
},
|
| 457 |
+
{
|
| 458 |
+
"epoch": 4.4,
|
| 459 |
+
"learning_rate": 1.0054723495346482e-07,
|
| 460 |
+
"loss": 0.0021,
|
| 461 |
+
"step": 700
|
| 462 |
+
},
|
| 463 |
+
{
|
| 464 |
+
"epoch": 4.47,
|
| 465 |
+
"learning_rate": 9.331100255592436e-08,
|
| 466 |
+
"loss": 0.0038,
|
| 467 |
+
"step": 710
|
| 468 |
+
},
|
| 469 |
+
{
|
| 470 |
+
"epoch": 4.53,
|
| 471 |
+
"learning_rate": 8.628481651367875e-08,
|
| 472 |
+
"loss": 0.0068,
|
| 473 |
+
"step": 720
|
| 474 |
+
},
|
| 475 |
+
{
|
| 476 |
+
"epoch": 4.59,
|
| 477 |
+
"learning_rate": 7.947809564230445e-08,
|
| 478 |
+
"loss": 0.0018,
|
| 479 |
+
"step": 730
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 4.65,
|
| 483 |
+
"learning_rate": 7.289996455765748e-08,
|
| 484 |
+
"loss": 0.0035,
|
| 485 |
+
"step": 740
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 4.72,
|
| 489 |
+
"learning_rate": 6.655924144404906e-08,
|
| 490 |
+
"loss": 0.0032,
|
| 491 |
+
"step": 750
|
| 492 |
+
},
|
| 493 |
+
{
|
| 494 |
+
"epoch": 4.78,
|
| 495 |
+
"learning_rate": 6.046442623320145e-08,
|
| 496 |
+
"loss": 0.0048,
|
| 497 |
+
"step": 760
|
| 498 |
+
},
|
| 499 |
+
{
|
| 500 |
+
"epoch": 4.84,
|
| 501 |
+
"learning_rate": 5.4623689209832484e-08,
|
| 502 |
+
"loss": 0.0034,
|
| 503 |
+
"step": 770
|
| 504 |
+
},
|
| 505 |
+
{
|
| 506 |
+
"epoch": 4.91,
|
| 507 |
+
"learning_rate": 4.904486005914027e-08,
|
| 508 |
+
"loss": 0.0031,
|
| 509 |
+
"step": 780
|
| 510 |
+
},
|
| 511 |
+
{
|
| 512 |
+
"epoch": 4.97,
|
| 513 |
+
"learning_rate": 4.373541737087263e-08,
|
| 514 |
+
"loss": 0.0021,
|
| 515 |
+
"step": 790
|
| 516 |
+
},
|
| 517 |
+
{
|
| 518 |
+
"epoch": 5.0,
|
| 519 |
+
"eval_loss": 0.06586528569459915,
|
| 520 |
+
"eval_runtime": 22.2256,
|
| 521 |
+
"eval_samples_per_second": 9.269,
|
| 522 |
+
"eval_steps_per_second": 0.405,
|
| 523 |
+
"step": 795
|
| 524 |
+
},
|
| 525 |
+
{
|
| 526 |
+
"epoch": 5.03,
|
| 527 |
+
"learning_rate": 3.8702478614051345e-08,
|
| 528 |
+
"loss": 0.0037,
|
| 529 |
+
"step": 800
|
| 530 |
+
},
|
| 531 |
+
{
|
| 532 |
+
"epoch": 5.09,
|
| 533 |
+
"learning_rate": 3.3952790595787986e-08,
|
| 534 |
+
"loss": 0.0015,
|
| 535 |
+
"step": 810
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 5.16,
|
| 539 |
+
"learning_rate": 2.9492720416985e-08,
|
| 540 |
+
"loss": 0.0007,
|
| 541 |
+
"step": 820
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 5.22,
|
| 545 |
+
"learning_rate": 2.5328246937043525e-08,
|
| 546 |
+
"loss": 0.002,
|
| 547 |
+
"step": 830
|
| 548 |
+
},
|
| 549 |
+
{
|
| 550 |
+
"epoch": 5.28,
|
| 551 |
+
"learning_rate": 2.1464952759020856e-08,
|
| 552 |
+
"loss": 0.0005,
|
| 553 |
+
"step": 840
|
| 554 |
+
},
|
| 555 |
+
{
|
| 556 |
+
"epoch": 5.35,
|
| 557 |
+
"learning_rate": 1.7908016745981856e-08,
|
| 558 |
+
"loss": 0.0012,
|
| 559 |
+
"step": 850
|
| 560 |
+
},
|
| 561 |
+
{
|
| 562 |
+
"epoch": 5.41,
|
| 563 |
+
"learning_rate": 1.4662207078575684e-08,
|
| 564 |
+
"loss": 0.0005,
|
| 565 |
+
"step": 860
|
| 566 |
+
},
|
| 567 |
+
{
|
| 568 |
+
"epoch": 5.47,
|
| 569 |
+
"learning_rate": 1.1731874863145142e-08,
|
| 570 |
+
"loss": 0.0024,
|
| 571 |
+
"step": 870
|
| 572 |
+
},
|
| 573 |
+
{
|
| 574 |
+
"epoch": 5.53,
|
| 575 |
+
"learning_rate": 9.12094829893642e-09,
|
| 576 |
+
"loss": 0.0011,
|
| 577 |
+
"step": 880
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 5.6,
|
| 581 |
+
"learning_rate": 6.832927412229017e-09,
|
| 582 |
+
"loss": 0.0006,
|
| 583 |
+
"step": 890
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 5.66,
|
| 587 |
+
"learning_rate": 4.8708793644441086e-09,
|
| 588 |
+
"loss": 0.0023,
|
| 589 |
+
"step": 900
|
| 590 |
+
},
|
| 591 |
+
{
|
| 592 |
+
"epoch": 5.72,
|
| 593 |
+
"learning_rate": 3.2374343405217884e-09,
|
| 594 |
+
"loss": 0.0023,
|
| 595 |
+
"step": 910
|
| 596 |
+
},
|
| 597 |
+
{
|
| 598 |
+
"epoch": 5.79,
|
| 599 |
+
"learning_rate": 1.9347820230782295e-09,
|
| 600 |
+
"loss": 0.0005,
|
| 601 |
+
"step": 920
|
| 602 |
+
},
|
| 603 |
+
{
|
| 604 |
+
"epoch": 5.85,
|
| 605 |
+
"learning_rate": 9.64668657069706e-10,
|
| 606 |
+
"loss": 0.0023,
|
| 607 |
+
"step": 930
|
| 608 |
+
},
|
| 609 |
+
{
|
| 610 |
+
"epoch": 5.91,
|
| 611 |
+
"learning_rate": 3.2839470889836627e-10,
|
| 612 |
+
"loss": 0.0011,
|
| 613 |
+
"step": 940
|
| 614 |
+
},
|
| 615 |
+
{
|
| 616 |
+
"epoch": 5.97,
|
| 617 |
+
"learning_rate": 2.6813123097352287e-11,
|
| 618 |
+
"loss": 0.0008,
|
| 619 |
+
"step": 950
|
| 620 |
+
},
|
| 621 |
+
{
|
| 622 |
+
"epoch": 6.0,
|
| 623 |
+
"eval_loss": 0.07076797634363174,
|
| 624 |
+
"eval_runtime": 21.958,
|
| 625 |
+
"eval_samples_per_second": 9.382,
|
| 626 |
+
"eval_steps_per_second": 0.41,
|
| 627 |
+
"step": 954
|
| 628 |
},
|
| 629 |
{
|
| 630 |
+
"epoch": 6.0,
|
| 631 |
+
"step": 954,
|
| 632 |
+
"total_flos": 126473593159680.0,
|
| 633 |
+
"train_loss": 0.18272512886366676,
|
| 634 |
+
"train_runtime": 9319.8502,
|
| 635 |
+
"train_samples_per_second": 2.443,
|
| 636 |
+
"train_steps_per_second": 0.102
|
| 637 |
}
|
| 638 |
],
|
| 639 |
"logging_steps": 10,
|
| 640 |
+
"max_steps": 954,
|
| 641 |
"num_input_tokens_seen": 0,
|
| 642 |
+
"num_train_epochs": 6,
|
| 643 |
"save_steps": 500,
|
| 644 |
+
"total_flos": 126473593159680.0,
|
| 645 |
+
"train_batch_size": 8,
|
| 646 |
"trial_name": null,
|
| 647 |
"trial_params": null
|
| 648 |
}
|
training_args.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 5371
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:55228b00492b67368a9b7ba885cdae50dcc19467b0776ba18e45dcad51e7aa44
|
| 3 |
size 5371
|