Model save
Browse files- README.md +19 -21
- all_results.json +6 -11
- generation_config.json +2 -3
- train_results.json +6 -6
- trainer_state.json +970 -772
README.md
CHANGED
|
@@ -1,17 +1,13 @@
|
|
| 1 |
---
|
| 2 |
library_name: transformers
|
| 3 |
-
license:
|
| 4 |
-
base_model:
|
| 5 |
tags:
|
| 6 |
-
- alignment-handbook
|
| 7 |
-
- trl
|
| 8 |
-
- sft
|
| 9 |
-
- generated_from_trainer
|
| 10 |
- trl
|
| 11 |
- sft
|
| 12 |
- generated_from_trainer
|
| 13 |
datasets:
|
| 14 |
-
-
|
| 15 |
model-index:
|
| 16 |
- name: zephyr-7b-gemma-sft
|
| 17 |
results: []
|
|
@@ -22,9 +18,9 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 22 |
|
| 23 |
# zephyr-7b-gemma-sft
|
| 24 |
|
| 25 |
-
This model is a fine-tuned version of [
|
| 26 |
It achieves the following results on the evaluation set:
|
| 27 |
-
- Loss:
|
| 28 |
|
| 29 |
## Model description
|
| 30 |
|
|
@@ -44,14 +40,14 @@ More information needed
|
|
| 44 |
|
| 45 |
The following hyperparameters were used during training:
|
| 46 |
- learning_rate: 2e-05
|
| 47 |
-
- train_batch_size:
|
| 48 |
-
- eval_batch_size:
|
| 49 |
- seed: 42
|
| 50 |
- distributed_type: multi-GPU
|
| 51 |
- num_devices: 8
|
| 52 |
-
- gradient_accumulation_steps:
|
| 53 |
- total_train_batch_size: 128
|
| 54 |
-
- total_eval_batch_size:
|
| 55 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 56 |
- lr_scheduler_type: cosine
|
| 57 |
- lr_scheduler_warmup_ratio: 0.1
|
|
@@ -61,14 +57,16 @@ The following hyperparameters were used during training:
|
|
| 61 |
|
| 62 |
| Training Loss | Epoch | Step | Validation Loss |
|
| 63 |
|:-------------:|:------:|:----:|:---------------:|
|
| 64 |
-
|
|
| 65 |
-
| 0.
|
| 66 |
-
| 0.
|
| 67 |
-
| 0.
|
| 68 |
-
| 0.
|
| 69 |
-
| 0.
|
| 70 |
-
| 0.
|
| 71 |
-
| 0.
|
|
|
|
|
|
|
| 72 |
|
| 73 |
|
| 74 |
### Framework versions
|
|
|
|
| 1 |
---
|
| 2 |
library_name: transformers
|
| 3 |
+
license: apache-2.0
|
| 4 |
+
base_model: mistralai/Mistral-7B-v0.1
|
| 5 |
tags:
|
|
|
|
|
|
|
|
|
|
|
|
|
| 6 |
- trl
|
| 7 |
- sft
|
| 8 |
- generated_from_trainer
|
| 9 |
datasets:
|
| 10 |
+
- generator
|
| 11 |
model-index:
|
| 12 |
- name: zephyr-7b-gemma-sft
|
| 13 |
results: []
|
|
|
|
| 18 |
|
| 19 |
# zephyr-7b-gemma-sft
|
| 20 |
|
| 21 |
+
This model is a fine-tuned version of [mistralai/Mistral-7B-v0.1](https://huggingface.co/mistralai/Mistral-7B-v0.1) on the generator dataset.
|
| 22 |
It achieves the following results on the evaluation set:
|
| 23 |
+
- Loss: 0.9465
|
| 24 |
|
| 25 |
## Model description
|
| 26 |
|
|
|
|
| 40 |
|
| 41 |
The following hyperparameters were used during training:
|
| 42 |
- learning_rate: 2e-05
|
| 43 |
+
- train_batch_size: 8
|
| 44 |
+
- eval_batch_size: 8
|
| 45 |
- seed: 42
|
| 46 |
- distributed_type: multi-GPU
|
| 47 |
- num_devices: 8
|
| 48 |
+
- gradient_accumulation_steps: 2
|
| 49 |
- total_train_batch_size: 128
|
| 50 |
+
- total_eval_batch_size: 64
|
| 51 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 52 |
- lr_scheduler_type: cosine
|
| 53 |
- lr_scheduler_warmup_ratio: 0.1
|
|
|
|
| 57 |
|
| 58 |
| Training Loss | Epoch | Step | Validation Loss |
|
| 59 |
|:-------------:|:------:|:----:|:---------------:|
|
| 60 |
+
| 0.8286 | 0.2924 | 100 | 0.8580 |
|
| 61 |
+
| 0.798 | 0.5848 | 200 | 0.8706 |
|
| 62 |
+
| 0.7677 | 0.8772 | 300 | 0.8727 |
|
| 63 |
+
| 0.5584 | 1.1696 | 400 | 0.9159 |
|
| 64 |
+
| 0.5186 | 1.4620 | 500 | 0.8920 |
|
| 65 |
+
| 0.5067 | 1.7544 | 600 | 0.8780 |
|
| 66 |
+
| 0.2223 | 2.0468 | 700 | 0.9511 |
|
| 67 |
+
| 0.2114 | 2.3392 | 800 | 0.9492 |
|
| 68 |
+
| 0.1983 | 2.6316 | 900 | 0.9508 |
|
| 69 |
+
| 0.2014 | 2.9240 | 1000 | 0.9465 |
|
| 70 |
|
| 71 |
|
| 72 |
### Framework versions
|
all_results.json
CHANGED
|
@@ -1,14 +1,9 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"
|
| 4 |
-
"
|
| 5 |
-
"
|
| 6 |
-
"eval_samples_per_second": 19.072,
|
| 7 |
-
"eval_steps_per_second": 0.598,
|
| 8 |
-
"total_flos": 246978202042368.0,
|
| 9 |
-
"train_loss": 0.8698950265695153,
|
| 10 |
-
"train_runtime": 25956.5038,
|
| 11 |
"train_samples": 9500,
|
| 12 |
-
"train_samples_per_second": 4.
|
| 13 |
-
"train_steps_per_second": 0.
|
| 14 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"total_flos": 429437673799680.0,
|
| 4 |
+
"train_loss": 0.5057662985420134,
|
| 5 |
+
"train_runtime": 26805.8841,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 6 |
"train_samples": 9500,
|
| 7 |
+
"train_samples_per_second": 4.896,
|
| 8 |
+
"train_steps_per_second": 0.038
|
| 9 |
}
|
generation_config.json
CHANGED
|
@@ -1,7 +1,6 @@
|
|
| 1 |
{
|
| 2 |
"_from_model_config": true,
|
| 3 |
-
"bos_token_id":
|
| 4 |
-
"eos_token_id":
|
| 5 |
-
"pad_token_id": 0,
|
| 6 |
"transformers_version": "4.45.2"
|
| 7 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"_from_model_config": true,
|
| 3 |
+
"bos_token_id": 1,
|
| 4 |
+
"eos_token_id": 2,
|
|
|
|
| 5 |
"transformers_version": "4.45.2"
|
| 6 |
}
|
train_results.json
CHANGED
|
@@ -1,9 +1,9 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"total_flos":
|
| 4 |
-
"train_loss": 0.
|
| 5 |
-
"train_runtime":
|
| 6 |
"train_samples": 9500,
|
| 7 |
-
"train_samples_per_second": 4.
|
| 8 |
-
"train_steps_per_second": 0.
|
| 9 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"total_flos": 429437673799680.0,
|
| 4 |
+
"train_loss": 0.5057662985420134,
|
| 5 |
+
"train_runtime": 26805.8841,
|
| 6 |
"train_samples": 9500,
|
| 7 |
+
"train_samples_per_second": 4.896,
|
| 8 |
+
"train_steps_per_second": 0.038
|
| 9 |
}
|
trainer_state.json
CHANGED
|
@@ -1,1349 +1,1547 @@
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
-
"epoch":
|
| 5 |
"eval_steps": 100,
|
| 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.
|
| 13 |
-
"grad_norm":
|
| 14 |
-
"learning_rate":
|
| 15 |
-
"loss":
|
| 16 |
"step": 1
|
| 17 |
},
|
| 18 |
{
|
| 19 |
-
"epoch": 0.
|
| 20 |
-
"grad_norm":
|
| 21 |
-
"learning_rate":
|
| 22 |
-
"loss":
|
| 23 |
"step": 5
|
| 24 |
},
|
| 25 |
{
|
| 26 |
-
"epoch": 0.
|
| 27 |
-
"grad_norm":
|
| 28 |
-
"learning_rate":
|
| 29 |
-
"loss":
|
| 30 |
"step": 10
|
| 31 |
},
|
| 32 |
{
|
| 33 |
-
"epoch": 0.
|
| 34 |
-
"grad_norm":
|
| 35 |
-
"learning_rate":
|
| 36 |
-
"loss":
|
| 37 |
"step": 15
|
| 38 |
},
|
| 39 |
{
|
| 40 |
-
"epoch": 0.
|
| 41 |
-
"grad_norm":
|
| 42 |
-
"learning_rate":
|
| 43 |
-
"loss":
|
| 44 |
"step": 20
|
| 45 |
},
|
| 46 |
{
|
| 47 |
-
"epoch": 0.
|
| 48 |
-
"grad_norm":
|
| 49 |
-
"learning_rate":
|
| 50 |
-
"loss":
|
| 51 |
"step": 25
|
| 52 |
},
|
| 53 |
{
|
| 54 |
-
"epoch": 0.
|
| 55 |
-
"grad_norm":
|
| 56 |
-
"learning_rate":
|
| 57 |
-
"loss":
|
| 58 |
"step": 30
|
| 59 |
},
|
| 60 |
{
|
| 61 |
-
"epoch": 0.
|
| 62 |
-
"grad_norm":
|
| 63 |
-
"learning_rate":
|
| 64 |
-
"loss":
|
| 65 |
"step": 35
|
| 66 |
},
|
| 67 |
{
|
| 68 |
-
"epoch": 0.
|
| 69 |
-
"grad_norm":
|
| 70 |
-
"learning_rate":
|
| 71 |
-
"loss":
|
| 72 |
"step": 40
|
| 73 |
},
|
| 74 |
{
|
| 75 |
-
"epoch": 0.
|
| 76 |
-
"grad_norm":
|
| 77 |
-
"learning_rate":
|
| 78 |
-
"loss":
|
| 79 |
"step": 45
|
| 80 |
},
|
| 81 |
{
|
| 82 |
-
"epoch": 0.
|
| 83 |
-
"grad_norm":
|
| 84 |
-
"learning_rate":
|
| 85 |
-
"loss":
|
| 86 |
"step": 50
|
| 87 |
},
|
| 88 |
{
|
| 89 |
-
"epoch": 0.
|
| 90 |
-
"grad_norm":
|
| 91 |
-
"learning_rate": 1.
|
| 92 |
-
"loss":
|
| 93 |
"step": 55
|
| 94 |
},
|
| 95 |
{
|
| 96 |
-
"epoch": 0.
|
| 97 |
-
"grad_norm":
|
| 98 |
-
"learning_rate": 1.
|
| 99 |
-
"loss":
|
| 100 |
"step": 60
|
| 101 |
},
|
| 102 |
{
|
| 103 |
-
"epoch": 0.
|
| 104 |
-
"grad_norm":
|
| 105 |
-
"learning_rate": 1.
|
| 106 |
-
"loss":
|
| 107 |
"step": 65
|
| 108 |
},
|
| 109 |
{
|
| 110 |
-
"epoch": 0.
|
| 111 |
-
"grad_norm":
|
| 112 |
-
"learning_rate": 1.
|
| 113 |
-
"loss":
|
| 114 |
"step": 70
|
| 115 |
},
|
| 116 |
{
|
| 117 |
-
"epoch": 0.
|
| 118 |
-
"grad_norm":
|
| 119 |
-
"learning_rate": 1.
|
| 120 |
-
"loss":
|
| 121 |
"step": 75
|
| 122 |
},
|
| 123 |
{
|
| 124 |
-
"epoch": 0.
|
| 125 |
-
"grad_norm":
|
| 126 |
-
"learning_rate": 1.
|
| 127 |
-
"loss":
|
| 128 |
"step": 80
|
| 129 |
},
|
| 130 |
{
|
| 131 |
-
"epoch": 0.
|
| 132 |
-
"grad_norm":
|
| 133 |
-
"learning_rate": 1.
|
| 134 |
-
"loss":
|
| 135 |
"step": 85
|
| 136 |
},
|
| 137 |
{
|
| 138 |
-
"epoch": 0.
|
| 139 |
-
"grad_norm":
|
| 140 |
-
"learning_rate":
|
| 141 |
-
"loss":
|
| 142 |
"step": 90
|
| 143 |
},
|
| 144 |
{
|
| 145 |
-
"epoch": 0.
|
| 146 |
-
"grad_norm":
|
| 147 |
-
"learning_rate": 1.
|
| 148 |
-
"loss":
|
| 149 |
"step": 95
|
| 150 |
},
|
| 151 |
{
|
| 152 |
-
"epoch": 0.
|
| 153 |
-
"grad_norm":
|
| 154 |
-
"learning_rate": 1.
|
| 155 |
-
"loss":
|
| 156 |
"step": 100
|
| 157 |
},
|
| 158 |
{
|
| 159 |
-
"epoch": 0.
|
| 160 |
-
"eval_loss":
|
| 161 |
-
"eval_runtime":
|
| 162 |
-
"eval_samples_per_second":
|
| 163 |
-
"eval_steps_per_second": 0.
|
| 164 |
"step": 100
|
| 165 |
},
|
| 166 |
{
|
| 167 |
-
"epoch": 0.
|
| 168 |
-
"grad_norm":
|
| 169 |
-
"learning_rate": 1.
|
| 170 |
-
"loss":
|
| 171 |
"step": 105
|
| 172 |
},
|
| 173 |
{
|
| 174 |
-
"epoch": 0.
|
| 175 |
-
"grad_norm":
|
| 176 |
-
"learning_rate": 1.
|
| 177 |
-
"loss":
|
| 178 |
"step": 110
|
| 179 |
},
|
| 180 |
{
|
| 181 |
-
"epoch": 0.
|
| 182 |
-
"grad_norm":
|
| 183 |
-
"learning_rate": 1.
|
| 184 |
-
"loss":
|
| 185 |
"step": 115
|
| 186 |
},
|
| 187 |
{
|
| 188 |
-
"epoch": 0.
|
| 189 |
-
"grad_norm":
|
| 190 |
-
"learning_rate": 1.
|
| 191 |
-
"loss":
|
| 192 |
"step": 120
|
| 193 |
},
|
| 194 |
{
|
| 195 |
-
"epoch": 0.
|
| 196 |
-
"grad_norm":
|
| 197 |
-
"learning_rate": 1.
|
| 198 |
-
"loss":
|
| 199 |
"step": 125
|
| 200 |
},
|
| 201 |
{
|
| 202 |
-
"epoch": 0.
|
| 203 |
-
"grad_norm":
|
| 204 |
-
"learning_rate": 1.
|
| 205 |
-
"loss":
|
| 206 |
"step": 130
|
| 207 |
},
|
| 208 |
{
|
| 209 |
-
"epoch": 0.
|
| 210 |
-
"grad_norm":
|
| 211 |
-
"learning_rate": 1.
|
| 212 |
-
"loss": 0.
|
| 213 |
"step": 135
|
| 214 |
},
|
| 215 |
{
|
| 216 |
-
"epoch": 0.
|
| 217 |
-
"grad_norm":
|
| 218 |
-
"learning_rate": 1.
|
| 219 |
-
"loss":
|
| 220 |
"step": 140
|
| 221 |
},
|
| 222 |
{
|
| 223 |
-
"epoch": 0.
|
| 224 |
-
"grad_norm":
|
| 225 |
-
"learning_rate": 1.
|
| 226 |
-
"loss":
|
| 227 |
"step": 145
|
| 228 |
},
|
| 229 |
{
|
| 230 |
-
"epoch": 0.
|
| 231 |
-
"grad_norm":
|
| 232 |
-
"learning_rate": 1.
|
| 233 |
-
"loss":
|
| 234 |
"step": 150
|
| 235 |
},
|
| 236 |
{
|
| 237 |
-
"epoch": 0.
|
| 238 |
-
"grad_norm":
|
| 239 |
-
"learning_rate": 1.
|
| 240 |
-
"loss":
|
| 241 |
"step": 155
|
| 242 |
},
|
| 243 |
{
|
| 244 |
-
"epoch": 0.
|
| 245 |
-
"grad_norm":
|
| 246 |
-
"learning_rate": 1.
|
| 247 |
-
"loss":
|
| 248 |
"step": 160
|
| 249 |
},
|
| 250 |
{
|
| 251 |
-
"epoch": 0.
|
| 252 |
-
"grad_norm":
|
| 253 |
-
"learning_rate": 1.
|
| 254 |
-
"loss":
|
| 255 |
"step": 165
|
| 256 |
},
|
| 257 |
{
|
| 258 |
-
"epoch": 0.
|
| 259 |
-
"grad_norm":
|
| 260 |
-
"learning_rate": 1.
|
| 261 |
-
"loss":
|
| 262 |
"step": 170
|
| 263 |
},
|
| 264 |
{
|
| 265 |
-
"epoch": 0.
|
| 266 |
-
"grad_norm":
|
| 267 |
-
"learning_rate": 1.
|
| 268 |
-
"loss":
|
| 269 |
"step": 175
|
| 270 |
},
|
| 271 |
{
|
| 272 |
-
"epoch": 0.
|
| 273 |
-
"grad_norm":
|
| 274 |
-
"learning_rate": 1.
|
| 275 |
-
"loss":
|
| 276 |
"step": 180
|
| 277 |
},
|
| 278 |
{
|
| 279 |
-
"epoch": 0.
|
| 280 |
-
"grad_norm":
|
| 281 |
-
"learning_rate": 1.
|
| 282 |
-
"loss":
|
| 283 |
"step": 185
|
| 284 |
},
|
| 285 |
{
|
| 286 |
-
"epoch": 0.
|
| 287 |
-
"grad_norm":
|
| 288 |
-
"learning_rate": 1.
|
| 289 |
-
"loss":
|
| 290 |
"step": 190
|
| 291 |
},
|
| 292 |
{
|
| 293 |
-
"epoch": 0.
|
| 294 |
-
"grad_norm":
|
| 295 |
-
"learning_rate": 1.
|
| 296 |
-
"loss":
|
| 297 |
"step": 195
|
| 298 |
},
|
| 299 |
{
|
| 300 |
-
"epoch": 0.
|
| 301 |
-
"grad_norm":
|
| 302 |
-
"learning_rate": 1.
|
| 303 |
-
"loss": 0.
|
| 304 |
"step": 200
|
| 305 |
},
|
| 306 |
{
|
| 307 |
-
"epoch": 0.
|
| 308 |
-
"eval_loss":
|
| 309 |
-
"eval_runtime":
|
| 310 |
-
"eval_samples_per_second":
|
| 311 |
-
"eval_steps_per_second": 0.
|
| 312 |
"step": 200
|
| 313 |
},
|
| 314 |
{
|
| 315 |
-
"epoch": 0.
|
| 316 |
-
"grad_norm":
|
| 317 |
-
"learning_rate": 1.
|
| 318 |
-
"loss":
|
| 319 |
"step": 205
|
| 320 |
},
|
| 321 |
{
|
| 322 |
-
"epoch": 0.
|
| 323 |
-
"grad_norm":
|
| 324 |
-
"learning_rate": 1.
|
| 325 |
-
"loss":
|
| 326 |
"step": 210
|
| 327 |
},
|
| 328 |
{
|
| 329 |
-
"epoch": 0.
|
| 330 |
-
"grad_norm":
|
| 331 |
-
"learning_rate": 1.
|
| 332 |
-
"loss":
|
| 333 |
"step": 215
|
| 334 |
},
|
| 335 |
{
|
| 336 |
-
"epoch": 0.
|
| 337 |
-
"grad_norm":
|
| 338 |
-
"learning_rate": 1.
|
| 339 |
-
"loss":
|
| 340 |
"step": 220
|
| 341 |
},
|
| 342 |
{
|
| 343 |
-
"epoch": 0.
|
| 344 |
-
"grad_norm":
|
| 345 |
-
"learning_rate": 1.
|
| 346 |
-
"loss":
|
| 347 |
"step": 225
|
| 348 |
},
|
| 349 |
{
|
| 350 |
-
"epoch": 0.
|
| 351 |
-
"grad_norm":
|
| 352 |
-
"learning_rate": 1.
|
| 353 |
-
"loss":
|
| 354 |
"step": 230
|
| 355 |
},
|
| 356 |
{
|
| 357 |
-
"epoch": 0.
|
| 358 |
-
"grad_norm":
|
| 359 |
-
"learning_rate": 1.
|
| 360 |
-
"loss":
|
| 361 |
"step": 235
|
| 362 |
},
|
| 363 |
{
|
| 364 |
-
"epoch": 0.
|
| 365 |
-
"grad_norm":
|
| 366 |
-
"learning_rate": 1.
|
| 367 |
-
"loss": 0.
|
| 368 |
"step": 240
|
| 369 |
},
|
| 370 |
{
|
| 371 |
-
"epoch": 0.
|
| 372 |
-
"grad_norm":
|
| 373 |
-
"learning_rate": 1.
|
| 374 |
-
"loss": 0.
|
| 375 |
"step": 245
|
| 376 |
},
|
| 377 |
{
|
| 378 |
-
"epoch": 0.
|
| 379 |
-
"grad_norm":
|
| 380 |
-
"learning_rate": 1.
|
| 381 |
-
"loss": 0.
|
| 382 |
"step": 250
|
| 383 |
},
|
| 384 |
{
|
| 385 |
-
"epoch": 0.
|
| 386 |
-
"grad_norm":
|
| 387 |
-
"learning_rate": 1.
|
| 388 |
-
"loss": 0.
|
| 389 |
"step": 255
|
| 390 |
},
|
| 391 |
{
|
| 392 |
-
"epoch": 0.
|
| 393 |
-
"grad_norm":
|
| 394 |
-
"learning_rate": 1.
|
| 395 |
-
"loss": 0.
|
| 396 |
"step": 260
|
| 397 |
},
|
| 398 |
{
|
| 399 |
-
"epoch": 0.
|
| 400 |
-
"grad_norm":
|
| 401 |
-
"learning_rate": 1.
|
| 402 |
-
"loss": 0.
|
| 403 |
"step": 265
|
| 404 |
},
|
| 405 |
{
|
| 406 |
-
"epoch": 0.
|
| 407 |
-
"grad_norm":
|
| 408 |
-
"learning_rate": 1.
|
| 409 |
-
"loss": 0.
|
| 410 |
"step": 270
|
| 411 |
},
|
| 412 |
{
|
| 413 |
-
"epoch": 0.
|
| 414 |
-
"grad_norm":
|
| 415 |
-
"learning_rate": 1.
|
| 416 |
-
"loss": 0.
|
| 417 |
"step": 275
|
| 418 |
},
|
| 419 |
{
|
| 420 |
-
"epoch": 0.
|
| 421 |
-
"grad_norm":
|
| 422 |
-
"learning_rate": 1.
|
| 423 |
-
"loss": 0.
|
| 424 |
"step": 280
|
| 425 |
},
|
| 426 |
{
|
| 427 |
-
"epoch": 0.
|
| 428 |
-
"grad_norm":
|
| 429 |
-
"learning_rate": 1.
|
| 430 |
-
"loss": 0.
|
| 431 |
"step": 285
|
| 432 |
},
|
| 433 |
{
|
| 434 |
-
"epoch": 0.
|
| 435 |
-
"grad_norm":
|
| 436 |
-
"learning_rate": 1.
|
| 437 |
-
"loss": 0.
|
| 438 |
"step": 290
|
| 439 |
},
|
| 440 |
{
|
| 441 |
-
"epoch": 0.
|
| 442 |
-
"grad_norm":
|
| 443 |
-
"learning_rate": 1.
|
| 444 |
-
"loss": 0.
|
| 445 |
"step": 295
|
| 446 |
},
|
| 447 |
{
|
| 448 |
-
"epoch":
|
| 449 |
-
"grad_norm":
|
| 450 |
-
"learning_rate": 1.
|
| 451 |
-
"loss": 0.
|
| 452 |
"step": 300
|
| 453 |
},
|
| 454 |
{
|
| 455 |
-
"epoch":
|
| 456 |
-
"eval_loss":
|
| 457 |
-
"eval_runtime":
|
| 458 |
-
"eval_samples_per_second":
|
| 459 |
-
"eval_steps_per_second": 0.
|
| 460 |
"step": 300
|
| 461 |
},
|
| 462 |
{
|
| 463 |
-
"epoch":
|
| 464 |
-
"grad_norm":
|
| 465 |
-
"learning_rate": 1.
|
| 466 |
-
"loss": 0.
|
| 467 |
"step": 305
|
| 468 |
},
|
| 469 |
{
|
| 470 |
-
"epoch":
|
| 471 |
-
"grad_norm":
|
| 472 |
-
"learning_rate": 1.
|
| 473 |
-
"loss": 0.
|
| 474 |
"step": 310
|
| 475 |
},
|
| 476 |
{
|
| 477 |
-
"epoch":
|
| 478 |
-
"grad_norm":
|
| 479 |
-
"learning_rate": 1.
|
| 480 |
-
"loss": 0.
|
| 481 |
"step": 315
|
| 482 |
},
|
| 483 |
{
|
| 484 |
-
"epoch":
|
| 485 |
-
"grad_norm":
|
| 486 |
-
"learning_rate": 1.
|
| 487 |
-
"loss": 0.
|
| 488 |
"step": 320
|
| 489 |
},
|
| 490 |
{
|
| 491 |
-
"epoch":
|
| 492 |
-
"grad_norm":
|
| 493 |
-
"learning_rate": 1.
|
| 494 |
-
"loss": 0.
|
| 495 |
"step": 325
|
| 496 |
},
|
| 497 |
{
|
| 498 |
-
"epoch":
|
| 499 |
-
"grad_norm":
|
| 500 |
-
"learning_rate": 1.
|
| 501 |
-
"loss": 0.
|
| 502 |
"step": 330
|
| 503 |
},
|
| 504 |
{
|
| 505 |
-
"epoch":
|
| 506 |
-
"grad_norm":
|
| 507 |
-
"learning_rate": 1.
|
| 508 |
-
"loss": 0.
|
| 509 |
"step": 335
|
| 510 |
},
|
| 511 |
{
|
| 512 |
-
"epoch":
|
| 513 |
-
"grad_norm":
|
| 514 |
-
"learning_rate": 1.
|
| 515 |
-
"loss": 0.
|
| 516 |
"step": 340
|
| 517 |
},
|
| 518 |
{
|
| 519 |
-
"epoch": 1.
|
| 520 |
-
"grad_norm":
|
| 521 |
-
"learning_rate": 1.
|
| 522 |
-
"loss": 0.
|
| 523 |
"step": 345
|
| 524 |
},
|
| 525 |
{
|
| 526 |
-
"epoch": 1.
|
| 527 |
-
"grad_norm":
|
| 528 |
-
"learning_rate": 1.
|
| 529 |
-
"loss": 0.
|
| 530 |
"step": 350
|
| 531 |
},
|
| 532 |
{
|
| 533 |
-
"epoch": 1.
|
| 534 |
-
"grad_norm":
|
| 535 |
-
"learning_rate": 1.
|
| 536 |
-
"loss": 0.
|
| 537 |
"step": 355
|
| 538 |
},
|
| 539 |
{
|
| 540 |
-
"epoch": 1.
|
| 541 |
-
"grad_norm":
|
| 542 |
-
"learning_rate": 1.
|
| 543 |
-
"loss": 0.
|
| 544 |
"step": 360
|
| 545 |
},
|
| 546 |
{
|
| 547 |
-
"epoch": 1.
|
| 548 |
-
"grad_norm":
|
| 549 |
-
"learning_rate": 1.
|
| 550 |
-
"loss": 0.
|
| 551 |
"step": 365
|
| 552 |
},
|
| 553 |
{
|
| 554 |
-
"epoch": 1.
|
| 555 |
-
"grad_norm":
|
| 556 |
-
"learning_rate": 1.
|
| 557 |
-
"loss": 0.
|
| 558 |
"step": 370
|
| 559 |
},
|
| 560 |
{
|
| 561 |
-
"epoch": 1.
|
| 562 |
-
"grad_norm":
|
| 563 |
-
"learning_rate": 1.
|
| 564 |
-
"loss": 0.
|
| 565 |
"step": 375
|
| 566 |
},
|
| 567 |
{
|
| 568 |
-
"epoch": 1.
|
| 569 |
-
"grad_norm":
|
| 570 |
-
"learning_rate": 1.
|
| 571 |
-
"loss": 0.
|
| 572 |
"step": 380
|
| 573 |
},
|
| 574 |
{
|
| 575 |
-
"epoch": 1.
|
| 576 |
-
"grad_norm":
|
| 577 |
-
"learning_rate": 1.
|
| 578 |
-
"loss": 0.
|
| 579 |
"step": 385
|
| 580 |
},
|
| 581 |
{
|
| 582 |
-
"epoch": 1.
|
| 583 |
-
"grad_norm":
|
| 584 |
-
"learning_rate": 1.
|
| 585 |
-
"loss": 0.
|
| 586 |
"step": 390
|
| 587 |
},
|
| 588 |
{
|
| 589 |
-
"epoch": 1.
|
| 590 |
-
"grad_norm":
|
| 591 |
-
"learning_rate": 1.
|
| 592 |
-
"loss": 0.
|
| 593 |
"step": 395
|
| 594 |
},
|
| 595 |
{
|
| 596 |
-
"epoch": 1.
|
| 597 |
-
"grad_norm":
|
| 598 |
-
"learning_rate": 1.
|
| 599 |
-
"loss": 0.
|
| 600 |
"step": 400
|
| 601 |
},
|
| 602 |
{
|
| 603 |
-
"epoch": 1.
|
| 604 |
-
"eval_loss":
|
| 605 |
-
"eval_runtime":
|
| 606 |
-
"eval_samples_per_second":
|
| 607 |
-
"eval_steps_per_second": 0.
|
| 608 |
"step": 400
|
| 609 |
},
|
| 610 |
{
|
| 611 |
-
"epoch": 1.
|
| 612 |
-
"grad_norm":
|
| 613 |
-
"learning_rate": 1.
|
| 614 |
-
"loss": 0.
|
| 615 |
"step": 405
|
| 616 |
},
|
| 617 |
{
|
| 618 |
-
"epoch": 1.
|
| 619 |
-
"grad_norm":
|
| 620 |
-
"learning_rate": 1.
|
| 621 |
-
"loss": 0.
|
| 622 |
"step": 410
|
| 623 |
},
|
| 624 |
{
|
| 625 |
-
"epoch": 1.
|
| 626 |
-
"grad_norm":
|
| 627 |
-
"learning_rate": 1.
|
| 628 |
-
"loss": 0.
|
| 629 |
"step": 415
|
| 630 |
},
|
| 631 |
{
|
| 632 |
-
"epoch": 1.
|
| 633 |
-
"grad_norm":
|
| 634 |
-
"learning_rate": 1.
|
| 635 |
-
"loss": 0.
|
| 636 |
"step": 420
|
| 637 |
},
|
| 638 |
{
|
| 639 |
-
"epoch": 1.
|
| 640 |
-
"grad_norm":
|
| 641 |
-
"learning_rate": 1.
|
| 642 |
-
"loss": 0.
|
| 643 |
"step": 425
|
| 644 |
},
|
| 645 |
{
|
| 646 |
-
"epoch": 1.
|
| 647 |
-
"grad_norm":
|
| 648 |
-
"learning_rate": 1.
|
| 649 |
-
"loss": 0.
|
| 650 |
"step": 430
|
| 651 |
},
|
| 652 |
{
|
| 653 |
-
"epoch": 1.
|
| 654 |
-
"grad_norm":
|
| 655 |
-
"learning_rate": 1.
|
| 656 |
-
"loss": 0.
|
| 657 |
"step": 435
|
| 658 |
},
|
| 659 |
{
|
| 660 |
-
"epoch": 1.
|
| 661 |
-
"grad_norm":
|
| 662 |
-
"learning_rate": 1.
|
| 663 |
-
"loss": 0.
|
| 664 |
"step": 440
|
| 665 |
},
|
| 666 |
{
|
| 667 |
-
"epoch": 1.
|
| 668 |
-
"grad_norm":
|
| 669 |
-
"learning_rate": 1.
|
| 670 |
-
"loss": 0.
|
| 671 |
"step": 445
|
| 672 |
},
|
| 673 |
{
|
| 674 |
-
"epoch": 1.
|
| 675 |
-
"grad_norm":
|
| 676 |
-
"learning_rate": 1.
|
| 677 |
-
"loss": 0.
|
| 678 |
"step": 450
|
| 679 |
},
|
| 680 |
{
|
| 681 |
-
"epoch": 1.
|
| 682 |
-
"grad_norm":
|
| 683 |
-
"learning_rate": 1.
|
| 684 |
-
"loss": 0.
|
| 685 |
"step": 455
|
| 686 |
},
|
| 687 |
{
|
| 688 |
-
"epoch": 1.
|
| 689 |
-
"grad_norm":
|
| 690 |
-
"learning_rate": 1.
|
| 691 |
-
"loss": 0.
|
| 692 |
"step": 460
|
| 693 |
},
|
| 694 |
{
|
| 695 |
-
"epoch": 1.
|
| 696 |
-
"grad_norm":
|
| 697 |
-
"learning_rate": 1.
|
| 698 |
-
"loss": 0.
|
| 699 |
"step": 465
|
| 700 |
},
|
| 701 |
{
|
| 702 |
-
"epoch": 1.
|
| 703 |
-
"grad_norm":
|
| 704 |
-
"learning_rate": 1.
|
| 705 |
-
"loss": 0.
|
| 706 |
"step": 470
|
| 707 |
},
|
| 708 |
{
|
| 709 |
-
"epoch": 1.
|
| 710 |
-
"grad_norm":
|
| 711 |
-
"learning_rate": 1.
|
| 712 |
-
"loss": 0.
|
| 713 |
"step": 475
|
| 714 |
},
|
| 715 |
{
|
| 716 |
-
"epoch": 1.
|
| 717 |
-
"grad_norm":
|
| 718 |
-
"learning_rate": 1.
|
| 719 |
-
"loss": 0.
|
| 720 |
"step": 480
|
| 721 |
},
|
| 722 |
{
|
| 723 |
-
"epoch": 1.
|
| 724 |
-
"grad_norm":
|
| 725 |
-
"learning_rate": 1.
|
| 726 |
-
"loss": 0.
|
| 727 |
"step": 485
|
| 728 |
},
|
| 729 |
{
|
| 730 |
-
"epoch": 1.
|
| 731 |
-
"grad_norm":
|
| 732 |
-
"learning_rate": 1.
|
| 733 |
-
"loss": 0.
|
| 734 |
"step": 490
|
| 735 |
},
|
| 736 |
{
|
| 737 |
-
"epoch": 1.
|
| 738 |
-
"grad_norm":
|
| 739 |
-
"learning_rate":
|
| 740 |
-
"loss": 0.
|
| 741 |
"step": 495
|
| 742 |
},
|
| 743 |
{
|
| 744 |
-
"epoch": 1.
|
| 745 |
-
"grad_norm":
|
| 746 |
-
"learning_rate":
|
| 747 |
-
"loss": 0.
|
| 748 |
"step": 500
|
| 749 |
},
|
| 750 |
{
|
| 751 |
-
"epoch": 1.
|
| 752 |
-
"eval_loss":
|
| 753 |
-
"eval_runtime":
|
| 754 |
-
"eval_samples_per_second":
|
| 755 |
-
"eval_steps_per_second": 0.
|
| 756 |
"step": 500
|
| 757 |
},
|
| 758 |
{
|
| 759 |
-
"epoch": 1.
|
| 760 |
-
"grad_norm":
|
| 761 |
-
"learning_rate":
|
| 762 |
-
"loss": 0.
|
| 763 |
"step": 505
|
| 764 |
},
|
| 765 |
{
|
| 766 |
-
"epoch": 1.
|
| 767 |
-
"grad_norm":
|
| 768 |
-
"learning_rate":
|
| 769 |
-
"loss": 0.
|
| 770 |
"step": 510
|
| 771 |
},
|
| 772 |
{
|
| 773 |
-
"epoch": 1.
|
| 774 |
-
"grad_norm":
|
| 775 |
-
"learning_rate":
|
| 776 |
-
"loss": 0.
|
| 777 |
"step": 515
|
| 778 |
},
|
| 779 |
{
|
| 780 |
-
"epoch": 1.
|
| 781 |
-
"grad_norm":
|
| 782 |
-
"learning_rate":
|
| 783 |
-
"loss": 0.
|
| 784 |
"step": 520
|
| 785 |
},
|
| 786 |
{
|
| 787 |
-
"epoch": 1.
|
| 788 |
-
"grad_norm":
|
| 789 |
-
"learning_rate":
|
| 790 |
-
"loss": 0.
|
| 791 |
"step": 525
|
| 792 |
},
|
| 793 |
{
|
| 794 |
-
"epoch": 1.
|
| 795 |
-
"grad_norm":
|
| 796 |
-
"learning_rate":
|
| 797 |
-
"loss": 0.
|
| 798 |
"step": 530
|
| 799 |
},
|
| 800 |
{
|
| 801 |
-
"epoch": 1.
|
| 802 |
-
"grad_norm":
|
| 803 |
-
"learning_rate":
|
| 804 |
-
"loss": 0.
|
| 805 |
"step": 535
|
| 806 |
},
|
| 807 |
{
|
| 808 |
-
"epoch": 1.
|
| 809 |
-
"grad_norm":
|
| 810 |
-
"learning_rate":
|
| 811 |
-
"loss": 0.
|
| 812 |
"step": 540
|
| 813 |
},
|
| 814 |
{
|
| 815 |
-
"epoch": 1.
|
| 816 |
-
"grad_norm":
|
| 817 |
-
"learning_rate":
|
| 818 |
-
"loss": 0.
|
| 819 |
"step": 545
|
| 820 |
},
|
| 821 |
{
|
| 822 |
-
"epoch": 1.
|
| 823 |
-
"grad_norm":
|
| 824 |
-
"learning_rate":
|
| 825 |
-
"loss": 0.
|
| 826 |
"step": 550
|
| 827 |
},
|
| 828 |
{
|
| 829 |
-
"epoch": 1.
|
| 830 |
-
"grad_norm":
|
| 831 |
-
"learning_rate":
|
| 832 |
-
"loss": 0.
|
| 833 |
"step": 555
|
| 834 |
},
|
| 835 |
{
|
| 836 |
-
"epoch": 1.
|
| 837 |
-
"grad_norm":
|
| 838 |
-
"learning_rate":
|
| 839 |
-
"loss": 0.
|
| 840 |
"step": 560
|
| 841 |
},
|
| 842 |
{
|
| 843 |
-
"epoch": 1.
|
| 844 |
-
"grad_norm":
|
| 845 |
-
"learning_rate":
|
| 846 |
-
"loss": 0.
|
| 847 |
"step": 565
|
| 848 |
},
|
| 849 |
{
|
| 850 |
-
"epoch": 1.
|
| 851 |
-
"grad_norm":
|
| 852 |
-
"learning_rate":
|
| 853 |
-
"loss": 0.
|
| 854 |
"step": 570
|
| 855 |
},
|
| 856 |
{
|
| 857 |
-
"epoch": 1.
|
| 858 |
-
"grad_norm":
|
| 859 |
-
"learning_rate":
|
| 860 |
-
"loss": 0.
|
| 861 |
"step": 575
|
| 862 |
},
|
| 863 |
{
|
| 864 |
-
"epoch": 1.
|
| 865 |
-
"grad_norm":
|
| 866 |
-
"learning_rate":
|
| 867 |
-
"loss": 0.
|
| 868 |
"step": 580
|
| 869 |
},
|
| 870 |
{
|
| 871 |
-
"epoch": 1.
|
| 872 |
-
"grad_norm":
|
| 873 |
-
"learning_rate":
|
| 874 |
-
"loss": 0.
|
| 875 |
"step": 585
|
| 876 |
},
|
| 877 |
{
|
| 878 |
-
"epoch": 1.
|
| 879 |
-
"grad_norm":
|
| 880 |
-
"learning_rate":
|
| 881 |
-
"loss": 0.
|
| 882 |
"step": 590
|
| 883 |
},
|
| 884 |
{
|
| 885 |
-
"epoch": 1.
|
| 886 |
-
"grad_norm":
|
| 887 |
-
"learning_rate":
|
| 888 |
-
"loss": 0.
|
| 889 |
"step": 595
|
| 890 |
},
|
| 891 |
{
|
| 892 |
-
"epoch":
|
| 893 |
-
"grad_norm":
|
| 894 |
-
"learning_rate":
|
| 895 |
-
"loss": 0.
|
| 896 |
"step": 600
|
| 897 |
},
|
| 898 |
{
|
| 899 |
-
"epoch":
|
| 900 |
-
"eval_loss":
|
| 901 |
-
"eval_runtime":
|
| 902 |
-
"eval_samples_per_second":
|
| 903 |
-
"eval_steps_per_second": 0.
|
| 904 |
"step": 600
|
| 905 |
},
|
| 906 |
{
|
| 907 |
-
"epoch":
|
| 908 |
-
"grad_norm":
|
| 909 |
-
"learning_rate":
|
| 910 |
-
"loss": 0.
|
| 911 |
"step": 605
|
| 912 |
},
|
| 913 |
{
|
| 914 |
-
"epoch":
|
| 915 |
-
"grad_norm":
|
| 916 |
-
"learning_rate":
|
| 917 |
-
"loss": 0.
|
| 918 |
"step": 610
|
| 919 |
},
|
| 920 |
{
|
| 921 |
-
"epoch":
|
| 922 |
-
"grad_norm":
|
| 923 |
-
"learning_rate":
|
| 924 |
-
"loss": 0.
|
| 925 |
"step": 615
|
| 926 |
},
|
| 927 |
{
|
| 928 |
-
"epoch":
|
| 929 |
-
"grad_norm":
|
| 930 |
-
"learning_rate":
|
| 931 |
-
"loss": 0.
|
| 932 |
"step": 620
|
| 933 |
},
|
| 934 |
{
|
| 935 |
-
"epoch":
|
| 936 |
-
"grad_norm":
|
| 937 |
-
"learning_rate":
|
| 938 |
-
"loss": 0.
|
| 939 |
"step": 625
|
| 940 |
},
|
| 941 |
{
|
| 942 |
-
"epoch":
|
| 943 |
-
"grad_norm":
|
| 944 |
-
"learning_rate":
|
| 945 |
-
"loss": 0.
|
| 946 |
"step": 630
|
| 947 |
},
|
| 948 |
{
|
| 949 |
-
"epoch":
|
| 950 |
-
"grad_norm":
|
| 951 |
-
"learning_rate":
|
| 952 |
-
"loss": 0.
|
| 953 |
"step": 635
|
| 954 |
},
|
| 955 |
{
|
| 956 |
-
"epoch":
|
| 957 |
-
"grad_norm":
|
| 958 |
-
"learning_rate":
|
| 959 |
-
"loss": 0.
|
| 960 |
"step": 640
|
| 961 |
},
|
| 962 |
{
|
| 963 |
-
"epoch":
|
| 964 |
-
"grad_norm":
|
| 965 |
-
"learning_rate":
|
| 966 |
-
"loss": 0.
|
| 967 |
"step": 645
|
| 968 |
},
|
| 969 |
{
|
| 970 |
-
"epoch":
|
| 971 |
-
"grad_norm":
|
| 972 |
-
"learning_rate":
|
| 973 |
-
"loss": 0.
|
| 974 |
"step": 650
|
| 975 |
},
|
| 976 |
{
|
| 977 |
-
"epoch":
|
| 978 |
-
"grad_norm":
|
| 979 |
-
"learning_rate":
|
| 980 |
-
"loss": 0.
|
| 981 |
"step": 655
|
| 982 |
},
|
| 983 |
{
|
| 984 |
-
"epoch":
|
| 985 |
-
"grad_norm":
|
| 986 |
-
"learning_rate":
|
| 987 |
-
"loss": 0.
|
| 988 |
"step": 660
|
| 989 |
},
|
| 990 |
{
|
| 991 |
-
"epoch":
|
| 992 |
-
"grad_norm":
|
| 993 |
-
"learning_rate":
|
| 994 |
-
"loss": 0.
|
| 995 |
"step": 665
|
| 996 |
},
|
| 997 |
{
|
| 998 |
-
"epoch":
|
| 999 |
-
"grad_norm":
|
| 1000 |
-
"learning_rate":
|
| 1001 |
-
"loss": 0.
|
| 1002 |
"step": 670
|
| 1003 |
},
|
| 1004 |
{
|
| 1005 |
-
"epoch":
|
| 1006 |
-
"grad_norm":
|
| 1007 |
-
"learning_rate":
|
| 1008 |
-
"loss": 0.
|
| 1009 |
"step": 675
|
| 1010 |
},
|
| 1011 |
{
|
| 1012 |
-
"epoch":
|
| 1013 |
-
"grad_norm":
|
| 1014 |
-
"learning_rate":
|
| 1015 |
-
"loss": 0.
|
| 1016 |
"step": 680
|
| 1017 |
},
|
| 1018 |
{
|
| 1019 |
-
"epoch": 2.
|
| 1020 |
-
"grad_norm":
|
| 1021 |
-
"learning_rate":
|
| 1022 |
-
"loss": 0.
|
| 1023 |
"step": 685
|
| 1024 |
},
|
| 1025 |
{
|
| 1026 |
-
"epoch": 2.
|
| 1027 |
-
"grad_norm": 2.
|
| 1028 |
-
"learning_rate":
|
| 1029 |
-
"loss": 0.
|
| 1030 |
"step": 690
|
| 1031 |
},
|
| 1032 |
{
|
| 1033 |
-
"epoch": 2.
|
| 1034 |
-
"grad_norm":
|
| 1035 |
-
"learning_rate":
|
| 1036 |
-
"loss": 0.
|
| 1037 |
"step": 695
|
| 1038 |
},
|
| 1039 |
{
|
| 1040 |
-
"epoch": 2.
|
| 1041 |
-
"grad_norm":
|
| 1042 |
-
"learning_rate":
|
| 1043 |
-
"loss": 0.
|
| 1044 |
"step": 700
|
| 1045 |
},
|
| 1046 |
{
|
| 1047 |
-
"epoch": 2.
|
| 1048 |
-
"eval_loss":
|
| 1049 |
-
"eval_runtime":
|
| 1050 |
-
"eval_samples_per_second":
|
| 1051 |
-
"eval_steps_per_second": 0.
|
| 1052 |
"step": 700
|
| 1053 |
},
|
| 1054 |
{
|
| 1055 |
-
"epoch": 2.
|
| 1056 |
-
"grad_norm":
|
| 1057 |
-
"learning_rate":
|
| 1058 |
-
"loss": 0.
|
| 1059 |
"step": 705
|
| 1060 |
},
|
| 1061 |
{
|
| 1062 |
-
"epoch": 2.
|
| 1063 |
-
"grad_norm":
|
| 1064 |
-
"learning_rate":
|
| 1065 |
-
"loss": 0.
|
| 1066 |
"step": 710
|
| 1067 |
},
|
| 1068 |
{
|
| 1069 |
-
"epoch": 2.
|
| 1070 |
-
"grad_norm":
|
| 1071 |
-
"learning_rate":
|
| 1072 |
-
"loss": 0.
|
| 1073 |
"step": 715
|
| 1074 |
},
|
| 1075 |
{
|
| 1076 |
-
"epoch": 2.
|
| 1077 |
-
"grad_norm":
|
| 1078 |
-
"learning_rate":
|
| 1079 |
-
"loss": 0.
|
| 1080 |
"step": 720
|
| 1081 |
},
|
| 1082 |
{
|
| 1083 |
-
"epoch": 2.
|
| 1084 |
-
"grad_norm": 1.
|
| 1085 |
-
"learning_rate":
|
| 1086 |
-
"loss": 0.
|
| 1087 |
"step": 725
|
| 1088 |
},
|
| 1089 |
{
|
| 1090 |
-
"epoch": 2.
|
| 1091 |
-
"grad_norm":
|
| 1092 |
-
"learning_rate":
|
| 1093 |
-
"loss": 0.
|
| 1094 |
"step": 730
|
| 1095 |
},
|
| 1096 |
{
|
| 1097 |
-
"epoch": 2.
|
| 1098 |
-
"grad_norm":
|
| 1099 |
-
"learning_rate":
|
| 1100 |
-
"loss": 0.
|
| 1101 |
"step": 735
|
| 1102 |
},
|
| 1103 |
{
|
| 1104 |
-
"epoch": 2.
|
| 1105 |
-
"grad_norm": 1.
|
| 1106 |
-
"learning_rate":
|
| 1107 |
-
"loss": 0.
|
| 1108 |
"step": 740
|
| 1109 |
},
|
| 1110 |
{
|
| 1111 |
-
"epoch": 2.
|
| 1112 |
-
"grad_norm":
|
| 1113 |
-
"learning_rate":
|
| 1114 |
-
"loss": 0.
|
| 1115 |
"step": 745
|
| 1116 |
},
|
| 1117 |
{
|
| 1118 |
-
"epoch": 2.
|
| 1119 |
-
"grad_norm":
|
| 1120 |
-
"learning_rate":
|
| 1121 |
-
"loss": 0.
|
| 1122 |
"step": 750
|
| 1123 |
},
|
| 1124 |
{
|
| 1125 |
-
"epoch": 2.
|
| 1126 |
-
"grad_norm": 1.
|
| 1127 |
-
"learning_rate":
|
| 1128 |
-
"loss": 0.
|
| 1129 |
"step": 755
|
| 1130 |
},
|
| 1131 |
{
|
| 1132 |
-
"epoch": 2.
|
| 1133 |
-
"grad_norm":
|
| 1134 |
-
"learning_rate":
|
| 1135 |
-
"loss": 0.
|
| 1136 |
"step": 760
|
| 1137 |
},
|
| 1138 |
{
|
| 1139 |
-
"epoch": 2.
|
| 1140 |
-
"grad_norm": 1.
|
| 1141 |
-
"learning_rate":
|
| 1142 |
-
"loss": 0.
|
| 1143 |
"step": 765
|
| 1144 |
},
|
| 1145 |
{
|
| 1146 |
-
"epoch": 2.
|
| 1147 |
-
"grad_norm": 1.
|
| 1148 |
-
"learning_rate":
|
| 1149 |
-
"loss": 0.
|
| 1150 |
"step": 770
|
| 1151 |
},
|
| 1152 |
{
|
| 1153 |
-
"epoch": 2.
|
| 1154 |
-
"grad_norm": 1.
|
| 1155 |
-
"learning_rate":
|
| 1156 |
-
"loss": 0.
|
| 1157 |
"step": 775
|
| 1158 |
},
|
| 1159 |
{
|
| 1160 |
-
"epoch": 2.
|
| 1161 |
-
"grad_norm":
|
| 1162 |
-
"learning_rate":
|
| 1163 |
-
"loss": 0.
|
| 1164 |
"step": 780
|
| 1165 |
},
|
| 1166 |
{
|
| 1167 |
-
"epoch": 2.
|
| 1168 |
-
"grad_norm": 1.
|
| 1169 |
-
"learning_rate":
|
| 1170 |
-
"loss": 0.
|
| 1171 |
"step": 785
|
| 1172 |
},
|
| 1173 |
{
|
| 1174 |
-
"epoch": 2.
|
| 1175 |
-
"grad_norm":
|
| 1176 |
-
"learning_rate":
|
| 1177 |
-
"loss": 0.
|
| 1178 |
"step": 790
|
| 1179 |
},
|
| 1180 |
{
|
| 1181 |
-
"epoch": 2.
|
| 1182 |
-
"grad_norm": 1.
|
| 1183 |
-
"learning_rate":
|
| 1184 |
-
"loss": 0.
|
| 1185 |
"step": 795
|
| 1186 |
},
|
| 1187 |
{
|
| 1188 |
-
"epoch": 2.
|
| 1189 |
-
"grad_norm": 1.
|
| 1190 |
-
"learning_rate":
|
| 1191 |
-
"loss": 0.
|
| 1192 |
"step": 800
|
| 1193 |
},
|
| 1194 |
{
|
| 1195 |
-
"epoch": 2.
|
| 1196 |
-
"eval_loss":
|
| 1197 |
-
"eval_runtime":
|
| 1198 |
-
"eval_samples_per_second":
|
| 1199 |
-
"eval_steps_per_second": 0.
|
| 1200 |
"step": 800
|
| 1201 |
},
|
| 1202 |
{
|
| 1203 |
-
"epoch": 2.
|
| 1204 |
-
"grad_norm": 1.
|
| 1205 |
-
"learning_rate":
|
| 1206 |
-
"loss": 0.
|
| 1207 |
"step": 805
|
| 1208 |
},
|
| 1209 |
{
|
| 1210 |
-
"epoch": 2.
|
| 1211 |
-
"grad_norm": 1.
|
| 1212 |
-
"learning_rate":
|
| 1213 |
-
"loss": 0.
|
| 1214 |
"step": 810
|
| 1215 |
},
|
| 1216 |
{
|
| 1217 |
-
"epoch": 2.
|
| 1218 |
-
"grad_norm": 1.
|
| 1219 |
-
"learning_rate":
|
| 1220 |
-
"loss": 0.
|
| 1221 |
"step": 815
|
| 1222 |
},
|
| 1223 |
{
|
| 1224 |
-
"epoch": 2.
|
| 1225 |
-
"grad_norm": 1.
|
| 1226 |
-
"learning_rate":
|
| 1227 |
-
"loss": 0.
|
| 1228 |
"step": 820
|
| 1229 |
},
|
| 1230 |
{
|
| 1231 |
-
"epoch": 2.
|
| 1232 |
-
"grad_norm": 1.
|
| 1233 |
-
"learning_rate":
|
| 1234 |
-
"loss": 0.
|
| 1235 |
"step": 825
|
| 1236 |
},
|
| 1237 |
{
|
| 1238 |
-
"epoch": 2.
|
| 1239 |
-
"grad_norm":
|
| 1240 |
-
"learning_rate":
|
| 1241 |
-
"loss": 0.
|
| 1242 |
"step": 830
|
| 1243 |
},
|
| 1244 |
{
|
| 1245 |
-
"epoch": 2.
|
| 1246 |
-
"grad_norm": 1.
|
| 1247 |
-
"learning_rate": 2.
|
| 1248 |
-
"loss": 0.
|
| 1249 |
"step": 835
|
| 1250 |
},
|
| 1251 |
{
|
| 1252 |
-
"epoch": 2.
|
| 1253 |
-
"grad_norm": 1.
|
| 1254 |
-
"learning_rate":
|
| 1255 |
-
"loss": 0.
|
| 1256 |
"step": 840
|
| 1257 |
},
|
| 1258 |
{
|
| 1259 |
-
"epoch": 2.
|
| 1260 |
-
"grad_norm": 1.
|
| 1261 |
-
"learning_rate":
|
| 1262 |
-
"loss": 0.
|
| 1263 |
"step": 845
|
| 1264 |
},
|
| 1265 |
{
|
| 1266 |
-
"epoch": 2.
|
| 1267 |
-
"grad_norm": 1.
|
| 1268 |
-
"learning_rate": 1.
|
| 1269 |
-
"loss": 0.
|
| 1270 |
"step": 850
|
| 1271 |
},
|
| 1272 |
{
|
| 1273 |
-
"epoch": 2.
|
| 1274 |
-
"grad_norm": 1.
|
| 1275 |
-
"learning_rate": 1.
|
| 1276 |
-
"loss": 0.
|
| 1277 |
"step": 855
|
| 1278 |
},
|
| 1279 |
{
|
| 1280 |
-
"epoch": 2.
|
| 1281 |
-
"grad_norm": 1.
|
| 1282 |
-
"learning_rate": 1.
|
| 1283 |
-
"loss": 0.
|
| 1284 |
"step": 860
|
| 1285 |
},
|
| 1286 |
{
|
| 1287 |
-
"epoch": 2.
|
| 1288 |
-
"grad_norm": 1.
|
| 1289 |
-
"learning_rate":
|
| 1290 |
-
"loss": 0.
|
| 1291 |
"step": 865
|
| 1292 |
},
|
| 1293 |
{
|
| 1294 |
-
"epoch": 2.
|
| 1295 |
-
"grad_norm": 1.
|
| 1296 |
-
"learning_rate":
|
| 1297 |
-
"loss": 0.
|
| 1298 |
"step": 870
|
| 1299 |
},
|
| 1300 |
{
|
| 1301 |
-
"epoch": 2.
|
| 1302 |
-
"grad_norm": 1.
|
| 1303 |
-
"learning_rate":
|
| 1304 |
-
"loss": 0.
|
| 1305 |
"step": 875
|
| 1306 |
},
|
| 1307 |
{
|
| 1308 |
-
"epoch": 2.
|
| 1309 |
-
"grad_norm": 1.
|
| 1310 |
-
"learning_rate":
|
| 1311 |
-
"loss": 0.
|
| 1312 |
"step": 880
|
| 1313 |
},
|
| 1314 |
{
|
| 1315 |
-
"epoch": 2.
|
| 1316 |
-
"grad_norm": 1.
|
| 1317 |
-
"learning_rate": 1.
|
| 1318 |
-
"loss": 0.
|
| 1319 |
"step": 885
|
| 1320 |
},
|
| 1321 |
{
|
| 1322 |
-
"epoch": 2.
|
| 1323 |
-
"grad_norm": 1.
|
| 1324 |
-
"learning_rate":
|
| 1325 |
-
"loss": 0.
|
| 1326 |
"step": 890
|
| 1327 |
},
|
| 1328 |
{
|
| 1329 |
-
"epoch": 2.
|
| 1330 |
-
"grad_norm": 1.
|
| 1331 |
-
"learning_rate":
|
| 1332 |
-
"loss": 0.
|
| 1333 |
"step": 895
|
| 1334 |
},
|
| 1335 |
{
|
| 1336 |
-
"epoch": 2.
|
| 1337 |
-
"
|
| 1338 |
-
"
|
| 1339 |
-
"
|
| 1340 |
-
"
|
| 1341 |
-
|
| 1342 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1343 |
}
|
| 1344 |
],
|
| 1345 |
"logging_steps": 5,
|
| 1346 |
-
"max_steps":
|
| 1347 |
"num_input_tokens_seen": 0,
|
| 1348 |
"num_train_epochs": 3,
|
| 1349 |
"save_steps": 100,
|
|
@@ -1359,8 +1557,8 @@
|
|
| 1359 |
"attributes": {}
|
| 1360 |
}
|
| 1361 |
},
|
| 1362 |
-
"total_flos":
|
| 1363 |
-
"train_batch_size":
|
| 1364 |
"trial_name": null,
|
| 1365 |
"trial_params": null
|
| 1366 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 3.0,
|
| 5 |
"eval_steps": 100,
|
| 6 |
+
"global_step": 1026,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
+
"epoch": 0.0029239766081871343,
|
| 13 |
+
"grad_norm": 6.1624870962638925,
|
| 14 |
+
"learning_rate": 1.941747572815534e-07,
|
| 15 |
+
"loss": 0.913,
|
| 16 |
"step": 1
|
| 17 |
},
|
| 18 |
{
|
| 19 |
+
"epoch": 0.014619883040935672,
|
| 20 |
+
"grad_norm": 11.812860926129314,
|
| 21 |
+
"learning_rate": 9.70873786407767e-07,
|
| 22 |
+
"loss": 0.8859,
|
| 23 |
"step": 5
|
| 24 |
},
|
| 25 |
{
|
| 26 |
+
"epoch": 0.029239766081871343,
|
| 27 |
+
"grad_norm": 2.6140064947786015,
|
| 28 |
+
"learning_rate": 1.941747572815534e-06,
|
| 29 |
+
"loss": 0.8409,
|
| 30 |
"step": 10
|
| 31 |
},
|
| 32 |
{
|
| 33 |
+
"epoch": 0.043859649122807015,
|
| 34 |
+
"grad_norm": 2.2391899324318154,
|
| 35 |
+
"learning_rate": 2.912621359223301e-06,
|
| 36 |
+
"loss": 0.8325,
|
| 37 |
"step": 15
|
| 38 |
},
|
| 39 |
{
|
| 40 |
+
"epoch": 0.05847953216374269,
|
| 41 |
+
"grad_norm": 2.11742528178852,
|
| 42 |
+
"learning_rate": 3.883495145631068e-06,
|
| 43 |
+
"loss": 0.8169,
|
| 44 |
"step": 20
|
| 45 |
},
|
| 46 |
{
|
| 47 |
+
"epoch": 0.07309941520467836,
|
| 48 |
+
"grad_norm": 2.088530291916816,
|
| 49 |
+
"learning_rate": 4.854368932038836e-06,
|
| 50 |
+
"loss": 0.8168,
|
| 51 |
"step": 25
|
| 52 |
},
|
| 53 |
{
|
| 54 |
+
"epoch": 0.08771929824561403,
|
| 55 |
+
"grad_norm": 2.028716635962563,
|
| 56 |
+
"learning_rate": 5.825242718446602e-06,
|
| 57 |
+
"loss": 0.8075,
|
| 58 |
"step": 30
|
| 59 |
},
|
| 60 |
{
|
| 61 |
+
"epoch": 0.1023391812865497,
|
| 62 |
+
"grad_norm": 2.3256620167562234,
|
| 63 |
+
"learning_rate": 6.79611650485437e-06,
|
| 64 |
+
"loss": 0.7953,
|
| 65 |
"step": 35
|
| 66 |
},
|
| 67 |
{
|
| 68 |
+
"epoch": 0.11695906432748537,
|
| 69 |
+
"grad_norm": 2.349007546868107,
|
| 70 |
+
"learning_rate": 7.766990291262136e-06,
|
| 71 |
+
"loss": 0.7841,
|
| 72 |
"step": 40
|
| 73 |
},
|
| 74 |
{
|
| 75 |
+
"epoch": 0.13157894736842105,
|
| 76 |
+
"grad_norm": 2.2423399474385666,
|
| 77 |
+
"learning_rate": 8.737864077669904e-06,
|
| 78 |
+
"loss": 0.7888,
|
| 79 |
"step": 45
|
| 80 |
},
|
| 81 |
{
|
| 82 |
+
"epoch": 0.14619883040935672,
|
| 83 |
+
"grad_norm": 1.942782247095399,
|
| 84 |
+
"learning_rate": 9.708737864077671e-06,
|
| 85 |
+
"loss": 0.8042,
|
| 86 |
"step": 50
|
| 87 |
},
|
| 88 |
{
|
| 89 |
+
"epoch": 0.1608187134502924,
|
| 90 |
+
"grad_norm": 2.1945226195890033,
|
| 91 |
+
"learning_rate": 1.0679611650485437e-05,
|
| 92 |
+
"loss": 0.8018,
|
| 93 |
"step": 55
|
| 94 |
},
|
| 95 |
{
|
| 96 |
+
"epoch": 0.17543859649122806,
|
| 97 |
+
"grad_norm": 1.8960792380527152,
|
| 98 |
+
"learning_rate": 1.1650485436893204e-05,
|
| 99 |
+
"loss": 0.7832,
|
| 100 |
"step": 60
|
| 101 |
},
|
| 102 |
{
|
| 103 |
+
"epoch": 0.19005847953216373,
|
| 104 |
+
"grad_norm": 2.050931671194075,
|
| 105 |
+
"learning_rate": 1.2621359223300974e-05,
|
| 106 |
+
"loss": 0.8048,
|
| 107 |
"step": 65
|
| 108 |
},
|
| 109 |
{
|
| 110 |
+
"epoch": 0.2046783625730994,
|
| 111 |
+
"grad_norm": 2.0914042301099043,
|
| 112 |
+
"learning_rate": 1.359223300970874e-05,
|
| 113 |
+
"loss": 0.7988,
|
| 114 |
"step": 70
|
| 115 |
},
|
| 116 |
{
|
| 117 |
+
"epoch": 0.21929824561403508,
|
| 118 |
+
"grad_norm": 2.1780402538204804,
|
| 119 |
+
"learning_rate": 1.4563106796116507e-05,
|
| 120 |
+
"loss": 0.8046,
|
| 121 |
"step": 75
|
| 122 |
},
|
| 123 |
{
|
| 124 |
+
"epoch": 0.23391812865497075,
|
| 125 |
+
"grad_norm": 2.2507384711871077,
|
| 126 |
+
"learning_rate": 1.5533980582524273e-05,
|
| 127 |
+
"loss": 0.8101,
|
| 128 |
"step": 80
|
| 129 |
},
|
| 130 |
{
|
| 131 |
+
"epoch": 0.24853801169590642,
|
| 132 |
+
"grad_norm": 2.0740107954310987,
|
| 133 |
+
"learning_rate": 1.650485436893204e-05,
|
| 134 |
+
"loss": 0.7992,
|
| 135 |
"step": 85
|
| 136 |
},
|
| 137 |
{
|
| 138 |
+
"epoch": 0.2631578947368421,
|
| 139 |
+
"grad_norm": 2.1703649782676426,
|
| 140 |
+
"learning_rate": 1.7475728155339808e-05,
|
| 141 |
+
"loss": 0.8168,
|
| 142 |
"step": 90
|
| 143 |
},
|
| 144 |
{
|
| 145 |
+
"epoch": 0.2777777777777778,
|
| 146 |
+
"grad_norm": 2.4978786575354817,
|
| 147 |
+
"learning_rate": 1.8446601941747575e-05,
|
| 148 |
+
"loss": 0.806,
|
| 149 |
"step": 95
|
| 150 |
},
|
| 151 |
{
|
| 152 |
+
"epoch": 0.29239766081871343,
|
| 153 |
+
"grad_norm": 2.0698254347384983,
|
| 154 |
+
"learning_rate": 1.9417475728155343e-05,
|
| 155 |
+
"loss": 0.8286,
|
| 156 |
"step": 100
|
| 157 |
},
|
| 158 |
{
|
| 159 |
+
"epoch": 0.29239766081871343,
|
| 160 |
+
"eval_loss": 0.8580158948898315,
|
| 161 |
+
"eval_runtime": 94.9126,
|
| 162 |
+
"eval_samples_per_second": 22.99,
|
| 163 |
+
"eval_steps_per_second": 0.369,
|
| 164 |
"step": 100
|
| 165 |
},
|
| 166 |
{
|
| 167 |
+
"epoch": 0.30701754385964913,
|
| 168 |
+
"grad_norm": 2.0735037748593723,
|
| 169 |
+
"learning_rate": 1.999976830073192e-05,
|
| 170 |
+
"loss": 0.8281,
|
| 171 |
"step": 105
|
| 172 |
},
|
| 173 |
{
|
| 174 |
+
"epoch": 0.3216374269005848,
|
| 175 |
+
"grad_norm": 2.095073626207817,
|
| 176 |
+
"learning_rate": 1.99971618072711e-05,
|
| 177 |
+
"loss": 0.8174,
|
| 178 |
"step": 110
|
| 179 |
},
|
| 180 |
{
|
| 181 |
+
"epoch": 0.3362573099415205,
|
| 182 |
+
"grad_norm": 2.1603082170396455,
|
| 183 |
+
"learning_rate": 1.9991659953668803e-05,
|
| 184 |
+
"loss": 0.8217,
|
| 185 |
"step": 115
|
| 186 |
},
|
| 187 |
{
|
| 188 |
+
"epoch": 0.3508771929824561,
|
| 189 |
+
"grad_norm": 2.0326092354167637,
|
| 190 |
+
"learning_rate": 1.998326433336204e-05,
|
| 191 |
+
"loss": 0.8381,
|
| 192 |
"step": 120
|
| 193 |
},
|
| 194 |
{
|
| 195 |
+
"epoch": 0.3654970760233918,
|
| 196 |
+
"grad_norm": 2.226630427942925,
|
| 197 |
+
"learning_rate": 1.997197737787537e-05,
|
| 198 |
+
"loss": 0.8261,
|
| 199 |
"step": 125
|
| 200 |
},
|
| 201 |
{
|
| 202 |
+
"epoch": 0.38011695906432746,
|
| 203 |
+
"grad_norm": 2.119406491731454,
|
| 204 |
+
"learning_rate": 1.9957802356116665e-05,
|
| 205 |
+
"loss": 0.8208,
|
| 206 |
"step": 130
|
| 207 |
},
|
| 208 |
{
|
| 209 |
+
"epoch": 0.39473684210526316,
|
| 210 |
+
"grad_norm": 2.8253278464362084,
|
| 211 |
+
"learning_rate": 1.994074337343041e-05,
|
| 212 |
+
"loss": 0.8129,
|
| 213 |
"step": 135
|
| 214 |
},
|
| 215 |
{
|
| 216 |
+
"epoch": 0.4093567251461988,
|
| 217 |
+
"grad_norm": 2.4882232695538464,
|
| 218 |
+
"learning_rate": 1.9920805370408695e-05,
|
| 219 |
+
"loss": 0.8135,
|
| 220 |
"step": 140
|
| 221 |
},
|
| 222 |
{
|
| 223 |
+
"epoch": 0.4239766081871345,
|
| 224 |
+
"grad_norm": 4.780582760365185,
|
| 225 |
+
"learning_rate": 1.9897994121460322e-05,
|
| 226 |
+
"loss": 0.8364,
|
| 227 |
"step": 145
|
| 228 |
},
|
| 229 |
{
|
| 230 |
+
"epoch": 0.43859649122807015,
|
| 231 |
+
"grad_norm": 11.5182408239238,
|
| 232 |
+
"learning_rate": 1.9872316233138463e-05,
|
| 233 |
+
"loss": 0.8277,
|
| 234 |
"step": 150
|
| 235 |
},
|
| 236 |
{
|
| 237 |
+
"epoch": 0.45321637426900585,
|
| 238 |
+
"grad_norm": 12.627096663789404,
|
| 239 |
+
"learning_rate": 1.9843779142227258e-05,
|
| 240 |
+
"loss": 0.8196,
|
| 241 |
"step": 155
|
| 242 |
},
|
| 243 |
{
|
| 244 |
+
"epoch": 0.4678362573099415,
|
| 245 |
+
"grad_norm": 2.740111270404608,
|
| 246 |
+
"learning_rate": 1.981239111358799e-05,
|
| 247 |
+
"loss": 0.8042,
|
| 248 |
"step": 160
|
| 249 |
},
|
| 250 |
{
|
| 251 |
+
"epoch": 0.4824561403508772,
|
| 252 |
+
"grad_norm": 2.174778620165311,
|
| 253 |
+
"learning_rate": 1.9778161237765438e-05,
|
| 254 |
+
"loss": 0.8064,
|
| 255 |
"step": 165
|
| 256 |
},
|
| 257 |
{
|
| 258 |
+
"epoch": 0.49707602339181284,
|
| 259 |
+
"grad_norm": 1.9246861870071321,
|
| 260 |
+
"learning_rate": 1.9741099428355075e-05,
|
| 261 |
+
"loss": 0.8024,
|
| 262 |
"step": 170
|
| 263 |
},
|
| 264 |
{
|
| 265 |
+
"epoch": 0.5116959064327485,
|
| 266 |
+
"grad_norm": 9.798315852556506,
|
| 267 |
+
"learning_rate": 1.9701216419131934e-05,
|
| 268 |
+
"loss": 0.8352,
|
| 269 |
"step": 175
|
| 270 |
},
|
| 271 |
{
|
| 272 |
+
"epoch": 0.5263157894736842,
|
| 273 |
+
"grad_norm": 2.9139360459829082,
|
| 274 |
+
"learning_rate": 1.9658523760941882e-05,
|
| 275 |
+
"loss": 0.8033,
|
| 276 |
"step": 180
|
| 277 |
},
|
| 278 |
{
|
| 279 |
+
"epoch": 0.5409356725146199,
|
| 280 |
+
"grad_norm": 14.543721579870974,
|
| 281 |
+
"learning_rate": 1.9613033818356322e-05,
|
| 282 |
+
"loss": 0.8631,
|
| 283 |
"step": 185
|
| 284 |
},
|
| 285 |
{
|
| 286 |
+
"epoch": 0.5555555555555556,
|
| 287 |
+
"grad_norm": 11.532560091003726,
|
| 288 |
+
"learning_rate": 1.956475976609114e-05,
|
| 289 |
+
"loss": 0.8304,
|
| 290 |
"step": 190
|
| 291 |
},
|
| 292 |
{
|
| 293 |
+
"epoch": 0.5701754385964912,
|
| 294 |
+
"grad_norm": 10.191690209836944,
|
| 295 |
+
"learning_rate": 1.951371558519111e-05,
|
| 296 |
+
"loss": 0.8021,
|
| 297 |
"step": 195
|
| 298 |
},
|
| 299 |
{
|
| 300 |
+
"epoch": 0.5847953216374269,
|
| 301 |
+
"grad_norm": 2.490323350347935,
|
| 302 |
+
"learning_rate": 1.9459916058980707e-05,
|
| 303 |
+
"loss": 0.798,
|
| 304 |
"step": 200
|
| 305 |
},
|
| 306 |
{
|
| 307 |
+
"epoch": 0.5847953216374269,
|
| 308 |
+
"eval_loss": 0.8706345558166504,
|
| 309 |
+
"eval_runtime": 94.8283,
|
| 310 |
+
"eval_samples_per_second": 23.01,
|
| 311 |
+
"eval_steps_per_second": 0.369,
|
| 312 |
"step": 200
|
| 313 |
},
|
| 314 |
{
|
| 315 |
+
"epoch": 0.5994152046783626,
|
| 316 |
+
"grad_norm": 2.025711297854962,
|
| 317 |
+
"learning_rate": 1.9403376768782592e-05,
|
| 318 |
+
"loss": 0.7884,
|
| 319 |
"step": 205
|
| 320 |
},
|
| 321 |
{
|
| 322 |
+
"epoch": 0.6140350877192983,
|
| 323 |
+
"grad_norm": 1.9488358870140705,
|
| 324 |
+
"learning_rate": 1.9344114089404983e-05,
|
| 325 |
+
"loss": 0.7804,
|
| 326 |
"step": 210
|
| 327 |
},
|
| 328 |
{
|
| 329 |
+
"epoch": 0.6286549707602339,
|
| 330 |
+
"grad_norm": 1.8564140890417253,
|
| 331 |
+
"learning_rate": 1.9282145184399197e-05,
|
| 332 |
+
"loss": 0.808,
|
| 333 |
"step": 215
|
| 334 |
},
|
| 335 |
{
|
| 336 |
+
"epoch": 0.6432748538011696,
|
| 337 |
+
"grad_norm": 2.0881170345282727,
|
| 338 |
+
"learning_rate": 1.9217488001088784e-05,
|
| 339 |
+
"loss": 0.799,
|
| 340 |
"step": 220
|
| 341 |
},
|
| 342 |
{
|
| 343 |
+
"epoch": 0.6578947368421053,
|
| 344 |
+
"grad_norm": 1.9678597098298427,
|
| 345 |
+
"learning_rate": 1.9150161265371663e-05,
|
| 346 |
+
"loss": 0.795,
|
| 347 |
"step": 225
|
| 348 |
},
|
| 349 |
{
|
| 350 |
+
"epoch": 0.672514619883041,
|
| 351 |
+
"grad_norm": 2.865460519803586,
|
| 352 |
+
"learning_rate": 1.908018447629674e-05,
|
| 353 |
+
"loss": 0.7915,
|
| 354 |
"step": 230
|
| 355 |
},
|
| 356 |
{
|
| 357 |
+
"epoch": 0.6871345029239766,
|
| 358 |
+
"grad_norm": 2.398699191637217,
|
| 359 |
+
"learning_rate": 1.9007577900416648e-05,
|
| 360 |
+
"loss": 0.7814,
|
| 361 |
"step": 235
|
| 362 |
},
|
| 363 |
{
|
| 364 |
+
"epoch": 0.7017543859649122,
|
| 365 |
+
"grad_norm": 2.1260051093101353,
|
| 366 |
+
"learning_rate": 1.8932362565918167e-05,
|
| 367 |
+
"loss": 0.7983,
|
| 368 |
"step": 240
|
| 369 |
},
|
| 370 |
{
|
| 371 |
+
"epoch": 0.716374269005848,
|
| 372 |
+
"grad_norm": 5.80073712189035,
|
| 373 |
+
"learning_rate": 1.8854560256532098e-05,
|
| 374 |
+
"loss": 0.7691,
|
| 375 |
"step": 245
|
| 376 |
},
|
| 377 |
{
|
| 378 |
+
"epoch": 0.7309941520467836,
|
| 379 |
+
"grad_norm": 2.5341535901655896,
|
| 380 |
+
"learning_rate": 1.877419350522429e-05,
|
| 381 |
+
"loss": 0.8155,
|
| 382 |
"step": 250
|
| 383 |
},
|
| 384 |
{
|
| 385 |
+
"epoch": 0.7456140350877193,
|
| 386 |
+
"grad_norm": 1.95626689403789,
|
| 387 |
+
"learning_rate": 1.869128558766965e-05,
|
| 388 |
+
"loss": 0.7878,
|
| 389 |
"step": 255
|
| 390 |
},
|
| 391 |
{
|
| 392 |
+
"epoch": 0.7602339181286549,
|
| 393 |
+
"grad_norm": 15.120401535093198,
|
| 394 |
+
"learning_rate": 1.8605860515511158e-05,
|
| 395 |
+
"loss": 0.7883,
|
| 396 |
"step": 260
|
| 397 |
},
|
| 398 |
{
|
| 399 |
+
"epoch": 0.7748538011695907,
|
| 400 |
+
"grad_norm": 2.2607309079656135,
|
| 401 |
+
"learning_rate": 1.8517943029405577e-05,
|
| 402 |
+
"loss": 0.7999,
|
| 403 |
"step": 265
|
| 404 |
},
|
| 405 |
{
|
| 406 |
+
"epoch": 0.7894736842105263,
|
| 407 |
+
"grad_norm": 2.295750513860874,
|
| 408 |
+
"learning_rate": 1.8427558591858164e-05,
|
| 409 |
+
"loss": 0.7697,
|
| 410 |
"step": 270
|
| 411 |
},
|
| 412 |
{
|
| 413 |
+
"epoch": 0.804093567251462,
|
| 414 |
+
"grad_norm": 4.1135088544849445,
|
| 415 |
+
"learning_rate": 1.833473337984823e-05,
|
| 416 |
+
"loss": 0.8065,
|
| 417 |
"step": 275
|
| 418 |
},
|
| 419 |
{
|
| 420 |
+
"epoch": 0.8187134502923976,
|
| 421 |
+
"grad_norm": 3.2241188905244895,
|
| 422 |
+
"learning_rate": 1.823949427724785e-05,
|
| 423 |
+
"loss": 0.799,
|
| 424 |
"step": 280
|
| 425 |
},
|
| 426 |
{
|
| 427 |
+
"epoch": 0.8333333333333334,
|
| 428 |
+
"grad_norm": 1.6843445283218286,
|
| 429 |
+
"learning_rate": 1.8141868867035745e-05,
|
| 430 |
+
"loss": 0.7909,
|
| 431 |
"step": 285
|
| 432 |
},
|
| 433 |
{
|
| 434 |
+
"epoch": 0.847953216374269,
|
| 435 |
+
"grad_norm": 1.7626650921963332,
|
| 436 |
+
"learning_rate": 1.8041885423308808e-05,
|
| 437 |
+
"loss": 0.781,
|
| 438 |
"step": 290
|
| 439 |
},
|
| 440 |
{
|
| 441 |
+
"epoch": 0.8625730994152047,
|
| 442 |
+
"grad_norm": 10.38812175281534,
|
| 443 |
+
"learning_rate": 1.7939572903093383e-05,
|
| 444 |
+
"loss": 0.7923,
|
| 445 |
"step": 295
|
| 446 |
},
|
| 447 |
{
|
| 448 |
+
"epoch": 0.8771929824561403,
|
| 449 |
+
"grad_norm": 3.1451964354292596,
|
| 450 |
+
"learning_rate": 1.7834960937958775e-05,
|
| 451 |
+
"loss": 0.7677,
|
| 452 |
"step": 300
|
| 453 |
},
|
| 454 |
{
|
| 455 |
+
"epoch": 0.8771929824561403,
|
| 456 |
+
"eval_loss": 0.8726791739463806,
|
| 457 |
+
"eval_runtime": 93.9416,
|
| 458 |
+
"eval_samples_per_second": 23.227,
|
| 459 |
+
"eval_steps_per_second": 0.373,
|
| 460 |
"step": 300
|
| 461 |
},
|
| 462 |
{
|
| 463 |
+
"epoch": 0.8918128654970761,
|
| 464 |
+
"grad_norm": 1.9970765983454766,
|
| 465 |
+
"learning_rate": 1.7728079825435426e-05,
|
| 466 |
+
"loss": 0.7966,
|
| 467 |
"step": 305
|
| 468 |
},
|
| 469 |
{
|
| 470 |
+
"epoch": 0.9064327485380117,
|
| 471 |
+
"grad_norm": 1.8313570465257998,
|
| 472 |
+
"learning_rate": 1.761896052024019e-05,
|
| 473 |
+
"loss": 0.7784,
|
| 474 |
"step": 310
|
| 475 |
},
|
| 476 |
{
|
| 477 |
+
"epoch": 0.9210526315789473,
|
| 478 |
+
"grad_norm": 2.028945141590537,
|
| 479 |
+
"learning_rate": 1.7507634625311283e-05,
|
| 480 |
+
"loss": 0.7569,
|
| 481 |
"step": 315
|
| 482 |
},
|
| 483 |
{
|
| 484 |
+
"epoch": 0.935672514619883,
|
| 485 |
+
"grad_norm": 3.1385176933971706,
|
| 486 |
+
"learning_rate": 1.7394134382655496e-05,
|
| 487 |
+
"loss": 0.7805,
|
| 488 |
"step": 320
|
| 489 |
},
|
| 490 |
{
|
| 491 |
+
"epoch": 0.9502923976608187,
|
| 492 |
+
"grad_norm": 1.7723196199368927,
|
| 493 |
+
"learning_rate": 1.7278492664010326e-05,
|
| 494 |
+
"loss": 0.7681,
|
| 495 |
"step": 325
|
| 496 |
},
|
| 497 |
{
|
| 498 |
+
"epoch": 0.9649122807017544,
|
| 499 |
+
"grad_norm": 1.7577488585231846,
|
| 500 |
+
"learning_rate": 1.7160742961323714e-05,
|
| 501 |
+
"loss": 0.7672,
|
| 502 |
"step": 330
|
| 503 |
},
|
| 504 |
{
|
| 505 |
+
"epoch": 0.97953216374269,
|
| 506 |
+
"grad_norm": 2.3636063665348757,
|
| 507 |
+
"learning_rate": 1.7040919377054182e-05,
|
| 508 |
+
"loss": 0.7795,
|
| 509 |
"step": 335
|
| 510 |
},
|
| 511 |
{
|
| 512 |
+
"epoch": 0.9941520467836257,
|
| 513 |
+
"grad_norm": 1.7709587208644213,
|
| 514 |
+
"learning_rate": 1.6919056614294133e-05,
|
| 515 |
+
"loss": 0.7624,
|
| 516 |
"step": 340
|
| 517 |
},
|
| 518 |
{
|
| 519 |
+
"epoch": 1.0087719298245614,
|
| 520 |
+
"grad_norm": 3.8073195101773765,
|
| 521 |
+
"learning_rate": 1.6795189966719182e-05,
|
| 522 |
+
"loss": 0.6426,
|
| 523 |
"step": 345
|
| 524 |
},
|
| 525 |
{
|
| 526 |
+
"epoch": 1.023391812865497,
|
| 527 |
+
"grad_norm": 2.787413972058394,
|
| 528 |
+
"learning_rate": 1.666935530836651e-05,
|
| 529 |
+
"loss": 0.5366,
|
| 530 |
"step": 350
|
| 531 |
},
|
| 532 |
{
|
| 533 |
+
"epoch": 1.0380116959064327,
|
| 534 |
+
"grad_norm": 2.362170165993678,
|
| 535 |
+
"learning_rate": 1.654158908324504e-05,
|
| 536 |
+
"loss": 0.5377,
|
| 537 |
"step": 355
|
| 538 |
},
|
| 539 |
{
|
| 540 |
+
"epoch": 1.0526315789473684,
|
| 541 |
+
"grad_norm": 2.311139734884461,
|
| 542 |
+
"learning_rate": 1.6411928294780626e-05,
|
| 543 |
+
"loss": 0.5252,
|
| 544 |
"step": 360
|
| 545 |
},
|
| 546 |
{
|
| 547 |
+
"epoch": 1.067251461988304,
|
| 548 |
+
"grad_norm": 1.886267380269756,
|
| 549 |
+
"learning_rate": 1.6280410495099165e-05,
|
| 550 |
+
"loss": 0.5028,
|
| 551 |
"step": 365
|
| 552 |
},
|
| 553 |
{
|
| 554 |
+
"epoch": 1.0818713450292399,
|
| 555 |
+
"grad_norm": 1.8701236760118256,
|
| 556 |
+
"learning_rate": 1.6147073774150834e-05,
|
| 557 |
+
"loss": 0.5121,
|
| 558 |
"step": 370
|
| 559 |
},
|
| 560 |
{
|
| 561 |
+
"epoch": 1.0964912280701755,
|
| 562 |
+
"grad_norm": 1.7743194110147134,
|
| 563 |
+
"learning_rate": 1.601195674867853e-05,
|
| 564 |
+
"loss": 0.5189,
|
| 565 |
"step": 375
|
| 566 |
},
|
| 567 |
{
|
| 568 |
+
"epoch": 1.1111111111111112,
|
| 569 |
+
"grad_norm": 36.55479744540479,
|
| 570 |
+
"learning_rate": 1.5875098551033765e-05,
|
| 571 |
+
"loss": 0.5071,
|
| 572 |
"step": 380
|
| 573 |
},
|
| 574 |
{
|
| 575 |
+
"epoch": 1.1257309941520468,
|
| 576 |
+
"grad_norm": 2.1840870340337823,
|
| 577 |
+
"learning_rate": 1.5736538817843228e-05,
|
| 578 |
+
"loss": 0.5489,
|
| 579 |
"step": 385
|
| 580 |
},
|
| 581 |
{
|
| 582 |
+
"epoch": 1.1403508771929824,
|
| 583 |
+
"grad_norm": 2.096047543627006,
|
| 584 |
+
"learning_rate": 1.559631767852929e-05,
|
| 585 |
+
"loss": 0.5115,
|
| 586 |
"step": 390
|
| 587 |
},
|
| 588 |
{
|
| 589 |
+
"epoch": 1.154970760233918,
|
| 590 |
+
"grad_norm": 8.700584160621043,
|
| 591 |
+
"learning_rate": 1.54544757436878e-05,
|
| 592 |
+
"loss": 0.5061,
|
| 593 |
"step": 395
|
| 594 |
},
|
| 595 |
{
|
| 596 |
+
"epoch": 1.1695906432748537,
|
| 597 |
+
"grad_norm": 1.7774756036197978,
|
| 598 |
+
"learning_rate": 1.5311054093326508e-05,
|
| 599 |
+
"loss": 0.5584,
|
| 600 |
"step": 400
|
| 601 |
},
|
| 602 |
{
|
| 603 |
+
"epoch": 1.1695906432748537,
|
| 604 |
+
"eval_loss": 0.9159147143363953,
|
| 605 |
+
"eval_runtime": 94.2366,
|
| 606 |
+
"eval_samples_per_second": 23.154,
|
| 607 |
+
"eval_steps_per_second": 0.371,
|
| 608 |
"step": 400
|
| 609 |
},
|
| 610 |
{
|
| 611 |
+
"epoch": 1.1842105263157894,
|
| 612 |
+
"grad_norm": 2.550427212943654,
|
| 613 |
+
"learning_rate": 1.5166094264967553e-05,
|
| 614 |
+
"loss": 0.5166,
|
| 615 |
"step": 405
|
| 616 |
},
|
| 617 |
{
|
| 618 |
+
"epoch": 1.198830409356725,
|
| 619 |
+
"grad_norm": 1.9907997528305226,
|
| 620 |
+
"learning_rate": 1.5019638241617429e-05,
|
| 621 |
+
"loss": 0.5131,
|
| 622 |
"step": 410
|
| 623 |
},
|
| 624 |
{
|
| 625 |
+
"epoch": 1.213450292397661,
|
| 626 |
+
"grad_norm": 1.7648223360572315,
|
| 627 |
+
"learning_rate": 1.4871728439607967e-05,
|
| 628 |
+
"loss": 0.5182,
|
| 629 |
"step": 415
|
| 630 |
},
|
| 631 |
{
|
| 632 |
+
"epoch": 1.2280701754385965,
|
| 633 |
+
"grad_norm": 2.23899567188947,
|
| 634 |
+
"learning_rate": 1.472240769631176e-05,
|
| 635 |
+
"loss": 0.5062,
|
| 636 |
"step": 420
|
| 637 |
},
|
| 638 |
{
|
| 639 |
+
"epoch": 1.2426900584795322,
|
| 640 |
+
"grad_norm": 1.763198949927548,
|
| 641 |
+
"learning_rate": 1.4571719257735702e-05,
|
| 642 |
+
"loss": 0.5203,
|
| 643 |
"step": 425
|
| 644 |
},
|
| 645 |
{
|
| 646 |
+
"epoch": 1.2573099415204678,
|
| 647 |
+
"grad_norm": 4.030641844892038,
|
| 648 |
+
"learning_rate": 1.4419706765996153e-05,
|
| 649 |
+
"loss": 0.5176,
|
| 650 |
"step": 430
|
| 651 |
},
|
| 652 |
{
|
| 653 |
+
"epoch": 1.2719298245614035,
|
| 654 |
+
"grad_norm": 1.8591284929042133,
|
| 655 |
+
"learning_rate": 1.4266414246679379e-05,
|
| 656 |
+
"loss": 0.5099,
|
| 657 |
"step": 435
|
| 658 |
},
|
| 659 |
{
|
| 660 |
+
"epoch": 1.286549707602339,
|
| 661 |
+
"grad_norm": 1.8559628967021533,
|
| 662 |
+
"learning_rate": 1.4111886096090953e-05,
|
| 663 |
+
"loss": 0.5129,
|
| 664 |
"step": 440
|
| 665 |
},
|
| 666 |
{
|
| 667 |
+
"epoch": 1.301169590643275,
|
| 668 |
+
"grad_norm": 2.862306979981545,
|
| 669 |
+
"learning_rate": 1.3956167068397756e-05,
|
| 670 |
+
"loss": 0.5075,
|
| 671 |
"step": 445
|
| 672 |
},
|
| 673 |
{
|
| 674 |
+
"epoch": 1.3157894736842106,
|
| 675 |
+
"grad_norm": 1.8928522544167012,
|
| 676 |
+
"learning_rate": 1.3799302262666388e-05,
|
| 677 |
+
"loss": 0.5177,
|
| 678 |
"step": 450
|
| 679 |
},
|
| 680 |
{
|
| 681 |
+
"epoch": 1.3304093567251463,
|
| 682 |
+
"grad_norm": 1.8170476120999561,
|
| 683 |
+
"learning_rate": 1.364133710980162e-05,
|
| 684 |
+
"loss": 0.516,
|
| 685 |
"step": 455
|
| 686 |
},
|
| 687 |
{
|
| 688 |
+
"epoch": 1.345029239766082,
|
| 689 |
+
"grad_norm": 1.735070100858528,
|
| 690 |
+
"learning_rate": 1.3482317359388806e-05,
|
| 691 |
+
"loss": 0.5086,
|
| 692 |
"step": 460
|
| 693 |
},
|
| 694 |
{
|
| 695 |
+
"epoch": 1.3596491228070176,
|
| 696 |
+
"grad_norm": 2.0140408558653737,
|
| 697 |
+
"learning_rate": 1.3322289066443947e-05,
|
| 698 |
+
"loss": 0.5169,
|
| 699 |
"step": 465
|
| 700 |
},
|
| 701 |
{
|
| 702 |
+
"epoch": 1.3742690058479532,
|
| 703 |
+
"grad_norm": 1.7311234684729015,
|
| 704 |
+
"learning_rate": 1.316129857807534e-05,
|
| 705 |
+
"loss": 0.5077,
|
| 706 |
"step": 470
|
| 707 |
},
|
| 708 |
{
|
| 709 |
+
"epoch": 1.3888888888888888,
|
| 710 |
+
"grad_norm": 1.7834637989309219,
|
| 711 |
+
"learning_rate": 1.2999392520060594e-05,
|
| 712 |
+
"loss": 0.5181,
|
| 713 |
"step": 475
|
| 714 |
},
|
| 715 |
{
|
| 716 |
+
"epoch": 1.4035087719298245,
|
| 717 |
+
"grad_norm": 1.725680916190487,
|
| 718 |
+
"learning_rate": 1.2836617783342968e-05,
|
| 719 |
+
"loss": 0.5102,
|
| 720 |
"step": 480
|
| 721 |
},
|
| 722 |
{
|
| 723 |
+
"epoch": 1.4181286549707601,
|
| 724 |
+
"grad_norm": 1.8726265727670828,
|
| 725 |
+
"learning_rate": 1.2673021510450893e-05,
|
| 726 |
+
"loss": 0.5169,
|
| 727 |
"step": 485
|
| 728 |
},
|
| 729 |
{
|
| 730 |
+
"epoch": 1.4327485380116958,
|
| 731 |
+
"grad_norm": 2.056653138512768,
|
| 732 |
+
"learning_rate": 1.250865108184464e-05,
|
| 733 |
+
"loss": 0.5085,
|
| 734 |
"step": 490
|
| 735 |
},
|
| 736 |
{
|
| 737 |
+
"epoch": 1.4473684210526316,
|
| 738 |
+
"grad_norm": 1.6627221599390933,
|
| 739 |
+
"learning_rate": 1.2343554102194073e-05,
|
| 740 |
+
"loss": 0.4981,
|
| 741 |
"step": 495
|
| 742 |
},
|
| 743 |
{
|
| 744 |
+
"epoch": 1.4619883040935673,
|
| 745 |
+
"grad_norm": 1.7476089665186805,
|
| 746 |
+
"learning_rate": 1.2177778386591475e-05,
|
| 747 |
+
"loss": 0.5186,
|
| 748 |
"step": 500
|
| 749 |
},
|
| 750 |
{
|
| 751 |
+
"epoch": 1.4619883040935673,
|
| 752 |
+
"eval_loss": 0.8920474648475647,
|
| 753 |
+
"eval_runtime": 93.9752,
|
| 754 |
+
"eval_samples_per_second": 23.219,
|
| 755 |
+
"eval_steps_per_second": 0.372,
|
| 756 |
"step": 500
|
| 757 |
},
|
| 758 |
{
|
| 759 |
+
"epoch": 1.476608187134503,
|
| 760 |
+
"grad_norm": 1.6946408040860308,
|
| 761 |
+
"learning_rate": 1.2011371946703416e-05,
|
| 762 |
+
"loss": 0.5197,
|
| 763 |
"step": 505
|
| 764 |
},
|
| 765 |
{
|
| 766 |
+
"epoch": 1.4912280701754386,
|
| 767 |
+
"grad_norm": 1.7569189927952173,
|
| 768 |
+
"learning_rate": 1.1844382976865714e-05,
|
| 769 |
+
"loss": 0.515,
|
| 770 |
"step": 510
|
| 771 |
},
|
| 772 |
{
|
| 773 |
+
"epoch": 1.5058479532163744,
|
| 774 |
+
"grad_norm": 1.9990864745103767,
|
| 775 |
+
"learning_rate": 1.1676859840125468e-05,
|
| 776 |
+
"loss": 0.5188,
|
| 777 |
"step": 515
|
| 778 |
},
|
| 779 |
{
|
| 780 |
+
"epoch": 1.52046783625731,
|
| 781 |
+
"grad_norm": 1.659671819522036,
|
| 782 |
+
"learning_rate": 1.1508851054234236e-05,
|
| 783 |
+
"loss": 0.4952,
|
| 784 |
"step": 520
|
| 785 |
},
|
| 786 |
{
|
| 787 |
+
"epoch": 1.5350877192982457,
|
| 788 |
+
"grad_norm": 1.8125623588878788,
|
| 789 |
+
"learning_rate": 1.1340405277596427e-05,
|
| 790 |
+
"loss": 0.5029,
|
| 791 |
"step": 525
|
| 792 |
},
|
| 793 |
{
|
| 794 |
+
"epoch": 1.5497076023391814,
|
| 795 |
+
"grad_norm": 1.6209359235184773,
|
| 796 |
+
"learning_rate": 1.1171571295176915e-05,
|
| 797 |
+
"loss": 0.5016,
|
| 798 |
"step": 530
|
| 799 |
},
|
| 800 |
{
|
| 801 |
+
"epoch": 1.564327485380117,
|
| 802 |
+
"grad_norm": 1.7381191339168205,
|
| 803 |
+
"learning_rate": 1.1002398004372048e-05,
|
| 804 |
+
"loss": 0.4938,
|
| 805 |
"step": 535
|
| 806 |
},
|
| 807 |
{
|
| 808 |
+
"epoch": 1.5789473684210527,
|
| 809 |
+
"grad_norm": 1.7440328988243268,
|
| 810 |
+
"learning_rate": 1.0832934400848063e-05,
|
| 811 |
+
"loss": 0.508,
|
| 812 |
"step": 540
|
| 813 |
},
|
| 814 |
{
|
| 815 |
+
"epoch": 1.5935672514619883,
|
| 816 |
+
"grad_norm": 1.6873580611700445,
|
| 817 |
+
"learning_rate": 1.066322956435104e-05,
|
| 818 |
+
"loss": 0.4994,
|
| 819 |
"step": 545
|
| 820 |
},
|
| 821 |
{
|
| 822 |
+
"epoch": 1.608187134502924,
|
| 823 |
+
"grad_norm": 1.6955737592784719,
|
| 824 |
+
"learning_rate": 1.0493332644492534e-05,
|
| 825 |
+
"loss": 0.494,
|
| 826 |
"step": 550
|
| 827 |
},
|
| 828 |
{
|
| 829 |
+
"epoch": 1.6228070175438596,
|
| 830 |
+
"grad_norm": 1.8639586363579819,
|
| 831 |
+
"learning_rate": 1.0323292846514927e-05,
|
| 832 |
+
"loss": 0.5086,
|
| 833 |
"step": 555
|
| 834 |
},
|
| 835 |
{
|
| 836 |
+
"epoch": 1.6374269005847952,
|
| 837 |
+
"grad_norm": 1.6602403945581443,
|
| 838 |
+
"learning_rate": 1.015315941704071e-05,
|
| 839 |
+
"loss": 0.4958,
|
| 840 |
"step": 560
|
| 841 |
},
|
| 842 |
{
|
| 843 |
+
"epoch": 1.6520467836257309,
|
| 844 |
+
"grad_norm": 1.6463228018665035,
|
| 845 |
+
"learning_rate": 9.982981629809776e-06,
|
| 846 |
+
"loss": 0.4947,
|
| 847 |
"step": 565
|
| 848 |
},
|
| 849 |
{
|
| 850 |
+
"epoch": 1.6666666666666665,
|
| 851 |
+
"grad_norm": 1.8439520823673021,
|
| 852 |
+
"learning_rate": 9.812808771408804e-06,
|
| 853 |
+
"loss": 0.4943,
|
| 854 |
"step": 570
|
| 855 |
},
|
| 856 |
{
|
| 857 |
+
"epoch": 1.6812865497076024,
|
| 858 |
+
"grad_norm": 1.6567753225002235,
|
| 859 |
+
"learning_rate": 9.64269012699702e-06,
|
| 860 |
+
"loss": 0.4815,
|
| 861 |
"step": 575
|
| 862 |
},
|
| 863 |
{
|
| 864 |
+
"epoch": 1.695906432748538,
|
| 865 |
+
"grad_norm": 1.6625638574779524,
|
| 866 |
+
"learning_rate": 9.472674966032276e-06,
|
| 867 |
+
"loss": 0.5006,
|
| 868 |
"step": 580
|
| 869 |
},
|
| 870 |
{
|
| 871 |
+
"epoch": 1.7105263157894737,
|
| 872 |
+
"grad_norm": 1.898719426059897,
|
| 873 |
+
"learning_rate": 9.302812528001741e-06,
|
| 874 |
+
"loss": 0.5086,
|
| 875 |
"step": 585
|
| 876 |
},
|
| 877 |
{
|
| 878 |
+
"epoch": 1.7251461988304093,
|
| 879 |
+
"grad_norm": 1.8795535435362318,
|
| 880 |
+
"learning_rate": 9.133152008161235e-06,
|
| 881 |
+
"loss": 0.5094,
|
| 882 |
"step": 590
|
| 883 |
},
|
| 884 |
{
|
| 885 |
+
"epoch": 1.7397660818713452,
|
| 886 |
+
"grad_norm": 1.7067878022572074,
|
| 887 |
+
"learning_rate": 8.963742543287397e-06,
|
| 888 |
+
"loss": 0.5001,
|
| 889 |
"step": 595
|
| 890 |
},
|
| 891 |
{
|
| 892 |
+
"epoch": 1.7543859649122808,
|
| 893 |
+
"grad_norm": 1.6684322823190847,
|
| 894 |
+
"learning_rate": 8.79463319744677e-06,
|
| 895 |
+
"loss": 0.5067,
|
| 896 |
"step": 600
|
| 897 |
},
|
| 898 |
{
|
| 899 |
+
"epoch": 1.7543859649122808,
|
| 900 |
+
"eval_loss": 0.8780078291893005,
|
| 901 |
+
"eval_runtime": 94.1156,
|
| 902 |
+
"eval_samples_per_second": 23.184,
|
| 903 |
+
"eval_steps_per_second": 0.372,
|
| 904 |
"step": 600
|
| 905 |
},
|
| 906 |
{
|
| 907 |
+
"epoch": 1.7690058479532165,
|
| 908 |
+
"grad_norm": 1.575226401438412,
|
| 909 |
+
"learning_rate": 8.625872947785968e-06,
|
| 910 |
+
"loss": 0.5163,
|
| 911 |
"step": 605
|
| 912 |
},
|
| 913 |
{
|
| 914 |
+
"epoch": 1.7836257309941521,
|
| 915 |
+
"grad_norm": 1.665839831059325,
|
| 916 |
+
"learning_rate": 8.457510670346976e-06,
|
| 917 |
+
"loss": 0.5185,
|
| 918 |
"step": 610
|
| 919 |
},
|
| 920 |
{
|
| 921 |
+
"epoch": 1.7982456140350878,
|
| 922 |
+
"grad_norm": 1.668476308962127,
|
| 923 |
+
"learning_rate": 8.2895951259118e-06,
|
| 924 |
+
"loss": 0.4891,
|
| 925 |
"step": 615
|
| 926 |
},
|
| 927 |
{
|
| 928 |
+
"epoch": 1.8128654970760234,
|
| 929 |
+
"grad_norm": 1.6099746797995087,
|
| 930 |
+
"learning_rate": 8.122174945880409e-06,
|
| 931 |
+
"loss": 0.4941,
|
| 932 |
"step": 620
|
| 933 |
},
|
| 934 |
{
|
| 935 |
+
"epoch": 1.827485380116959,
|
| 936 |
+
"grad_norm": 1.5972670521003787,
|
| 937 |
+
"learning_rate": 7.955298618186227e-06,
|
| 938 |
+
"loss": 0.4983,
|
| 939 |
"step": 625
|
| 940 |
},
|
| 941 |
{
|
| 942 |
+
"epoch": 1.8421052631578947,
|
| 943 |
+
"grad_norm": 1.6567002288888941,
|
| 944 |
+
"learning_rate": 7.78901447325314e-06,
|
| 945 |
+
"loss": 0.4923,
|
| 946 |
"step": 630
|
| 947 |
},
|
| 948 |
{
|
| 949 |
+
"epoch": 1.8567251461988303,
|
| 950 |
+
"grad_norm": 1.633625117649634,
|
| 951 |
+
"learning_rate": 7.623370669998115e-06,
|
| 952 |
+
"loss": 0.4952,
|
| 953 |
"step": 635
|
| 954 |
},
|
| 955 |
{
|
| 956 |
+
"epoch": 1.871345029239766,
|
| 957 |
+
"grad_norm": 1.6702231621007573,
|
| 958 |
+
"learning_rate": 7.458415181883506e-06,
|
| 959 |
+
"loss": 0.4781,
|
| 960 |
"step": 640
|
| 961 |
},
|
| 962 |
{
|
| 963 |
+
"epoch": 1.8859649122807016,
|
| 964 |
+
"grad_norm": 1.5695315130710974,
|
| 965 |
+
"learning_rate": 7.294195783023072e-06,
|
| 966 |
+
"loss": 0.4772,
|
| 967 |
"step": 645
|
| 968 |
},
|
| 969 |
{
|
| 970 |
+
"epoch": 1.9005847953216373,
|
| 971 |
+
"grad_norm": 1.5932464772328656,
|
| 972 |
+
"learning_rate": 7.130760034345708e-06,
|
| 973 |
+
"loss": 0.4766,
|
| 974 |
"step": 650
|
| 975 |
},
|
| 976 |
{
|
| 977 |
+
"epoch": 1.9152046783625731,
|
| 978 |
+
"grad_norm": 1.5938375889822713,
|
| 979 |
+
"learning_rate": 6.968155269820951e-06,
|
| 980 |
+
"loss": 0.4954,
|
| 981 |
"step": 655
|
| 982 |
},
|
| 983 |
{
|
| 984 |
+
"epoch": 1.9298245614035088,
|
| 985 |
+
"grad_norm": 1.6068162124707588,
|
| 986 |
+
"learning_rate": 6.806428582750191e-06,
|
| 987 |
+
"loss": 0.5031,
|
| 988 |
"step": 660
|
| 989 |
},
|
| 990 |
{
|
| 991 |
+
"epoch": 1.9444444444444444,
|
| 992 |
+
"grad_norm": 1.6369099221197159,
|
| 993 |
+
"learning_rate": 6.645626812127588e-06,
|
| 994 |
+
"loss": 0.4871,
|
| 995 |
"step": 665
|
| 996 |
},
|
| 997 |
{
|
| 998 |
+
"epoch": 1.95906432748538,
|
| 999 |
+
"grad_norm": 1.8488127966750827,
|
| 1000 |
+
"learning_rate": 6.485796529074662e-06,
|
| 1001 |
+
"loss": 0.4793,
|
| 1002 |
"step": 670
|
| 1003 |
},
|
| 1004 |
{
|
| 1005 |
+
"epoch": 1.973684210526316,
|
| 1006 |
+
"grad_norm": 1.5795572120040184,
|
| 1007 |
+
"learning_rate": 6.326984023352435e-06,
|
| 1008 |
+
"loss": 0.4728,
|
| 1009 |
"step": 675
|
| 1010 |
},
|
| 1011 |
{
|
| 1012 |
+
"epoch": 1.9883040935672516,
|
| 1013 |
+
"grad_norm": 1.8151786682783004,
|
| 1014 |
+
"learning_rate": 6.169235289955073e-06,
|
| 1015 |
+
"loss": 0.4659,
|
| 1016 |
"step": 680
|
| 1017 |
},
|
| 1018 |
{
|
| 1019 |
+
"epoch": 2.002923976608187,
|
| 1020 |
+
"grad_norm": 3.8300611957194977,
|
| 1021 |
+
"learning_rate": 6.012596015788903e-06,
|
| 1022 |
+
"loss": 0.4261,
|
| 1023 |
"step": 685
|
| 1024 |
},
|
| 1025 |
{
|
| 1026 |
+
"epoch": 2.017543859649123,
|
| 1027 |
+
"grad_norm": 2.621763538813387,
|
| 1028 |
+
"learning_rate": 5.8571115664406655e-06,
|
| 1029 |
+
"loss": 0.2426,
|
| 1030 |
"step": 690
|
| 1031 |
},
|
| 1032 |
{
|
| 1033 |
+
"epoch": 2.0321637426900585,
|
| 1034 |
+
"grad_norm": 1.8343232864531938,
|
| 1035 |
+
"learning_rate": 5.702826973038776e-06,
|
| 1036 |
+
"loss": 0.24,
|
| 1037 |
"step": 695
|
| 1038 |
},
|
| 1039 |
{
|
| 1040 |
+
"epoch": 2.046783625730994,
|
| 1041 |
+
"grad_norm": 1.9497619965124309,
|
| 1042 |
+
"learning_rate": 5.549786919211532e-06,
|
| 1043 |
+
"loss": 0.2223,
|
| 1044 |
"step": 700
|
| 1045 |
},
|
| 1046 |
{
|
| 1047 |
+
"epoch": 2.046783625730994,
|
| 1048 |
+
"eval_loss": 0.951133131980896,
|
| 1049 |
+
"eval_runtime": 93.9878,
|
| 1050 |
+
"eval_samples_per_second": 23.216,
|
| 1051 |
+
"eval_steps_per_second": 0.372,
|
| 1052 |
"step": 700
|
| 1053 |
},
|
| 1054 |
{
|
| 1055 |
+
"epoch": 2.06140350877193,
|
| 1056 |
+
"grad_norm": 1.7439743078429233,
|
| 1057 |
+
"learning_rate": 5.3980357281459e-06,
|
| 1058 |
+
"loss": 0.229,
|
| 1059 |
"step": 705
|
| 1060 |
},
|
| 1061 |
{
|
| 1062 |
+
"epoch": 2.0760233918128654,
|
| 1063 |
+
"grad_norm": 1.5354726993504848,
|
| 1064 |
+
"learning_rate": 5.247617349750717e-06,
|
| 1065 |
+
"loss": 0.2083,
|
| 1066 |
"step": 710
|
| 1067 |
},
|
| 1068 |
{
|
| 1069 |
+
"epoch": 2.090643274853801,
|
| 1070 |
+
"grad_norm": 7.788063462658685,
|
| 1071 |
+
"learning_rate": 5.0985753479279824e-06,
|
| 1072 |
+
"loss": 0.2391,
|
| 1073 |
"step": 715
|
| 1074 |
},
|
| 1075 |
{
|
| 1076 |
+
"epoch": 2.1052631578947367,
|
| 1077 |
+
"grad_norm": 1.5764797068116867,
|
| 1078 |
+
"learning_rate": 4.950952887955992e-06,
|
| 1079 |
+
"loss": 0.2146,
|
| 1080 |
"step": 720
|
| 1081 |
},
|
| 1082 |
{
|
| 1083 |
+
"epoch": 2.1198830409356724,
|
| 1084 |
+
"grad_norm": 1.5338239957853983,
|
| 1085 |
+
"learning_rate": 4.80479272398786e-06,
|
| 1086 |
+
"loss": 0.2145,
|
| 1087 |
"step": 725
|
| 1088 |
},
|
| 1089 |
{
|
| 1090 |
+
"epoch": 2.134502923976608,
|
| 1091 |
+
"grad_norm": 1.5400470674701994,
|
| 1092 |
+
"learning_rate": 4.660137186669131e-06,
|
| 1093 |
+
"loss": 0.2131,
|
| 1094 |
"step": 730
|
| 1095 |
},
|
| 1096 |
{
|
| 1097 |
+
"epoch": 2.1491228070175437,
|
| 1098 |
+
"grad_norm": 1.5891715585976705,
|
| 1099 |
+
"learning_rate": 4.5170281708780865e-06,
|
| 1100 |
+
"loss": 0.2273,
|
| 1101 |
"step": 735
|
| 1102 |
},
|
| 1103 |
{
|
| 1104 |
+
"epoch": 2.1637426900584797,
|
| 1105 |
+
"grad_norm": 1.6323196857228797,
|
| 1106 |
+
"learning_rate": 4.375507123592194e-06,
|
| 1107 |
+
"loss": 0.2221,
|
| 1108 |
"step": 740
|
| 1109 |
},
|
| 1110 |
{
|
| 1111 |
+
"epoch": 2.1783625730994154,
|
| 1112 |
+
"grad_norm": 3.4202645186771896,
|
| 1113 |
+
"learning_rate": 4.235615031884326e-06,
|
| 1114 |
+
"loss": 0.2139,
|
| 1115 |
"step": 745
|
| 1116 |
},
|
| 1117 |
{
|
| 1118 |
+
"epoch": 2.192982456140351,
|
| 1119 |
+
"grad_norm": 1.6425009664630226,
|
| 1120 |
+
"learning_rate": 4.097392411052149e-06,
|
| 1121 |
+
"loss": 0.2165,
|
| 1122 |
"step": 750
|
| 1123 |
},
|
| 1124 |
{
|
| 1125 |
+
"epoch": 2.2076023391812867,
|
| 1126 |
+
"grad_norm": 1.5960953283192305,
|
| 1127 |
+
"learning_rate": 3.9608792928841596e-06,
|
| 1128 |
+
"loss": 0.201,
|
| 1129 |
"step": 755
|
| 1130 |
},
|
| 1131 |
{
|
| 1132 |
+
"epoch": 2.2222222222222223,
|
| 1133 |
+
"grad_norm": 1.5527267754127971,
|
| 1134 |
+
"learning_rate": 3.826115214065739e-06,
|
| 1135 |
+
"loss": 0.2162,
|
| 1136 |
"step": 760
|
| 1137 |
},
|
| 1138 |
{
|
| 1139 |
+
"epoch": 2.236842105263158,
|
| 1140 |
+
"grad_norm": 1.5345330165293922,
|
| 1141 |
+
"learning_rate": 3.693139204728623e-06,
|
| 1142 |
+
"loss": 0.2033,
|
| 1143 |
"step": 765
|
| 1144 |
},
|
| 1145 |
{
|
| 1146 |
+
"epoch": 2.2514619883040936,
|
| 1147 |
+
"grad_norm": 1.5602344960819443,
|
| 1148 |
+
"learning_rate": 3.561989777147059e-06,
|
| 1149 |
+
"loss": 0.2175,
|
| 1150 |
"step": 770
|
| 1151 |
},
|
| 1152 |
{
|
| 1153 |
+
"epoch": 2.2660818713450293,
|
| 1154 |
+
"grad_norm": 1.543288744735399,
|
| 1155 |
+
"learning_rate": 3.4327049145839496e-06,
|
| 1156 |
+
"loss": 0.2044,
|
| 1157 |
"step": 775
|
| 1158 |
},
|
| 1159 |
{
|
| 1160 |
+
"epoch": 2.280701754385965,
|
| 1161 |
+
"grad_norm": 1.5668680143830462,
|
| 1162 |
+
"learning_rate": 3.3053220602902057e-06,
|
| 1163 |
+
"loss": 0.2067,
|
| 1164 |
"step": 780
|
| 1165 |
},
|
| 1166 |
{
|
| 1167 |
+
"epoch": 2.2953216374269005,
|
| 1168 |
+
"grad_norm": 1.4653721970803488,
|
| 1169 |
+
"learning_rate": 3.1798781066605076e-06,
|
| 1170 |
+
"loss": 0.2072,
|
| 1171 |
"step": 785
|
| 1172 |
},
|
| 1173 |
{
|
| 1174 |
+
"epoch": 2.309941520467836,
|
| 1175 |
+
"grad_norm": 1.5895314411422212,
|
| 1176 |
+
"learning_rate": 3.056409384548575e-06,
|
| 1177 |
+
"loss": 0.2077,
|
| 1178 |
"step": 790
|
| 1179 |
},
|
| 1180 |
{
|
| 1181 |
+
"epoch": 2.324561403508772,
|
| 1182 |
+
"grad_norm": 1.5324432023481178,
|
| 1183 |
+
"learning_rate": 2.934951652745123e-06,
|
| 1184 |
+
"loss": 0.2067,
|
| 1185 |
"step": 795
|
| 1186 |
},
|
| 1187 |
{
|
| 1188 |
+
"epoch": 2.3391812865497075,
|
| 1189 |
+
"grad_norm": 1.6877695775778203,
|
| 1190 |
+
"learning_rate": 2.8155400876214365e-06,
|
| 1191 |
+
"loss": 0.2114,
|
| 1192 |
"step": 800
|
| 1193 |
},
|
| 1194 |
{
|
| 1195 |
+
"epoch": 2.3391812865497075,
|
| 1196 |
+
"eval_loss": 0.9491869807243347,
|
| 1197 |
+
"eval_runtime": 94.0592,
|
| 1198 |
+
"eval_samples_per_second": 23.198,
|
| 1199 |
+
"eval_steps_per_second": 0.372,
|
| 1200 |
"step": 800
|
| 1201 |
},
|
| 1202 |
{
|
| 1203 |
+
"epoch": 2.353801169590643,
|
| 1204 |
+
"grad_norm": 1.797333972810049,
|
| 1205 |
+
"learning_rate": 2.698209272941659e-06,
|
| 1206 |
+
"loss": 0.2066,
|
| 1207 |
"step": 805
|
| 1208 |
},
|
| 1209 |
{
|
| 1210 |
+
"epoch": 2.3684210526315788,
|
| 1211 |
+
"grad_norm": 1.4817024828655516,
|
| 1212 |
+
"learning_rate": 2.5829931898467143e-06,
|
| 1213 |
+
"loss": 0.2003,
|
| 1214 |
"step": 810
|
| 1215 |
},
|
| 1216 |
{
|
| 1217 |
+
"epoch": 2.383040935672515,
|
| 1218 |
+
"grad_norm": 1.5944694035375513,
|
| 1219 |
+
"learning_rate": 2.469925207012741e-06,
|
| 1220 |
+
"loss": 0.2111,
|
| 1221 |
"step": 815
|
| 1222 |
},
|
| 1223 |
{
|
| 1224 |
+
"epoch": 2.39766081871345,
|
| 1225 |
+
"grad_norm": 1.5437718528241764,
|
| 1226 |
+
"learning_rate": 2.3590380709869175e-06,
|
| 1227 |
+
"loss": 0.2036,
|
| 1228 |
"step": 820
|
| 1229 |
},
|
| 1230 |
{
|
| 1231 |
+
"epoch": 2.412280701754386,
|
| 1232 |
+
"grad_norm": 1.5545155179428847,
|
| 1233 |
+
"learning_rate": 2.2503638967034668e-06,
|
| 1234 |
+
"loss": 0.2119,
|
| 1235 |
"step": 825
|
| 1236 |
},
|
| 1237 |
{
|
| 1238 |
+
"epoch": 2.426900584795322,
|
| 1239 |
+
"grad_norm": 1.5090412576695187,
|
| 1240 |
+
"learning_rate": 2.1439341581825855e-06,
|
| 1241 |
+
"loss": 0.2079,
|
| 1242 |
"step": 830
|
| 1243 |
},
|
| 1244 |
{
|
| 1245 |
+
"epoch": 2.4415204678362574,
|
| 1246 |
+
"grad_norm": 1.5651792272296057,
|
| 1247 |
+
"learning_rate": 2.039779679414996e-06,
|
| 1248 |
+
"loss": 0.2049,
|
| 1249 |
"step": 835
|
| 1250 |
},
|
| 1251 |
{
|
| 1252 |
+
"epoch": 2.456140350877193,
|
| 1253 |
+
"grad_norm": 1.629599771732464,
|
| 1254 |
+
"learning_rate": 1.9379306254347487e-06,
|
| 1255 |
+
"loss": 0.2063,
|
| 1256 |
"step": 840
|
| 1257 |
},
|
| 1258 |
{
|
| 1259 |
+
"epoch": 2.4707602339181287,
|
| 1260 |
+
"grad_norm": 1.517424552525727,
|
| 1261 |
+
"learning_rate": 1.838416493582893e-06,
|
| 1262 |
+
"loss": 0.2054,
|
| 1263 |
"step": 845
|
| 1264 |
},
|
| 1265 |
{
|
| 1266 |
+
"epoch": 2.4853801169590644,
|
| 1267 |
+
"grad_norm": 1.5665838306665794,
|
| 1268 |
+
"learning_rate": 1.7412661049645097e-06,
|
| 1269 |
+
"loss": 0.2072,
|
| 1270 |
"step": 850
|
| 1271 |
},
|
| 1272 |
{
|
| 1273 |
+
"epoch": 2.5,
|
| 1274 |
+
"grad_norm": 1.5709890486248266,
|
| 1275 |
+
"learning_rate": 1.6465075961015697e-06,
|
| 1276 |
+
"loss": 0.1969,
|
| 1277 |
"step": 855
|
| 1278 |
},
|
| 1279 |
{
|
| 1280 |
+
"epoch": 2.5146198830409356,
|
| 1281 |
+
"grad_norm": 1.433926747309705,
|
| 1282 |
+
"learning_rate": 1.554168410784117e-06,
|
| 1283 |
+
"loss": 0.2025,
|
| 1284 |
"step": 860
|
| 1285 |
},
|
| 1286 |
{
|
| 1287 |
+
"epoch": 2.5292397660818713,
|
| 1288 |
+
"grad_norm": 1.56024876676377,
|
| 1289 |
+
"learning_rate": 1.4642752921220272e-06,
|
| 1290 |
+
"loss": 0.2019,
|
| 1291 |
"step": 865
|
| 1292 |
},
|
| 1293 |
{
|
| 1294 |
+
"epoch": 2.543859649122807,
|
| 1295 |
+
"grad_norm": 1.4917168419123434,
|
| 1296 |
+
"learning_rate": 1.3768542747997215e-06,
|
| 1297 |
+
"loss": 0.2092,
|
| 1298 |
"step": 870
|
| 1299 |
},
|
| 1300 |
{
|
| 1301 |
+
"epoch": 2.5584795321637426,
|
| 1302 |
+
"grad_norm": 1.5176036989413515,
|
| 1303 |
+
"learning_rate": 1.2919306775360495e-06,
|
| 1304 |
+
"loss": 0.2053,
|
| 1305 |
"step": 875
|
| 1306 |
},
|
| 1307 |
{
|
| 1308 |
+
"epoch": 2.573099415204678,
|
| 1309 |
+
"grad_norm": 1.5727527454387282,
|
| 1310 |
+
"learning_rate": 1.209529095751527e-06,
|
| 1311 |
+
"loss": 0.2014,
|
| 1312 |
"step": 880
|
| 1313 |
},
|
| 1314 |
{
|
| 1315 |
+
"epoch": 2.587719298245614,
|
| 1316 |
+
"grad_norm": 1.489918411889695,
|
| 1317 |
+
"learning_rate": 1.1296733944450445e-06,
|
| 1318 |
+
"loss": 0.2012,
|
| 1319 |
"step": 885
|
| 1320 |
},
|
| 1321 |
{
|
| 1322 |
+
"epoch": 2.60233918128655,
|
| 1323 |
+
"grad_norm": 1.5532738891332698,
|
| 1324 |
+
"learning_rate": 1.0523867012821444e-06,
|
| 1325 |
+
"loss": 0.2031,
|
| 1326 |
"step": 890
|
| 1327 |
},
|
| 1328 |
{
|
| 1329 |
+
"epoch": 2.616959064327485,
|
| 1330 |
+
"grad_norm": 1.4483198492727338,
|
| 1331 |
+
"learning_rate": 9.776913998968196e-07,
|
| 1332 |
+
"loss": 0.2012,
|
| 1333 |
"step": 895
|
| 1334 |
},
|
| 1335 |
{
|
| 1336 |
+
"epoch": 2.6315789473684212,
|
| 1337 |
+
"grad_norm": 1.5331083693362297,
|
| 1338 |
+
"learning_rate": 9.056091234088038e-07,
|
| 1339 |
+
"loss": 0.1983,
|
| 1340 |
+
"step": 900
|
| 1341 |
+
},
|
| 1342 |
+
{
|
| 1343 |
+
"epoch": 2.6315789473684212,
|
| 1344 |
+
"eval_loss": 0.9508408904075623,
|
| 1345 |
+
"eval_runtime": 93.8904,
|
| 1346 |
+
"eval_samples_per_second": 23.24,
|
| 1347 |
+
"eval_steps_per_second": 0.373,
|
| 1348 |
+
"step": 900
|
| 1349 |
+
},
|
| 1350 |
+
{
|
| 1351 |
+
"epoch": 2.646198830409357,
|
| 1352 |
+
"grad_norm": 1.4959941578534954,
|
| 1353 |
+
"learning_rate": 8.361607481582312e-07,
|
| 1354 |
+
"loss": 0.2031,
|
| 1355 |
+
"step": 905
|
| 1356 |
+
},
|
| 1357 |
+
{
|
| 1358 |
+
"epoch": 2.6608187134502925,
|
| 1359 |
+
"grad_norm": 1.533846604598029,
|
| 1360 |
+
"learning_rate": 7.693663876594648e-07,
|
| 1361 |
+
"loss": 0.2086,
|
| 1362 |
+
"step": 910
|
| 1363 |
+
},
|
| 1364 |
+
{
|
| 1365 |
+
"epoch": 2.675438596491228,
|
| 1366 |
+
"grad_norm": 1.4882252878887594,
|
| 1367 |
+
"learning_rate": 7.052453867758525e-07,
|
| 1368 |
+
"loss": 0.2044,
|
| 1369 |
+
"step": 915
|
| 1370 |
+
},
|
| 1371 |
+
{
|
| 1372 |
+
"epoch": 2.690058479532164,
|
| 1373 |
+
"grad_norm": 1.463561383109587,
|
| 1374 |
+
"learning_rate": 6.438163161171096e-07,
|
| 1375 |
+
"loss": 0.199,
|
| 1376 |
+
"step": 920
|
| 1377 |
+
},
|
| 1378 |
+
{
|
| 1379 |
+
"epoch": 2.7046783625730995,
|
| 1380 |
+
"grad_norm": 1.4904086029449222,
|
| 1381 |
+
"learning_rate": 5.850969666609363e-07,
|
| 1382 |
+
"loss": 0.1949,
|
| 1383 |
+
"step": 925
|
| 1384 |
+
},
|
| 1385 |
+
{
|
| 1386 |
+
"epoch": 2.719298245614035,
|
| 1387 |
+
"grad_norm": 1.4484608706071034,
|
| 1388 |
+
"learning_rate": 5.291043446004074e-07,
|
| 1389 |
+
"loss": 0.1941,
|
| 1390 |
+
"step": 930
|
| 1391 |
+
},
|
| 1392 |
+
{
|
| 1393 |
+
"epoch": 2.7339181286549707,
|
| 1394 |
+
"grad_norm": 1.4693647586602272,
|
| 1395 |
+
"learning_rate": 4.7585466641868696e-07,
|
| 1396 |
+
"loss": 0.1925,
|
| 1397 |
+
"step": 935
|
| 1398 |
+
},
|
| 1399 |
+
{
|
| 1400 |
+
"epoch": 2.7485380116959064,
|
| 1401 |
+
"grad_norm": 1.4804496226260073,
|
| 1402 |
+
"learning_rate": 4.25363354192434e-07,
|
| 1403 |
+
"loss": 0.2027,
|
| 1404 |
+
"step": 940
|
| 1405 |
+
},
|
| 1406 |
+
{
|
| 1407 |
+
"epoch": 2.763157894736842,
|
| 1408 |
+
"grad_norm": 1.5177590267070082,
|
| 1409 |
+
"learning_rate": 3.776450311252866e-07,
|
| 1410 |
+
"loss": 0.2068,
|
| 1411 |
+
"step": 945
|
| 1412 |
+
},
|
| 1413 |
+
{
|
| 1414 |
+
"epoch": 2.7777777777777777,
|
| 1415 |
+
"grad_norm": 1.4962603716375011,
|
| 1416 |
+
"learning_rate": 3.3271351731271717e-07,
|
| 1417 |
+
"loss": 0.1993,
|
| 1418 |
+
"step": 950
|
| 1419 |
+
},
|
| 1420 |
+
{
|
| 1421 |
+
"epoch": 2.7923976608187133,
|
| 1422 |
+
"grad_norm": 1.53541672293609,
|
| 1423 |
+
"learning_rate": 2.905818257394799e-07,
|
| 1424 |
+
"loss": 0.2126,
|
| 1425 |
+
"step": 955
|
| 1426 |
+
},
|
| 1427 |
+
{
|
| 1428 |
+
"epoch": 2.807017543859649,
|
| 1429 |
+
"grad_norm": 1.4669550373356948,
|
| 1430 |
+
"learning_rate": 2.512621585108155e-07,
|
| 1431 |
+
"loss": 0.2021,
|
| 1432 |
+
"step": 960
|
| 1433 |
+
},
|
| 1434 |
+
{
|
| 1435 |
+
"epoch": 2.821637426900585,
|
| 1436 |
+
"grad_norm": 1.4475885818879743,
|
| 1437 |
+
"learning_rate": 2.1476590331849566e-07,
|
| 1438 |
+
"loss": 0.195,
|
| 1439 |
+
"step": 965
|
| 1440 |
+
},
|
| 1441 |
+
{
|
| 1442 |
+
"epoch": 2.8362573099415203,
|
| 1443 |
+
"grad_norm": 1.487795902053431,
|
| 1444 |
+
"learning_rate": 1.811036301427449e-07,
|
| 1445 |
+
"loss": 0.2073,
|
| 1446 |
+
"step": 970
|
| 1447 |
+
},
|
| 1448 |
+
{
|
| 1449 |
+
"epoch": 2.8508771929824563,
|
| 1450 |
+
"grad_norm": 1.4972431353793751,
|
| 1451 |
+
"learning_rate": 1.502850881909801e-07,
|
| 1452 |
+
"loss": 0.2017,
|
| 1453 |
+
"step": 975
|
| 1454 |
+
},
|
| 1455 |
+
{
|
| 1456 |
+
"epoch": 2.8654970760233915,
|
| 1457 |
+
"grad_norm": 1.506977096918749,
|
| 1458 |
+
"learning_rate": 1.2231920307425927e-07,
|
| 1459 |
+
"loss": 0.2081,
|
| 1460 |
+
"step": 980
|
| 1461 |
+
},
|
| 1462 |
+
{
|
| 1463 |
+
"epoch": 2.8801169590643276,
|
| 1464 |
+
"grad_norm": 1.5752046829905806,
|
| 1465 |
+
"learning_rate": 9.721407422226492e-08,
|
| 1466 |
+
"loss": 0.2089,
|
| 1467 |
+
"step": 985
|
| 1468 |
+
},
|
| 1469 |
+
{
|
| 1470 |
+
"epoch": 2.8947368421052633,
|
| 1471 |
+
"grad_norm": 1.4564645802512894,
|
| 1472 |
+
"learning_rate": 7.497697253756265e-08,
|
| 1473 |
+
"loss": 0.1933,
|
| 1474 |
+
"step": 990
|
| 1475 |
+
},
|
| 1476 |
+
{
|
| 1477 |
+
"epoch": 2.909356725146199,
|
| 1478 |
+
"grad_norm": 1.493498674859337,
|
| 1479 |
+
"learning_rate": 5.5614338289812216e-08,
|
| 1480 |
+
"loss": 0.1974,
|
| 1481 |
+
"step": 995
|
| 1482 |
+
},
|
| 1483 |
+
{
|
| 1484 |
+
"epoch": 2.9239766081871346,
|
| 1485 |
+
"grad_norm": 1.4762567685805037,
|
| 1486 |
+
"learning_rate": 3.913177925055189e-08,
|
| 1487 |
+
"loss": 0.2014,
|
| 1488 |
+
"step": 1000
|
| 1489 |
+
},
|
| 1490 |
+
{
|
| 1491 |
+
"epoch": 2.9239766081871346,
|
| 1492 |
+
"eval_loss": 0.946516215801239,
|
| 1493 |
+
"eval_runtime": 94.0428,
|
| 1494 |
+
"eval_samples_per_second": 23.202,
|
| 1495 |
+
"eval_steps_per_second": 0.372,
|
| 1496 |
+
"step": 1000
|
| 1497 |
+
},
|
| 1498 |
+
{
|
| 1499 |
+
"epoch": 2.93859649122807,
|
| 1500 |
+
"grad_norm": 1.47647651184245,
|
| 1501 |
+
"learning_rate": 2.5534069069081957e-08,
|
| 1502 |
+
"loss": 0.2044,
|
| 1503 |
+
"step": 1005
|
| 1504 |
+
},
|
| 1505 |
+
{
|
| 1506 |
+
"epoch": 2.953216374269006,
|
| 1507 |
+
"grad_norm": 1.4575732503367402,
|
| 1508 |
+
"learning_rate": 1.482514588993067e-08,
|
| 1509 |
+
"loss": 0.2001,
|
| 1510 |
+
"step": 1010
|
| 1511 |
+
},
|
| 1512 |
+
{
|
| 1513 |
+
"epoch": 2.9678362573099415,
|
| 1514 |
+
"grad_norm": 1.5453006830413416,
|
| 1515 |
+
"learning_rate": 7.0081112122966086e-09,
|
| 1516 |
+
"loss": 0.2042,
|
| 1517 |
+
"step": 1015
|
| 1518 |
+
},
|
| 1519 |
+
{
|
| 1520 |
+
"epoch": 2.982456140350877,
|
| 1521 |
+
"grad_norm": 1.7014593329310401,
|
| 1522 |
+
"learning_rate": 2.0852289917971947e-09,
|
| 1523 |
+
"loss": 0.1995,
|
| 1524 |
+
"step": 1020
|
| 1525 |
+
},
|
| 1526 |
+
{
|
| 1527 |
+
"epoch": 2.997076023391813,
|
| 1528 |
+
"grad_norm": 1.5593115324810627,
|
| 1529 |
+
"learning_rate": 5.792498478651709e-11,
|
| 1530 |
+
"loss": 0.2055,
|
| 1531 |
+
"step": 1025
|
| 1532 |
+
},
|
| 1533 |
+
{
|
| 1534 |
+
"epoch": 3.0,
|
| 1535 |
+
"step": 1026,
|
| 1536 |
+
"total_flos": 429437673799680.0,
|
| 1537 |
+
"train_loss": 0.5057662985420134,
|
| 1538 |
+
"train_runtime": 26805.8841,
|
| 1539 |
+
"train_samples_per_second": 4.896,
|
| 1540 |
+
"train_steps_per_second": 0.038
|
| 1541 |
}
|
| 1542 |
],
|
| 1543 |
"logging_steps": 5,
|
| 1544 |
+
"max_steps": 1026,
|
| 1545 |
"num_input_tokens_seen": 0,
|
| 1546 |
"num_train_epochs": 3,
|
| 1547 |
"save_steps": 100,
|
|
|
|
| 1557 |
"attributes": {}
|
| 1558 |
}
|
| 1559 |
},
|
| 1560 |
+
"total_flos": 429437673799680.0,
|
| 1561 |
+
"train_batch_size": 8,
|
| 1562 |
"trial_name": null,
|
| 1563 |
"trial_params": null
|
| 1564 |
}
|