Model save
Browse files- README.md +4 -6
- all_results.json +4 -9
- train_results.json +4 -4
- trainer_state.json +912 -928
README.md
CHANGED
|
@@ -1,16 +1,14 @@
|
|
| 1 |
---
|
| 2 |
-
datasets: HuggingFaceH4/Bespoke-Stratos-17k
|
| 3 |
library_name: transformers
|
| 4 |
model_name: Qwen2.5-1.5B-Open-R1-Distill
|
| 5 |
tags:
|
| 6 |
- generated_from_trainer
|
| 7 |
-
- open-r1
|
| 8 |
licence: license
|
| 9 |
---
|
| 10 |
|
| 11 |
# Model Card for Qwen2.5-1.5B-Open-R1-Distill
|
| 12 |
|
| 13 |
-
This model is a fine-tuned version of [None](https://huggingface.co/None)
|
| 14 |
It has been trained using [TRL](https://github.com/huggingface/trl).
|
| 15 |
|
| 16 |
## Quick start
|
|
@@ -26,7 +24,7 @@ print(output["generated_text"])
|
|
| 26 |
|
| 27 |
## Training procedure
|
| 28 |
|
| 29 |
-
[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="150" height="24"/>](https://wandb.ai/2741919970-hustvl/huggingface/runs/
|
| 30 |
|
| 31 |
|
| 32 |
This model was trained with SFT.
|
|
@@ -35,8 +33,8 @@ This model was trained with SFT.
|
|
| 35 |
|
| 36 |
- TRL: 0.16.0.dev0
|
| 37 |
- Transformers: 4.50.0.dev0
|
| 38 |
-
- Pytorch: 2.
|
| 39 |
-
- Datasets: 3.3.
|
| 40 |
- Tokenizers: 0.21.0
|
| 41 |
|
| 42 |
## Citations
|
|
|
|
| 1 |
---
|
|
|
|
| 2 |
library_name: transformers
|
| 3 |
model_name: Qwen2.5-1.5B-Open-R1-Distill
|
| 4 |
tags:
|
| 5 |
- generated_from_trainer
|
|
|
|
| 6 |
licence: license
|
| 7 |
---
|
| 8 |
|
| 9 |
# Model Card for Qwen2.5-1.5B-Open-R1-Distill
|
| 10 |
|
| 11 |
+
This model is a fine-tuned version of [None](https://huggingface.co/None).
|
| 12 |
It has been trained using [TRL](https://github.com/huggingface/trl).
|
| 13 |
|
| 14 |
## Quick start
|
|
|
|
| 24 |
|
| 25 |
## Training procedure
|
| 26 |
|
| 27 |
+
[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="150" height="24"/>](https://wandb.ai/2741919970-hustvl/huggingface/runs/3jcwqhpk)
|
| 28 |
|
| 29 |
|
| 30 |
This model was trained with SFT.
|
|
|
|
| 33 |
|
| 34 |
- TRL: 0.16.0.dev0
|
| 35 |
- Transformers: 4.50.0.dev0
|
| 36 |
+
- Pytorch: 2.6.0
|
| 37 |
+
- Datasets: 3.3.2
|
| 38 |
- Tokenizers: 0.21.0
|
| 39 |
|
| 40 |
## Citations
|
all_results.json
CHANGED
|
@@ -1,13 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"eval_loss": 0.9801562428474426,
|
| 3 |
-
"eval_runtime": 8.4113,
|
| 4 |
-
"eval_samples": 100,
|
| 5 |
-
"eval_samples_per_second": 11.889,
|
| 6 |
-
"eval_steps_per_second": 1.546,
|
| 7 |
"total_flos": 0.0,
|
| 8 |
-
"train_loss":
|
| 9 |
-
"train_runtime":
|
| 10 |
"train_samples": 16610,
|
| 11 |
-
"train_samples_per_second":
|
| 12 |
-
"train_steps_per_second": 0.
|
| 13 |
}
|
|
|
|
| 1 |
{
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 2 |
"total_flos": 0.0,
|
| 3 |
+
"train_loss": 0.9946218774200528,
|
| 4 |
+
"train_runtime": 21041.6427,
|
| 5 |
"train_samples": 16610,
|
| 6 |
+
"train_samples_per_second": 1.566,
|
| 7 |
+
"train_steps_per_second": 0.049
|
| 8 |
}
|
train_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
"total_flos": 0.0,
|
| 3 |
-
"train_loss":
|
| 4 |
-
"train_runtime":
|
| 5 |
"train_samples": 16610,
|
| 6 |
-
"train_samples_per_second":
|
| 7 |
-
"train_steps_per_second": 0.
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"total_flos": 0.0,
|
| 3 |
+
"train_loss": 0.9946218774200528,
|
| 4 |
+
"train_runtime": 21041.6427,
|
| 5 |
"train_samples": 16610,
|
| 6 |
+
"train_samples_per_second": 1.566,
|
| 7 |
+
"train_steps_per_second": 0.049
|
| 8 |
}
|
trainer_state.json
CHANGED
|
@@ -1,1774 +1,1758 @@
|
|
| 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": 6.
|
| 15 |
-
"loss":
|
| 16 |
-
"mean_token_accuracy": 0.
|
| 17 |
"step": 5
|
| 18 |
},
|
| 19 |
{
|
| 20 |
-
"epoch": 0.
|
| 21 |
-
"grad_norm":
|
| 22 |
-
"learning_rate": 9.
|
| 23 |
-
"loss":
|
| 24 |
-
"mean_token_accuracy": 0.
|
| 25 |
"step": 10
|
| 26 |
},
|
| 27 |
{
|
| 28 |
-
"epoch": 0.
|
| 29 |
-
"grad_norm":
|
| 30 |
-
"learning_rate": 1.
|
| 31 |
-
"loss": 1.
|
| 32 |
-
"mean_token_accuracy": 0.
|
| 33 |
"step": 15
|
| 34 |
},
|
| 35 |
{
|
| 36 |
-
"epoch": 0.
|
| 37 |
-
"grad_norm": 0.
|
| 38 |
-
"learning_rate": 1.
|
| 39 |
-
"loss": 1.
|
| 40 |
-
"mean_token_accuracy": 0.
|
| 41 |
"step": 20
|
| 42 |
},
|
| 43 |
{
|
| 44 |
-
"epoch": 0.
|
| 45 |
-
"grad_norm": 0.
|
| 46 |
-
"learning_rate": 1.
|
| 47 |
-
"loss": 1.
|
| 48 |
-
"mean_token_accuracy": 0.
|
| 49 |
"step": 25
|
| 50 |
},
|
| 51 |
{
|
| 52 |
-
"epoch": 0.
|
| 53 |
-
"grad_norm": 0.
|
| 54 |
-
"learning_rate": 1.
|
| 55 |
-
"loss": 1.
|
| 56 |
-
"mean_token_accuracy": 0.
|
| 57 |
"step": 30
|
| 58 |
},
|
| 59 |
{
|
| 60 |
-
"epoch": 0.
|
| 61 |
-
"grad_norm": 0.
|
| 62 |
-
"learning_rate": 1.
|
| 63 |
-
"loss": 1.
|
| 64 |
-
"mean_token_accuracy": 0.
|
| 65 |
"step": 35
|
| 66 |
},
|
| 67 |
{
|
| 68 |
-
"epoch": 0.
|
| 69 |
-
"grad_norm": 0.
|
| 70 |
-
"learning_rate": 1.
|
| 71 |
-
"loss": 1.
|
| 72 |
-
"mean_token_accuracy": 0.
|
| 73 |
"step": 40
|
| 74 |
},
|
| 75 |
{
|
| 76 |
-
"epoch": 0.
|
| 77 |
-
"grad_norm": 0.
|
| 78 |
-
"learning_rate": 1.
|
| 79 |
-
"loss": 1.
|
| 80 |
-
"mean_token_accuracy": 0.
|
| 81 |
"step": 45
|
| 82 |
},
|
| 83 |
{
|
| 84 |
-
"epoch": 0.
|
| 85 |
-
"grad_norm": 0.
|
| 86 |
-
"learning_rate": 1.
|
| 87 |
-
"loss": 1.
|
| 88 |
-
"mean_token_accuracy": 0.
|
| 89 |
"step": 50
|
| 90 |
},
|
| 91 |
{
|
| 92 |
-
"epoch": 0.
|
| 93 |
-
"grad_norm": 0.
|
| 94 |
-
"learning_rate": 1.
|
| 95 |
-
"loss": 1.
|
| 96 |
-
"mean_token_accuracy": 0.
|
| 97 |
"step": 55
|
| 98 |
},
|
| 99 |
{
|
| 100 |
-
"epoch": 0.
|
| 101 |
-
"grad_norm": 0.
|
| 102 |
-
"learning_rate": 1.
|
| 103 |
-
"loss": 1.
|
| 104 |
-
"mean_token_accuracy": 0.
|
| 105 |
"step": 60
|
| 106 |
},
|
| 107 |
{
|
| 108 |
-
"epoch": 0.
|
| 109 |
-
"grad_norm": 0.
|
| 110 |
-
"learning_rate": 1.
|
| 111 |
-
"loss": 1.
|
| 112 |
-
"mean_token_accuracy": 0.
|
| 113 |
"step": 65
|
| 114 |
},
|
| 115 |
{
|
| 116 |
-
"epoch": 0.
|
| 117 |
-
"grad_norm": 0.
|
| 118 |
-
"learning_rate": 1.
|
| 119 |
-
"loss": 1.
|
| 120 |
-
"mean_token_accuracy": 0.
|
| 121 |
"step": 70
|
| 122 |
},
|
| 123 |
{
|
| 124 |
-
"epoch": 0.
|
| 125 |
-
"grad_norm": 0.
|
| 126 |
-
"learning_rate": 1.
|
| 127 |
-
"loss": 1.
|
| 128 |
-
"mean_token_accuracy": 0.
|
| 129 |
"step": 75
|
| 130 |
},
|
| 131 |
{
|
| 132 |
-
"epoch": 0.
|
| 133 |
-
"grad_norm": 0.
|
| 134 |
-
"learning_rate": 1.
|
| 135 |
-
"loss": 1.
|
| 136 |
-
"mean_token_accuracy": 0.
|
| 137 |
"step": 80
|
| 138 |
},
|
| 139 |
{
|
| 140 |
-
"epoch": 0.
|
| 141 |
-
"grad_norm": 0.
|
| 142 |
-
"learning_rate": 1.
|
| 143 |
-
"loss": 1.
|
| 144 |
-
"mean_token_accuracy": 0.
|
| 145 |
"step": 85
|
| 146 |
},
|
| 147 |
{
|
| 148 |
-
"epoch": 0.
|
| 149 |
-
"grad_norm": 0.
|
| 150 |
-
"learning_rate": 1.
|
| 151 |
-
"loss": 1.
|
| 152 |
-
"mean_token_accuracy": 0.
|
| 153 |
"step": 90
|
| 154 |
},
|
| 155 |
{
|
| 156 |
-
"epoch": 0.
|
| 157 |
-
"grad_norm": 0.
|
| 158 |
-
"learning_rate": 1.
|
| 159 |
-
"loss": 1.
|
| 160 |
-
"mean_token_accuracy": 0.
|
| 161 |
"step": 95
|
| 162 |
},
|
| 163 |
{
|
| 164 |
-
"epoch": 0.
|
| 165 |
-
"grad_norm": 0.
|
| 166 |
-
"learning_rate": 1.
|
| 167 |
-
"loss": 1.
|
| 168 |
-
"mean_token_accuracy": 0.
|
| 169 |
"step": 100
|
| 170 |
},
|
| 171 |
{
|
| 172 |
-
"epoch": 0.
|
| 173 |
-
"eval_loss": 1.
|
| 174 |
-
"eval_mean_token_accuracy": 0.
|
| 175 |
-
"eval_runtime":
|
| 176 |
-
"eval_samples_per_second":
|
| 177 |
-
"eval_steps_per_second":
|
| 178 |
"step": 100
|
| 179 |
},
|
| 180 |
{
|
| 181 |
-
"epoch": 0.
|
| 182 |
-
"grad_norm": 0.
|
| 183 |
"learning_rate": 2e-05,
|
| 184 |
-
"loss": 1.
|
| 185 |
-
"mean_token_accuracy": 0.
|
| 186 |
"step": 105
|
| 187 |
},
|
| 188 |
{
|
| 189 |
-
"epoch": 0.
|
| 190 |
-
"grad_norm": 0.
|
| 191 |
"learning_rate": 2e-05,
|
| 192 |
-
"loss": 1.
|
| 193 |
-
"mean_token_accuracy": 0.
|
| 194 |
"step": 110
|
| 195 |
},
|
| 196 |
{
|
| 197 |
-
"epoch": 0.
|
| 198 |
-
"grad_norm": 0.
|
| 199 |
"learning_rate": 2e-05,
|
| 200 |
-
"loss": 1.
|
| 201 |
-
"mean_token_accuracy": 0.
|
| 202 |
"step": 115
|
| 203 |
},
|
| 204 |
{
|
| 205 |
-
"epoch": 0.
|
| 206 |
-
"grad_norm": 0.
|
| 207 |
"learning_rate": 2e-05,
|
| 208 |
-
"loss": 1.
|
| 209 |
-
"mean_token_accuracy": 0.
|
| 210 |
"step": 120
|
| 211 |
},
|
| 212 |
{
|
| 213 |
-
"epoch": 0.
|
| 214 |
-
"grad_norm": 0.
|
| 215 |
"learning_rate": 2e-05,
|
| 216 |
-
"loss": 1.
|
| 217 |
-
"mean_token_accuracy": 0.
|
| 218 |
"step": 125
|
| 219 |
},
|
| 220 |
{
|
| 221 |
-
"epoch": 0.
|
| 222 |
-
"grad_norm": 0.
|
| 223 |
"learning_rate": 2e-05,
|
| 224 |
-
"loss": 1.
|
| 225 |
-
"mean_token_accuracy": 0.
|
| 226 |
"step": 130
|
| 227 |
},
|
| 228 |
{
|
| 229 |
-
"epoch": 0.
|
| 230 |
-
"grad_norm": 0.
|
| 231 |
"learning_rate": 2e-05,
|
| 232 |
-
"loss": 1.
|
| 233 |
-
"mean_token_accuracy": 0.
|
| 234 |
"step": 135
|
| 235 |
},
|
| 236 |
{
|
| 237 |
-
"epoch": 0.
|
| 238 |
-
"grad_norm": 0.
|
| 239 |
"learning_rate": 2e-05,
|
| 240 |
-
"loss": 1.
|
| 241 |
-
"mean_token_accuracy": 0.
|
| 242 |
"step": 140
|
| 243 |
},
|
| 244 |
{
|
| 245 |
-
"epoch": 0.
|
| 246 |
-
"grad_norm": 0.
|
| 247 |
"learning_rate": 2e-05,
|
| 248 |
-
"loss": 1.
|
| 249 |
-
"mean_token_accuracy": 0.
|
| 250 |
"step": 145
|
| 251 |
},
|
| 252 |
{
|
| 253 |
-
"epoch": 0.
|
| 254 |
-
"grad_norm": 0.
|
| 255 |
"learning_rate": 2e-05,
|
| 256 |
-
"loss": 1.
|
| 257 |
-
"mean_token_accuracy": 0.
|
| 258 |
"step": 150
|
| 259 |
},
|
| 260 |
{
|
| 261 |
-
"epoch": 0.
|
| 262 |
-
"grad_norm": 0.
|
| 263 |
"learning_rate": 2e-05,
|
| 264 |
-
"loss": 1.
|
| 265 |
-
"mean_token_accuracy": 0.
|
| 266 |
"step": 155
|
| 267 |
},
|
| 268 |
{
|
| 269 |
-
"epoch": 0.
|
| 270 |
-
"grad_norm": 0.
|
| 271 |
"learning_rate": 2e-05,
|
| 272 |
-
"loss": 1.
|
| 273 |
-
"mean_token_accuracy": 0.
|
| 274 |
"step": 160
|
| 275 |
},
|
| 276 |
{
|
| 277 |
-
"epoch": 0.
|
| 278 |
-
"grad_norm": 0.
|
| 279 |
"learning_rate": 2e-05,
|
| 280 |
-
"loss": 1.
|
| 281 |
-
"mean_token_accuracy": 0.
|
| 282 |
"step": 165
|
| 283 |
},
|
| 284 |
{
|
| 285 |
-
"epoch": 0.
|
| 286 |
-
"grad_norm": 0.
|
| 287 |
"learning_rate": 2e-05,
|
| 288 |
-
"loss": 1.
|
| 289 |
-
"mean_token_accuracy": 0.
|
| 290 |
"step": 170
|
| 291 |
},
|
| 292 |
{
|
| 293 |
-
"epoch": 0.
|
| 294 |
-
"grad_norm": 0.
|
| 295 |
"learning_rate": 2e-05,
|
| 296 |
-
"loss": 1.
|
| 297 |
-
"mean_token_accuracy": 0.
|
| 298 |
"step": 175
|
| 299 |
},
|
| 300 |
{
|
| 301 |
-
"epoch": 0.
|
| 302 |
-
"grad_norm": 0.
|
| 303 |
"learning_rate": 2e-05,
|
| 304 |
-
"loss": 1.
|
| 305 |
-
"mean_token_accuracy": 0.
|
| 306 |
"step": 180
|
| 307 |
},
|
| 308 |
{
|
| 309 |
-
"epoch": 0.
|
| 310 |
-
"grad_norm": 0.
|
| 311 |
"learning_rate": 2e-05,
|
| 312 |
-
"loss": 1.
|
| 313 |
-
"mean_token_accuracy": 0.
|
| 314 |
"step": 185
|
| 315 |
},
|
| 316 |
{
|
| 317 |
-
"epoch": 0.
|
| 318 |
-
"grad_norm": 0.
|
| 319 |
"learning_rate": 2e-05,
|
| 320 |
-
"loss": 1.
|
| 321 |
-
"mean_token_accuracy": 0.
|
| 322 |
"step": 190
|
| 323 |
},
|
| 324 |
{
|
| 325 |
-
"epoch": 0.
|
| 326 |
-
"grad_norm": 0.
|
| 327 |
"learning_rate": 2e-05,
|
| 328 |
-
"loss": 1.
|
| 329 |
-
"mean_token_accuracy": 0.
|
| 330 |
"step": 195
|
| 331 |
},
|
| 332 |
{
|
| 333 |
-
"epoch": 0.
|
| 334 |
-
"grad_norm": 0.
|
| 335 |
"learning_rate": 2e-05,
|
| 336 |
-
"loss": 1.
|
| 337 |
-
"mean_token_accuracy": 0.
|
| 338 |
"step": 200
|
| 339 |
},
|
| 340 |
{
|
| 341 |
-
"epoch": 0.
|
| 342 |
-
"eval_loss": 1.
|
| 343 |
-
"eval_mean_token_accuracy": 0.
|
| 344 |
-
"eval_runtime":
|
| 345 |
-
"eval_samples_per_second":
|
| 346 |
-
"eval_steps_per_second":
|
| 347 |
"step": 200
|
| 348 |
},
|
| 349 |
{
|
| 350 |
-
"epoch": 0.
|
| 351 |
-
"grad_norm": 0.
|
| 352 |
"learning_rate": 2e-05,
|
| 353 |
-
"loss": 1.
|
| 354 |
-
"mean_token_accuracy": 0.
|
| 355 |
"step": 205
|
| 356 |
},
|
| 357 |
{
|
| 358 |
-
"epoch": 0.
|
| 359 |
-
"grad_norm": 0.
|
| 360 |
"learning_rate": 2e-05,
|
| 361 |
-
"loss": 1.
|
| 362 |
-
"mean_token_accuracy": 0.
|
| 363 |
"step": 210
|
| 364 |
},
|
| 365 |
{
|
| 366 |
-
"epoch": 0.
|
| 367 |
-
"grad_norm": 0.
|
| 368 |
"learning_rate": 2e-05,
|
| 369 |
-
"loss": 1.
|
| 370 |
-
"mean_token_accuracy": 0.
|
| 371 |
"step": 215
|
| 372 |
},
|
| 373 |
{
|
| 374 |
-
"epoch": 0.
|
| 375 |
-
"grad_norm": 0.
|
| 376 |
"learning_rate": 2e-05,
|
| 377 |
-
"loss": 1.
|
| 378 |
-
"mean_token_accuracy": 0.
|
| 379 |
"step": 220
|
| 380 |
},
|
| 381 |
{
|
| 382 |
-
"epoch": 0.
|
| 383 |
-
"grad_norm": 0.
|
| 384 |
"learning_rate": 2e-05,
|
| 385 |
-
"loss": 1.
|
| 386 |
-
"mean_token_accuracy": 0.
|
| 387 |
"step": 225
|
| 388 |
},
|
| 389 |
{
|
| 390 |
-
"epoch": 0.
|
| 391 |
-
"grad_norm": 0.
|
| 392 |
"learning_rate": 2e-05,
|
| 393 |
-
"loss": 1.
|
| 394 |
-
"mean_token_accuracy": 0.
|
| 395 |
"step": 230
|
| 396 |
},
|
| 397 |
{
|
| 398 |
-
"epoch": 0.
|
| 399 |
-
"grad_norm": 0.
|
| 400 |
"learning_rate": 2e-05,
|
| 401 |
-
"loss": 1.
|
| 402 |
-
"mean_token_accuracy": 0.
|
| 403 |
"step": 235
|
| 404 |
},
|
| 405 |
{
|
| 406 |
-
"epoch": 0.
|
| 407 |
-
"grad_norm": 0.
|
| 408 |
"learning_rate": 2e-05,
|
| 409 |
-
"loss": 1.
|
| 410 |
-
"mean_token_accuracy": 0.
|
| 411 |
"step": 240
|
| 412 |
},
|
| 413 |
{
|
| 414 |
-
"epoch": 0.
|
| 415 |
-
"grad_norm": 0.
|
| 416 |
"learning_rate": 2e-05,
|
| 417 |
-
"loss": 1.
|
| 418 |
-
"mean_token_accuracy": 0.
|
| 419 |
"step": 245
|
| 420 |
},
|
| 421 |
{
|
| 422 |
-
"epoch": 0.
|
| 423 |
-
"grad_norm": 0.
|
| 424 |
"learning_rate": 2e-05,
|
| 425 |
-
"loss": 1.
|
| 426 |
-
"mean_token_accuracy": 0.
|
| 427 |
"step": 250
|
| 428 |
},
|
| 429 |
{
|
| 430 |
-
"epoch": 0.
|
| 431 |
-
"grad_norm": 0.
|
| 432 |
"learning_rate": 2e-05,
|
| 433 |
-
"loss": 1.
|
| 434 |
-
"mean_token_accuracy": 0.
|
| 435 |
"step": 255
|
| 436 |
},
|
| 437 |
{
|
| 438 |
-
"epoch": 0.
|
| 439 |
-
"grad_norm": 0.
|
| 440 |
"learning_rate": 2e-05,
|
| 441 |
-
"loss": 1.
|
| 442 |
-
"mean_token_accuracy": 0.
|
| 443 |
"step": 260
|
| 444 |
},
|
| 445 |
{
|
| 446 |
-
"epoch": 0.
|
| 447 |
-
"grad_norm": 0.
|
| 448 |
"learning_rate": 2e-05,
|
| 449 |
-
"loss": 1.
|
| 450 |
-
"mean_token_accuracy": 0.
|
| 451 |
"step": 265
|
| 452 |
},
|
| 453 |
{
|
| 454 |
-
"epoch": 0.
|
| 455 |
-
"grad_norm": 0.
|
| 456 |
"learning_rate": 2e-05,
|
| 457 |
-
"loss": 1.
|
| 458 |
-
"mean_token_accuracy": 0.
|
| 459 |
"step": 270
|
| 460 |
},
|
| 461 |
{
|
| 462 |
-
"epoch": 0.
|
| 463 |
-
"grad_norm": 0.
|
| 464 |
"learning_rate": 2e-05,
|
| 465 |
-
"loss": 1.
|
| 466 |
-
"mean_token_accuracy": 0.
|
| 467 |
"step": 275
|
| 468 |
},
|
| 469 |
{
|
| 470 |
-
"epoch": 0.
|
| 471 |
-
"grad_norm": 0.
|
| 472 |
"learning_rate": 2e-05,
|
| 473 |
-
"loss": 1.
|
| 474 |
-
"mean_token_accuracy": 0.
|
| 475 |
"step": 280
|
| 476 |
},
|
| 477 |
{
|
| 478 |
-
"epoch": 0.
|
| 479 |
-
"grad_norm": 0.
|
| 480 |
"learning_rate": 2e-05,
|
| 481 |
-
"loss": 1.
|
| 482 |
-
"mean_token_accuracy": 0.
|
| 483 |
"step": 285
|
| 484 |
},
|
| 485 |
{
|
| 486 |
-
"epoch": 0.
|
| 487 |
-
"grad_norm": 0.
|
| 488 |
"learning_rate": 2e-05,
|
| 489 |
-
"loss":
|
| 490 |
-
"mean_token_accuracy": 0.
|
| 491 |
"step": 290
|
| 492 |
},
|
| 493 |
{
|
| 494 |
-
"epoch": 0.
|
| 495 |
-
"grad_norm": 0.
|
| 496 |
"learning_rate": 2e-05,
|
| 497 |
-
"loss": 1.
|
| 498 |
-
"mean_token_accuracy": 0.
|
| 499 |
"step": 295
|
| 500 |
},
|
| 501 |
{
|
| 502 |
-
"epoch": 0.
|
| 503 |
-
"grad_norm": 0.
|
| 504 |
"learning_rate": 2e-05,
|
| 505 |
-
"loss": 1.
|
| 506 |
-
"mean_token_accuracy": 0.
|
| 507 |
"step": 300
|
| 508 |
},
|
| 509 |
{
|
| 510 |
-
"epoch": 0.
|
| 511 |
-
"eval_loss": 1.
|
| 512 |
-
"eval_mean_token_accuracy": 0.
|
| 513 |
-
"eval_runtime":
|
| 514 |
-
"eval_samples_per_second":
|
| 515 |
-
"eval_steps_per_second":
|
| 516 |
"step": 300
|
| 517 |
},
|
| 518 |
{
|
| 519 |
-
"epoch": 0.
|
| 520 |
-
"grad_norm": 0.
|
| 521 |
"learning_rate": 2e-05,
|
| 522 |
-
"loss": 1.
|
| 523 |
-
"mean_token_accuracy": 0.
|
| 524 |
"step": 305
|
| 525 |
},
|
| 526 |
{
|
| 527 |
-
"epoch": 0.
|
| 528 |
-
"grad_norm": 0.
|
| 529 |
"learning_rate": 2e-05,
|
| 530 |
-
"loss": 1.
|
| 531 |
-
"mean_token_accuracy": 0.
|
| 532 |
"step": 310
|
| 533 |
},
|
| 534 |
{
|
| 535 |
-
"epoch": 0.
|
| 536 |
-
"grad_norm": 0.
|
| 537 |
"learning_rate": 2e-05,
|
| 538 |
-
"loss":
|
| 539 |
-
"mean_token_accuracy": 0.
|
| 540 |
"step": 315
|
| 541 |
},
|
| 542 |
{
|
| 543 |
-
"epoch": 0.
|
| 544 |
-
"grad_norm": 0.
|
| 545 |
"learning_rate": 2e-05,
|
| 546 |
-
"loss": 1.
|
| 547 |
-
"mean_token_accuracy": 0.
|
| 548 |
"step": 320
|
| 549 |
},
|
| 550 |
{
|
| 551 |
-
"epoch": 0.
|
| 552 |
-
"grad_norm": 0.
|
| 553 |
"learning_rate": 2e-05,
|
| 554 |
-
"loss": 1.
|
| 555 |
-
"mean_token_accuracy": 0.
|
| 556 |
"step": 325
|
| 557 |
},
|
| 558 |
{
|
| 559 |
-
"epoch": 0.
|
| 560 |
-
"grad_norm": 0.
|
| 561 |
"learning_rate": 2e-05,
|
| 562 |
-
"loss":
|
| 563 |
-
"mean_token_accuracy": 0.
|
| 564 |
"step": 330
|
| 565 |
},
|
| 566 |
{
|
| 567 |
-
"epoch": 0.
|
| 568 |
-
"grad_norm": 0.
|
| 569 |
"learning_rate": 2e-05,
|
| 570 |
-
"loss": 1.
|
| 571 |
-
"mean_token_accuracy": 0.
|
| 572 |
"step": 335
|
| 573 |
},
|
| 574 |
{
|
| 575 |
-
"epoch": 0.
|
| 576 |
-
"grad_norm": 0.
|
| 577 |
"learning_rate": 2e-05,
|
| 578 |
-
"loss":
|
| 579 |
-
"mean_token_accuracy": 0.
|
| 580 |
"step": 340
|
| 581 |
},
|
| 582 |
{
|
| 583 |
-
"epoch":
|
| 584 |
-
"grad_norm": 0.
|
| 585 |
"learning_rate": 2e-05,
|
| 586 |
-
"loss": 1.
|
| 587 |
-
"mean_token_accuracy": 0.
|
| 588 |
"step": 345
|
| 589 |
},
|
| 590 |
{
|
| 591 |
-
"epoch":
|
| 592 |
-
"grad_norm": 0.
|
| 593 |
"learning_rate": 2e-05,
|
| 594 |
-
"loss":
|
| 595 |
-
"mean_token_accuracy": 0.
|
| 596 |
"step": 350
|
| 597 |
},
|
| 598 |
{
|
| 599 |
-
"epoch":
|
| 600 |
-
"grad_norm": 0.
|
| 601 |
"learning_rate": 2e-05,
|
| 602 |
-
"loss":
|
| 603 |
-
"mean_token_accuracy": 0.
|
| 604 |
"step": 355
|
| 605 |
},
|
| 606 |
{
|
| 607 |
-
"epoch":
|
| 608 |
-
"grad_norm": 0.
|
| 609 |
"learning_rate": 2e-05,
|
| 610 |
-
"loss":
|
| 611 |
-
"mean_token_accuracy": 0.
|
| 612 |
"step": 360
|
| 613 |
},
|
| 614 |
{
|
| 615 |
-
"epoch":
|
| 616 |
-
"grad_norm": 0.
|
| 617 |
"learning_rate": 2e-05,
|
| 618 |
-
"loss":
|
| 619 |
-
"mean_token_accuracy": 0.
|
| 620 |
"step": 365
|
| 621 |
},
|
| 622 |
{
|
| 623 |
-
"epoch":
|
| 624 |
-
"grad_norm": 0.
|
| 625 |
"learning_rate": 2e-05,
|
| 626 |
-
"loss":
|
| 627 |
-
"mean_token_accuracy": 0.
|
| 628 |
"step": 370
|
| 629 |
},
|
| 630 |
{
|
| 631 |
-
"epoch":
|
| 632 |
-
"grad_norm": 0.
|
| 633 |
"learning_rate": 2e-05,
|
| 634 |
-
"loss":
|
| 635 |
-
"mean_token_accuracy": 0.
|
| 636 |
"step": 375
|
| 637 |
},
|
| 638 |
{
|
| 639 |
-
"epoch":
|
| 640 |
-
"grad_norm": 0.
|
| 641 |
"learning_rate": 2e-05,
|
| 642 |
-
"loss":
|
| 643 |
-
"mean_token_accuracy": 0.
|
| 644 |
"step": 380
|
| 645 |
},
|
| 646 |
{
|
| 647 |
-
"epoch":
|
| 648 |
-
"grad_norm": 0.
|
| 649 |
"learning_rate": 2e-05,
|
| 650 |
-
"loss":
|
| 651 |
-
"mean_token_accuracy": 0.
|
| 652 |
"step": 385
|
| 653 |
},
|
| 654 |
{
|
| 655 |
-
"epoch":
|
| 656 |
-
"grad_norm": 0.
|
| 657 |
"learning_rate": 2e-05,
|
| 658 |
-
"loss":
|
| 659 |
-
"mean_token_accuracy": 0.
|
| 660 |
"step": 390
|
| 661 |
},
|
| 662 |
{
|
| 663 |
-
"epoch":
|
| 664 |
-
"grad_norm": 0.
|
| 665 |
"learning_rate": 2e-05,
|
| 666 |
-
"loss":
|
| 667 |
-
"mean_token_accuracy": 0.
|
| 668 |
"step": 395
|
| 669 |
},
|
| 670 |
{
|
| 671 |
-
"epoch":
|
| 672 |
-
"grad_norm": 0.
|
| 673 |
"learning_rate": 2e-05,
|
| 674 |
-
"loss":
|
| 675 |
-
"mean_token_accuracy": 0.
|
| 676 |
"step": 400
|
| 677 |
},
|
| 678 |
{
|
| 679 |
-
"epoch":
|
| 680 |
-
"eval_loss":
|
| 681 |
-
"eval_mean_token_accuracy": 0.
|
| 682 |
-
"eval_runtime":
|
| 683 |
-
"eval_samples_per_second":
|
| 684 |
-
"eval_steps_per_second":
|
| 685 |
"step": 400
|
| 686 |
},
|
| 687 |
{
|
| 688 |
-
"epoch":
|
| 689 |
-
"grad_norm": 0.
|
| 690 |
"learning_rate": 2e-05,
|
| 691 |
-
"loss":
|
| 692 |
-
"mean_token_accuracy": 0.
|
| 693 |
"step": 405
|
| 694 |
},
|
| 695 |
{
|
| 696 |
-
"epoch":
|
| 697 |
-
"grad_norm": 0.
|
| 698 |
"learning_rate": 2e-05,
|
| 699 |
-
"loss":
|
| 700 |
-
"mean_token_accuracy": 0.
|
| 701 |
"step": 410
|
| 702 |
},
|
| 703 |
{
|
| 704 |
-
"epoch":
|
| 705 |
-
"grad_norm": 0.
|
| 706 |
"learning_rate": 2e-05,
|
| 707 |
-
"loss":
|
| 708 |
-
"mean_token_accuracy": 0.
|
| 709 |
"step": 415
|
| 710 |
},
|
| 711 |
{
|
| 712 |
-
"epoch":
|
| 713 |
-
"grad_norm": 0.
|
| 714 |
"learning_rate": 2e-05,
|
| 715 |
-
"loss":
|
| 716 |
-
"mean_token_accuracy": 0.
|
| 717 |
"step": 420
|
| 718 |
},
|
| 719 |
{
|
| 720 |
-
"epoch":
|
| 721 |
-
"grad_norm": 0.
|
| 722 |
"learning_rate": 2e-05,
|
| 723 |
-
"loss":
|
| 724 |
-
"mean_token_accuracy": 0.
|
| 725 |
"step": 425
|
| 726 |
},
|
| 727 |
{
|
| 728 |
-
"epoch":
|
| 729 |
-
"grad_norm": 0.
|
| 730 |
"learning_rate": 2e-05,
|
| 731 |
-
"loss":
|
| 732 |
-
"mean_token_accuracy": 0.
|
| 733 |
"step": 430
|
| 734 |
},
|
| 735 |
{
|
| 736 |
-
"epoch":
|
| 737 |
-
"grad_norm": 0.
|
| 738 |
"learning_rate": 2e-05,
|
| 739 |
-
"loss":
|
| 740 |
-
"mean_token_accuracy": 0.
|
| 741 |
"step": 435
|
| 742 |
},
|
| 743 |
{
|
| 744 |
-
"epoch":
|
| 745 |
-
"grad_norm": 0.
|
| 746 |
"learning_rate": 2e-05,
|
| 747 |
-
"loss":
|
| 748 |
-
"mean_token_accuracy": 0.
|
| 749 |
"step": 440
|
| 750 |
},
|
| 751 |
{
|
| 752 |
-
"epoch":
|
| 753 |
-
"grad_norm": 0.
|
| 754 |
"learning_rate": 2e-05,
|
| 755 |
-
"loss":
|
| 756 |
-
"mean_token_accuracy": 0.
|
| 757 |
"step": 445
|
| 758 |
},
|
| 759 |
{
|
| 760 |
-
"epoch":
|
| 761 |
-
"grad_norm": 0.
|
| 762 |
"learning_rate": 2e-05,
|
| 763 |
-
"loss":
|
| 764 |
-
"mean_token_accuracy": 0.
|
| 765 |
"step": 450
|
| 766 |
},
|
| 767 |
{
|
| 768 |
-
"epoch":
|
| 769 |
-
"grad_norm": 0.
|
| 770 |
"learning_rate": 2e-05,
|
| 771 |
-
"loss":
|
| 772 |
-
"mean_token_accuracy": 0.
|
| 773 |
"step": 455
|
| 774 |
},
|
| 775 |
{
|
| 776 |
-
"epoch":
|
| 777 |
-
"grad_norm": 0.
|
| 778 |
"learning_rate": 2e-05,
|
| 779 |
-
"loss":
|
| 780 |
-
"mean_token_accuracy": 0.
|
| 781 |
"step": 460
|
| 782 |
},
|
| 783 |
{
|
| 784 |
-
"epoch":
|
| 785 |
-
"grad_norm": 0.
|
| 786 |
"learning_rate": 2e-05,
|
| 787 |
-
"loss":
|
| 788 |
-
"mean_token_accuracy": 0.
|
| 789 |
"step": 465
|
| 790 |
},
|
| 791 |
{
|
| 792 |
-
"epoch":
|
| 793 |
-
"grad_norm": 0.
|
| 794 |
"learning_rate": 2e-05,
|
| 795 |
-
"loss":
|
| 796 |
-
"mean_token_accuracy": 0.
|
| 797 |
"step": 470
|
| 798 |
},
|
| 799 |
{
|
| 800 |
-
"epoch":
|
| 801 |
-
"grad_norm": 0.
|
| 802 |
"learning_rate": 2e-05,
|
| 803 |
-
"loss":
|
| 804 |
-
"mean_token_accuracy": 0.
|
| 805 |
"step": 475
|
| 806 |
},
|
| 807 |
{
|
| 808 |
-
"epoch":
|
| 809 |
-
"grad_norm": 0.
|
| 810 |
"learning_rate": 2e-05,
|
| 811 |
-
"loss":
|
| 812 |
-
"mean_token_accuracy": 0.
|
| 813 |
"step": 480
|
| 814 |
},
|
| 815 |
{
|
| 816 |
-
"epoch":
|
| 817 |
-
"grad_norm": 0.
|
| 818 |
"learning_rate": 2e-05,
|
| 819 |
-
"loss":
|
| 820 |
-
"mean_token_accuracy": 0.
|
| 821 |
"step": 485
|
| 822 |
},
|
| 823 |
{
|
| 824 |
-
"epoch":
|
| 825 |
-
"grad_norm": 0.
|
| 826 |
"learning_rate": 2e-05,
|
| 827 |
-
"loss":
|
| 828 |
-
"mean_token_accuracy": 0.
|
| 829 |
"step": 490
|
| 830 |
},
|
| 831 |
{
|
| 832 |
-
"epoch":
|
| 833 |
-
"grad_norm": 0.
|
| 834 |
"learning_rate": 2e-05,
|
| 835 |
-
"loss":
|
| 836 |
-
"mean_token_accuracy": 0.
|
| 837 |
"step": 495
|
| 838 |
},
|
| 839 |
{
|
| 840 |
-
"epoch":
|
| 841 |
-
"grad_norm": 0.
|
| 842 |
"learning_rate": 2e-05,
|
| 843 |
-
"loss": 0.
|
| 844 |
-
"mean_token_accuracy": 0.
|
| 845 |
"step": 500
|
| 846 |
},
|
| 847 |
{
|
| 848 |
-
"epoch":
|
| 849 |
-
"eval_loss":
|
| 850 |
-
"eval_mean_token_accuracy": 0.
|
| 851 |
-
"eval_runtime":
|
| 852 |
-
"eval_samples_per_second":
|
| 853 |
-
"eval_steps_per_second":
|
| 854 |
"step": 500
|
| 855 |
},
|
| 856 |
{
|
| 857 |
-
"epoch":
|
| 858 |
-
"grad_norm": 0.
|
| 859 |
"learning_rate": 2e-05,
|
| 860 |
-
"loss":
|
| 861 |
-
"mean_token_accuracy": 0.
|
| 862 |
"step": 505
|
| 863 |
},
|
| 864 |
{
|
| 865 |
-
"epoch":
|
| 866 |
-
"grad_norm": 0.
|
| 867 |
"learning_rate": 2e-05,
|
| 868 |
-
"loss":
|
| 869 |
-
"mean_token_accuracy": 0.
|
| 870 |
"step": 510
|
| 871 |
},
|
| 872 |
{
|
| 873 |
-
"epoch":
|
| 874 |
-
"grad_norm": 0.
|
| 875 |
"learning_rate": 2e-05,
|
| 876 |
-
"loss":
|
| 877 |
-
"mean_token_accuracy": 0.
|
| 878 |
"step": 515
|
| 879 |
},
|
| 880 |
{
|
| 881 |
-
"epoch":
|
| 882 |
-
"grad_norm": 0.
|
| 883 |
"learning_rate": 2e-05,
|
| 884 |
-
"loss":
|
| 885 |
-
"mean_token_accuracy": 0.
|
| 886 |
"step": 520
|
| 887 |
},
|
| 888 |
{
|
| 889 |
-
"epoch":
|
| 890 |
-
"grad_norm": 0.
|
| 891 |
"learning_rate": 2e-05,
|
| 892 |
-
"loss": 0.
|
| 893 |
-
"mean_token_accuracy": 0.
|
| 894 |
"step": 525
|
| 895 |
},
|
| 896 |
{
|
| 897 |
-
"epoch":
|
| 898 |
-
"grad_norm": 0.
|
| 899 |
"learning_rate": 2e-05,
|
| 900 |
-
"loss":
|
| 901 |
-
"mean_token_accuracy": 0.
|
| 902 |
"step": 530
|
| 903 |
},
|
| 904 |
{
|
| 905 |
-
"epoch":
|
| 906 |
-
"grad_norm": 0.
|
| 907 |
"learning_rate": 2e-05,
|
| 908 |
-
"loss":
|
| 909 |
-
"mean_token_accuracy": 0.
|
| 910 |
"step": 535
|
| 911 |
},
|
| 912 |
{
|
| 913 |
-
"epoch":
|
| 914 |
-
"grad_norm": 0.
|
| 915 |
"learning_rate": 2e-05,
|
| 916 |
-
"loss":
|
| 917 |
-
"mean_token_accuracy": 0.
|
| 918 |
"step": 540
|
| 919 |
},
|
| 920 |
{
|
| 921 |
-
"epoch":
|
| 922 |
-
"grad_norm": 0.
|
| 923 |
"learning_rate": 2e-05,
|
| 924 |
-
"loss":
|
| 925 |
-
"mean_token_accuracy": 0.
|
| 926 |
"step": 545
|
| 927 |
},
|
| 928 |
{
|
| 929 |
-
"epoch":
|
| 930 |
-
"grad_norm": 0.
|
| 931 |
"learning_rate": 2e-05,
|
| 932 |
-
"loss": 0.
|
| 933 |
-
"mean_token_accuracy": 0.
|
| 934 |
"step": 550
|
| 935 |
},
|
| 936 |
{
|
| 937 |
-
"epoch":
|
| 938 |
-
"grad_norm": 0.
|
| 939 |
"learning_rate": 2e-05,
|
| 940 |
-
"loss":
|
| 941 |
-
"mean_token_accuracy": 0.
|
| 942 |
"step": 555
|
| 943 |
},
|
| 944 |
{
|
| 945 |
-
"epoch":
|
| 946 |
-
"grad_norm": 0.
|
| 947 |
"learning_rate": 2e-05,
|
| 948 |
-
"loss":
|
| 949 |
-
"mean_token_accuracy": 0.
|
| 950 |
"step": 560
|
| 951 |
},
|
| 952 |
{
|
| 953 |
-
"epoch":
|
| 954 |
-
"grad_norm": 0.
|
| 955 |
"learning_rate": 2e-05,
|
| 956 |
-
"loss":
|
| 957 |
-
"mean_token_accuracy": 0.
|
| 958 |
"step": 565
|
| 959 |
},
|
| 960 |
{
|
| 961 |
-
"epoch":
|
| 962 |
-
"grad_norm": 0.
|
| 963 |
"learning_rate": 2e-05,
|
| 964 |
-
"loss": 0.
|
| 965 |
-
"mean_token_accuracy": 0.
|
| 966 |
"step": 570
|
| 967 |
},
|
| 968 |
{
|
| 969 |
-
"epoch":
|
| 970 |
-
"grad_norm": 0.
|
| 971 |
"learning_rate": 2e-05,
|
| 972 |
-
"loss":
|
| 973 |
-
"mean_token_accuracy": 0.
|
| 974 |
"step": 575
|
| 975 |
},
|
| 976 |
{
|
| 977 |
-
"epoch":
|
| 978 |
-
"grad_norm": 0.
|
| 979 |
"learning_rate": 2e-05,
|
| 980 |
-
"loss": 0.
|
| 981 |
-
"mean_token_accuracy": 0.
|
| 982 |
"step": 580
|
| 983 |
},
|
| 984 |
{
|
| 985 |
-
"epoch":
|
| 986 |
-
"grad_norm": 0.
|
| 987 |
"learning_rate": 2e-05,
|
| 988 |
-
"loss":
|
| 989 |
-
"mean_token_accuracy": 0.
|
| 990 |
"step": 585
|
| 991 |
},
|
| 992 |
{
|
| 993 |
-
"epoch":
|
| 994 |
-
"grad_norm": 0.
|
| 995 |
"learning_rate": 2e-05,
|
| 996 |
-
"loss":
|
| 997 |
-
"mean_token_accuracy": 0.
|
| 998 |
"step": 590
|
| 999 |
},
|
| 1000 |
{
|
| 1001 |
-
"epoch":
|
| 1002 |
-
"grad_norm": 0.
|
| 1003 |
"learning_rate": 2e-05,
|
| 1004 |
-
"loss":
|
| 1005 |
-
"mean_token_accuracy": 0.
|
| 1006 |
"step": 595
|
| 1007 |
},
|
| 1008 |
{
|
| 1009 |
-
"epoch":
|
| 1010 |
-
"grad_norm": 0.
|
| 1011 |
"learning_rate": 2e-05,
|
| 1012 |
-
"loss":
|
| 1013 |
-
"mean_token_accuracy": 0.
|
| 1014 |
"step": 600
|
| 1015 |
},
|
| 1016 |
{
|
| 1017 |
-
"epoch":
|
| 1018 |
-
"eval_loss":
|
| 1019 |
-
"eval_mean_token_accuracy": 0.
|
| 1020 |
-
"eval_runtime":
|
| 1021 |
-
"eval_samples_per_second":
|
| 1022 |
-
"eval_steps_per_second":
|
| 1023 |
"step": 600
|
| 1024 |
},
|
| 1025 |
{
|
| 1026 |
-
"epoch":
|
| 1027 |
-
"grad_norm": 0.
|
| 1028 |
"learning_rate": 2e-05,
|
| 1029 |
-
"loss":
|
| 1030 |
-
"mean_token_accuracy": 0.
|
| 1031 |
"step": 605
|
| 1032 |
},
|
| 1033 |
{
|
| 1034 |
-
"epoch":
|
| 1035 |
-
"grad_norm": 0.
|
| 1036 |
"learning_rate": 2e-05,
|
| 1037 |
-
"loss":
|
| 1038 |
-
"mean_token_accuracy": 0.
|
| 1039 |
"step": 610
|
| 1040 |
},
|
| 1041 |
{
|
| 1042 |
-
"epoch":
|
| 1043 |
-
"grad_norm": 0.
|
| 1044 |
"learning_rate": 2e-05,
|
| 1045 |
-
"loss": 0.
|
| 1046 |
-
"mean_token_accuracy": 0.
|
| 1047 |
"step": 615
|
| 1048 |
},
|
| 1049 |
{
|
| 1050 |
-
"epoch":
|
| 1051 |
-
"grad_norm": 0.
|
| 1052 |
"learning_rate": 2e-05,
|
| 1053 |
-
"loss":
|
| 1054 |
-
"mean_token_accuracy": 0.
|
| 1055 |
"step": 620
|
| 1056 |
},
|
| 1057 |
{
|
| 1058 |
-
"epoch":
|
| 1059 |
-
"grad_norm": 0.
|
| 1060 |
"learning_rate": 2e-05,
|
| 1061 |
-
"loss": 0.
|
| 1062 |
-
"mean_token_accuracy": 0.
|
| 1063 |
"step": 625
|
| 1064 |
},
|
| 1065 |
{
|
| 1066 |
-
"epoch":
|
| 1067 |
-
"grad_norm": 0.
|
| 1068 |
"learning_rate": 2e-05,
|
| 1069 |
-
"loss":
|
| 1070 |
-
"mean_token_accuracy": 0.
|
| 1071 |
"step": 630
|
| 1072 |
},
|
| 1073 |
{
|
| 1074 |
-
"epoch":
|
| 1075 |
-
"grad_norm": 0.
|
| 1076 |
"learning_rate": 2e-05,
|
| 1077 |
-
"loss":
|
| 1078 |
-
"mean_token_accuracy": 0.
|
| 1079 |
"step": 635
|
| 1080 |
},
|
| 1081 |
{
|
| 1082 |
-
"epoch":
|
| 1083 |
-
"grad_norm": 0.
|
| 1084 |
"learning_rate": 2e-05,
|
| 1085 |
-
"loss":
|
| 1086 |
-
"mean_token_accuracy": 0.
|
| 1087 |
"step": 640
|
| 1088 |
},
|
| 1089 |
{
|
| 1090 |
-
"epoch":
|
| 1091 |
-
"grad_norm": 0.
|
| 1092 |
"learning_rate": 2e-05,
|
| 1093 |
-
"loss":
|
| 1094 |
-
"mean_token_accuracy": 0.
|
| 1095 |
"step": 645
|
| 1096 |
},
|
| 1097 |
{
|
| 1098 |
-
"epoch":
|
| 1099 |
-
"grad_norm": 0.
|
| 1100 |
"learning_rate": 2e-05,
|
| 1101 |
-
"loss":
|
| 1102 |
-
"mean_token_accuracy": 0.
|
| 1103 |
"step": 650
|
| 1104 |
},
|
| 1105 |
{
|
| 1106 |
-
"epoch":
|
| 1107 |
-
"grad_norm": 0.
|
| 1108 |
"learning_rate": 2e-05,
|
| 1109 |
-
"loss":
|
| 1110 |
-
"mean_token_accuracy": 0.
|
| 1111 |
"step": 655
|
| 1112 |
},
|
| 1113 |
{
|
| 1114 |
-
"epoch":
|
| 1115 |
-
"grad_norm": 0.
|
| 1116 |
"learning_rate": 2e-05,
|
| 1117 |
-
"loss":
|
| 1118 |
-
"mean_token_accuracy": 0.
|
| 1119 |
"step": 660
|
| 1120 |
},
|
| 1121 |
{
|
| 1122 |
-
"epoch":
|
| 1123 |
-
"grad_norm": 0.
|
| 1124 |
"learning_rate": 2e-05,
|
| 1125 |
-
"loss": 0.
|
| 1126 |
-
"mean_token_accuracy": 0.
|
| 1127 |
"step": 665
|
| 1128 |
},
|
| 1129 |
{
|
| 1130 |
-
"epoch":
|
| 1131 |
-
"grad_norm": 0.
|
| 1132 |
"learning_rate": 2e-05,
|
| 1133 |
-
"loss":
|
| 1134 |
-
"mean_token_accuracy": 0.
|
| 1135 |
"step": 670
|
| 1136 |
},
|
| 1137 |
{
|
| 1138 |
-
"epoch":
|
| 1139 |
-
"grad_norm": 0.
|
| 1140 |
"learning_rate": 2e-05,
|
| 1141 |
-
"loss": 0.
|
| 1142 |
-
"mean_token_accuracy": 0.
|
| 1143 |
"step": 675
|
| 1144 |
},
|
| 1145 |
{
|
| 1146 |
-
"epoch":
|
| 1147 |
-
"grad_norm": 0.
|
| 1148 |
"learning_rate": 2e-05,
|
| 1149 |
-
"loss":
|
| 1150 |
-
"mean_token_accuracy": 0.
|
| 1151 |
"step": 680
|
| 1152 |
},
|
| 1153 |
{
|
| 1154 |
-
"epoch":
|
| 1155 |
-
"grad_norm": 0.
|
| 1156 |
"learning_rate": 2e-05,
|
| 1157 |
-
"loss": 0.
|
| 1158 |
-
"mean_token_accuracy": 0.
|
| 1159 |
"step": 685
|
| 1160 |
},
|
| 1161 |
{
|
| 1162 |
-
"epoch":
|
| 1163 |
-
"grad_norm": 0.
|
| 1164 |
"learning_rate": 2e-05,
|
| 1165 |
-
"loss":
|
| 1166 |
-
"mean_token_accuracy": 0.
|
| 1167 |
"step": 690
|
| 1168 |
},
|
| 1169 |
{
|
| 1170 |
-
"epoch":
|
| 1171 |
-
"grad_norm": 0.
|
| 1172 |
"learning_rate": 2e-05,
|
| 1173 |
-
"loss":
|
| 1174 |
-
"mean_token_accuracy": 0.
|
| 1175 |
"step": 695
|
| 1176 |
},
|
| 1177 |
{
|
| 1178 |
-
"epoch":
|
| 1179 |
-
"grad_norm": 0.
|
| 1180 |
"learning_rate": 2e-05,
|
| 1181 |
-
"loss":
|
| 1182 |
-
"mean_token_accuracy": 0.
|
| 1183 |
"step": 700
|
| 1184 |
},
|
| 1185 |
{
|
| 1186 |
-
"epoch":
|
| 1187 |
-
"eval_loss":
|
| 1188 |
-
"eval_mean_token_accuracy": 0.
|
| 1189 |
-
"eval_runtime":
|
| 1190 |
-
"eval_samples_per_second":
|
| 1191 |
-
"eval_steps_per_second":
|
| 1192 |
"step": 700
|
| 1193 |
},
|
| 1194 |
{
|
| 1195 |
-
"epoch":
|
| 1196 |
-
"grad_norm": 0.
|
| 1197 |
"learning_rate": 2e-05,
|
| 1198 |
-
"loss":
|
| 1199 |
-
"mean_token_accuracy": 0.
|
| 1200 |
"step": 705
|
| 1201 |
},
|
| 1202 |
{
|
| 1203 |
-
"epoch":
|
| 1204 |
-
"grad_norm": 0.
|
| 1205 |
"learning_rate": 2e-05,
|
| 1206 |
-
"loss":
|
| 1207 |
-
"mean_token_accuracy": 0.
|
| 1208 |
"step": 710
|
| 1209 |
},
|
| 1210 |
{
|
| 1211 |
-
"epoch":
|
| 1212 |
-
"grad_norm": 0.
|
| 1213 |
"learning_rate": 2e-05,
|
| 1214 |
-
"loss":
|
| 1215 |
-
"mean_token_accuracy": 0.
|
| 1216 |
"step": 715
|
| 1217 |
},
|
| 1218 |
{
|
| 1219 |
-
"epoch":
|
| 1220 |
-
"grad_norm": 0.
|
| 1221 |
"learning_rate": 2e-05,
|
| 1222 |
-
"loss":
|
| 1223 |
-
"mean_token_accuracy": 0.
|
| 1224 |
"step": 720
|
| 1225 |
},
|
| 1226 |
{
|
| 1227 |
-
"epoch":
|
| 1228 |
-
"grad_norm": 0.
|
| 1229 |
"learning_rate": 2e-05,
|
| 1230 |
-
"loss":
|
| 1231 |
-
"mean_token_accuracy": 0.
|
| 1232 |
"step": 725
|
| 1233 |
},
|
| 1234 |
{
|
| 1235 |
-
"epoch":
|
| 1236 |
-
"grad_norm": 0.
|
| 1237 |
"learning_rate": 2e-05,
|
| 1238 |
-
"loss": 0.
|
| 1239 |
-
"mean_token_accuracy": 0.
|
| 1240 |
"step": 730
|
| 1241 |
},
|
| 1242 |
{
|
| 1243 |
-
"epoch":
|
| 1244 |
-
"grad_norm": 0.
|
| 1245 |
"learning_rate": 2e-05,
|
| 1246 |
-
"loss":
|
| 1247 |
-
"mean_token_accuracy": 0.
|
| 1248 |
"step": 735
|
| 1249 |
},
|
| 1250 |
{
|
| 1251 |
-
"epoch":
|
| 1252 |
-
"grad_norm": 0.
|
| 1253 |
"learning_rate": 2e-05,
|
| 1254 |
-
"loss":
|
| 1255 |
-
"mean_token_accuracy": 0.
|
| 1256 |
"step": 740
|
| 1257 |
},
|
| 1258 |
{
|
| 1259 |
-
"epoch":
|
| 1260 |
-
"grad_norm": 0.
|
| 1261 |
"learning_rate": 2e-05,
|
| 1262 |
-
"loss": 0.
|
| 1263 |
-
"mean_token_accuracy": 0.
|
| 1264 |
"step": 745
|
| 1265 |
},
|
| 1266 |
{
|
| 1267 |
-
"epoch":
|
| 1268 |
-
"grad_norm": 0.
|
| 1269 |
"learning_rate": 2e-05,
|
| 1270 |
-
"loss": 0.
|
| 1271 |
-
"mean_token_accuracy": 0.
|
| 1272 |
"step": 750
|
| 1273 |
},
|
| 1274 |
{
|
| 1275 |
-
"epoch":
|
| 1276 |
-
"grad_norm": 0.
|
| 1277 |
"learning_rate": 2e-05,
|
| 1278 |
-
"loss": 0.
|
| 1279 |
-
"mean_token_accuracy": 0.
|
| 1280 |
"step": 755
|
| 1281 |
},
|
| 1282 |
{
|
| 1283 |
-
"epoch":
|
| 1284 |
-
"grad_norm": 0.
|
| 1285 |
"learning_rate": 2e-05,
|
| 1286 |
-
"loss":
|
| 1287 |
-
"mean_token_accuracy": 0.
|
| 1288 |
"step": 760
|
| 1289 |
},
|
| 1290 |
{
|
| 1291 |
-
"epoch":
|
| 1292 |
-
"grad_norm": 0.
|
| 1293 |
"learning_rate": 2e-05,
|
| 1294 |
-
"loss": 0.
|
| 1295 |
-
"mean_token_accuracy": 0.
|
| 1296 |
"step": 765
|
| 1297 |
},
|
| 1298 |
{
|
| 1299 |
-
"epoch":
|
| 1300 |
-
"grad_norm": 0.
|
| 1301 |
"learning_rate": 2e-05,
|
| 1302 |
-
"loss": 0.
|
| 1303 |
-
"mean_token_accuracy": 0.
|
| 1304 |
"step": 770
|
| 1305 |
},
|
| 1306 |
{
|
| 1307 |
-
"epoch":
|
| 1308 |
-
"grad_norm": 0.
|
| 1309 |
"learning_rate": 2e-05,
|
| 1310 |
-
"loss":
|
| 1311 |
-
"mean_token_accuracy": 0.
|
| 1312 |
"step": 775
|
| 1313 |
},
|
| 1314 |
{
|
| 1315 |
-
"epoch":
|
| 1316 |
-
"grad_norm": 0.
|
| 1317 |
"learning_rate": 2e-05,
|
| 1318 |
-
"loss": 0.
|
| 1319 |
-
"mean_token_accuracy": 0.
|
| 1320 |
"step": 780
|
| 1321 |
},
|
| 1322 |
{
|
| 1323 |
-
"epoch":
|
| 1324 |
-
"grad_norm": 0.
|
| 1325 |
"learning_rate": 2e-05,
|
| 1326 |
-
"loss": 0.
|
| 1327 |
-
"mean_token_accuracy": 0.
|
| 1328 |
"step": 785
|
| 1329 |
},
|
| 1330 |
{
|
| 1331 |
-
"epoch":
|
| 1332 |
-
"grad_norm": 0.
|
| 1333 |
"learning_rate": 2e-05,
|
| 1334 |
-
"loss":
|
| 1335 |
-
"mean_token_accuracy": 0.
|
| 1336 |
"step": 790
|
| 1337 |
},
|
| 1338 |
{
|
| 1339 |
-
"epoch":
|
| 1340 |
-
"grad_norm": 0.
|
| 1341 |
"learning_rate": 2e-05,
|
| 1342 |
-
"loss": 0.
|
| 1343 |
-
"mean_token_accuracy": 0.
|
| 1344 |
"step": 795
|
| 1345 |
},
|
| 1346 |
{
|
| 1347 |
-
"epoch":
|
| 1348 |
-
"grad_norm": 0.
|
| 1349 |
"learning_rate": 2e-05,
|
| 1350 |
-
"loss":
|
| 1351 |
-
"mean_token_accuracy": 0.
|
| 1352 |
"step": 800
|
| 1353 |
},
|
| 1354 |
{
|
| 1355 |
-
"epoch":
|
| 1356 |
-
"eval_loss":
|
| 1357 |
-
"eval_mean_token_accuracy": 0.
|
| 1358 |
-
"eval_runtime":
|
| 1359 |
-
"eval_samples_per_second":
|
| 1360 |
-
"eval_steps_per_second":
|
| 1361 |
"step": 800
|
| 1362 |
},
|
| 1363 |
{
|
| 1364 |
-
"epoch":
|
| 1365 |
-
"grad_norm": 0.
|
| 1366 |
"learning_rate": 2e-05,
|
| 1367 |
-
"loss":
|
| 1368 |
-
"mean_token_accuracy": 0.
|
| 1369 |
"step": 805
|
| 1370 |
},
|
| 1371 |
{
|
| 1372 |
-
"epoch":
|
| 1373 |
-
"grad_norm": 0.
|
| 1374 |
"learning_rate": 2e-05,
|
| 1375 |
-
"loss":
|
| 1376 |
-
"mean_token_accuracy": 0.
|
| 1377 |
"step": 810
|
| 1378 |
},
|
| 1379 |
{
|
| 1380 |
-
"epoch":
|
| 1381 |
-
"grad_norm": 0.
|
| 1382 |
"learning_rate": 2e-05,
|
| 1383 |
-
"loss":
|
| 1384 |
-
"mean_token_accuracy": 0.
|
| 1385 |
"step": 815
|
| 1386 |
},
|
| 1387 |
{
|
| 1388 |
-
"epoch":
|
| 1389 |
-
"grad_norm": 0.
|
| 1390 |
"learning_rate": 2e-05,
|
| 1391 |
-
"loss": 0.
|
| 1392 |
-
"mean_token_accuracy": 0.
|
| 1393 |
"step": 820
|
| 1394 |
},
|
| 1395 |
{
|
| 1396 |
-
"epoch":
|
| 1397 |
-
"grad_norm": 0.
|
| 1398 |
"learning_rate": 2e-05,
|
| 1399 |
-
"loss": 0.
|
| 1400 |
-
"mean_token_accuracy": 0.
|
| 1401 |
"step": 825
|
| 1402 |
},
|
| 1403 |
{
|
| 1404 |
-
"epoch":
|
| 1405 |
-
"grad_norm": 0.
|
| 1406 |
"learning_rate": 2e-05,
|
| 1407 |
-
"loss": 0.
|
| 1408 |
-
"mean_token_accuracy": 0.
|
| 1409 |
"step": 830
|
| 1410 |
},
|
| 1411 |
{
|
| 1412 |
-
"epoch":
|
| 1413 |
-
"grad_norm": 0.
|
| 1414 |
"learning_rate": 2e-05,
|
| 1415 |
-
"loss": 0.
|
| 1416 |
-
"mean_token_accuracy": 0.
|
| 1417 |
"step": 835
|
| 1418 |
},
|
| 1419 |
{
|
| 1420 |
-
"epoch":
|
| 1421 |
-
"grad_norm": 0.
|
| 1422 |
"learning_rate": 2e-05,
|
| 1423 |
-
"loss": 0.
|
| 1424 |
-
"mean_token_accuracy": 0.
|
| 1425 |
"step": 840
|
| 1426 |
},
|
| 1427 |
{
|
| 1428 |
-
"epoch":
|
| 1429 |
-
"grad_norm": 0.
|
| 1430 |
"learning_rate": 2e-05,
|
| 1431 |
-
"loss":
|
| 1432 |
-
"mean_token_accuracy": 0.
|
| 1433 |
"step": 845
|
| 1434 |
},
|
| 1435 |
{
|
| 1436 |
-
"epoch":
|
| 1437 |
-
"grad_norm": 0.
|
| 1438 |
"learning_rate": 2e-05,
|
| 1439 |
-
"loss":
|
| 1440 |
-
"mean_token_accuracy": 0.
|
| 1441 |
"step": 850
|
| 1442 |
},
|
| 1443 |
{
|
| 1444 |
-
"epoch":
|
| 1445 |
-
"grad_norm": 0.
|
| 1446 |
"learning_rate": 2e-05,
|
| 1447 |
-
"loss": 0.
|
| 1448 |
-
"mean_token_accuracy": 0.
|
| 1449 |
"step": 855
|
| 1450 |
},
|
| 1451 |
{
|
| 1452 |
-
"epoch":
|
| 1453 |
-
"grad_norm": 0.
|
| 1454 |
"learning_rate": 2e-05,
|
| 1455 |
-
"loss":
|
| 1456 |
-
"mean_token_accuracy": 0.
|
| 1457 |
"step": 860
|
| 1458 |
},
|
| 1459 |
{
|
| 1460 |
-
"epoch":
|
| 1461 |
-
"grad_norm": 0.
|
| 1462 |
"learning_rate": 2e-05,
|
| 1463 |
-
"loss":
|
| 1464 |
-
"mean_token_accuracy": 0.
|
| 1465 |
"step": 865
|
| 1466 |
},
|
| 1467 |
{
|
| 1468 |
-
"epoch":
|
| 1469 |
-
"grad_norm": 0.
|
| 1470 |
"learning_rate": 2e-05,
|
| 1471 |
-
"loss": 0.
|
| 1472 |
-
"mean_token_accuracy": 0.
|
| 1473 |
"step": 870
|
| 1474 |
},
|
| 1475 |
{
|
| 1476 |
-
"epoch":
|
| 1477 |
-
"grad_norm": 0.
|
| 1478 |
"learning_rate": 2e-05,
|
| 1479 |
-
"loss": 0.
|
| 1480 |
-
"mean_token_accuracy": 0.
|
| 1481 |
"step": 875
|
| 1482 |
},
|
| 1483 |
{
|
| 1484 |
-
"epoch":
|
| 1485 |
-
"grad_norm": 0.
|
| 1486 |
"learning_rate": 2e-05,
|
| 1487 |
-
"loss": 0.
|
| 1488 |
-
"mean_token_accuracy": 0.
|
| 1489 |
"step": 880
|
| 1490 |
},
|
| 1491 |
{
|
| 1492 |
-
"epoch":
|
| 1493 |
-
"grad_norm": 0.
|
| 1494 |
"learning_rate": 2e-05,
|
| 1495 |
-
"loss":
|
| 1496 |
-
"mean_token_accuracy": 0.
|
| 1497 |
"step": 885
|
| 1498 |
},
|
| 1499 |
{
|
| 1500 |
-
"epoch":
|
| 1501 |
-
"grad_norm": 0.
|
| 1502 |
"learning_rate": 2e-05,
|
| 1503 |
-
"loss": 0.
|
| 1504 |
-
"mean_token_accuracy": 0.
|
| 1505 |
"step": 890
|
| 1506 |
},
|
| 1507 |
{
|
| 1508 |
-
"epoch":
|
| 1509 |
-
"grad_norm": 0.
|
| 1510 |
"learning_rate": 2e-05,
|
| 1511 |
-
"loss": 0.
|
| 1512 |
-
"mean_token_accuracy": 0.
|
| 1513 |
"step": 895
|
| 1514 |
},
|
| 1515 |
{
|
| 1516 |
-
"epoch":
|
| 1517 |
-
"grad_norm": 0.
|
| 1518 |
"learning_rate": 2e-05,
|
| 1519 |
-
"loss": 0.
|
| 1520 |
-
"mean_token_accuracy": 0.
|
| 1521 |
"step": 900
|
| 1522 |
},
|
| 1523 |
{
|
| 1524 |
-
"epoch":
|
| 1525 |
-
"eval_loss": 0.
|
| 1526 |
-
"eval_mean_token_accuracy": 0.
|
| 1527 |
-
"eval_runtime":
|
| 1528 |
-
"eval_samples_per_second":
|
| 1529 |
-
"eval_steps_per_second":
|
| 1530 |
"step": 900
|
| 1531 |
},
|
| 1532 |
{
|
| 1533 |
-
"epoch":
|
| 1534 |
-
"grad_norm": 0.
|
| 1535 |
"learning_rate": 2e-05,
|
| 1536 |
-
"loss": 0.
|
| 1537 |
-
"mean_token_accuracy": 0.
|
| 1538 |
"step": 905
|
| 1539 |
},
|
| 1540 |
{
|
| 1541 |
-
"epoch":
|
| 1542 |
-
"grad_norm": 0.
|
| 1543 |
"learning_rate": 2e-05,
|
| 1544 |
-
"loss": 0.
|
| 1545 |
-
"mean_token_accuracy": 0.
|
| 1546 |
"step": 910
|
| 1547 |
},
|
| 1548 |
{
|
| 1549 |
-
"epoch":
|
| 1550 |
-
"grad_norm": 0.
|
| 1551 |
"learning_rate": 2e-05,
|
| 1552 |
-
"loss": 0.
|
| 1553 |
-
"mean_token_accuracy": 0.
|
| 1554 |
"step": 915
|
| 1555 |
},
|
| 1556 |
{
|
| 1557 |
-
"epoch":
|
| 1558 |
-
"grad_norm": 0.
|
| 1559 |
"learning_rate": 2e-05,
|
| 1560 |
-
"loss": 0.
|
| 1561 |
-
"mean_token_accuracy": 0.
|
| 1562 |
"step": 920
|
| 1563 |
},
|
| 1564 |
{
|
| 1565 |
-
"epoch":
|
| 1566 |
-
"grad_norm": 0.
|
| 1567 |
"learning_rate": 2e-05,
|
| 1568 |
-
"loss":
|
| 1569 |
-
"mean_token_accuracy": 0.
|
| 1570 |
"step": 925
|
| 1571 |
},
|
| 1572 |
{
|
| 1573 |
-
"epoch":
|
| 1574 |
-
"grad_norm": 0.
|
| 1575 |
"learning_rate": 2e-05,
|
| 1576 |
-
"loss": 0.
|
| 1577 |
-
"mean_token_accuracy": 0.
|
| 1578 |
"step": 930
|
| 1579 |
},
|
| 1580 |
{
|
| 1581 |
-
"epoch":
|
| 1582 |
-
"grad_norm": 0.
|
| 1583 |
"learning_rate": 2e-05,
|
| 1584 |
-
"loss": 0.
|
| 1585 |
-
"mean_token_accuracy": 0.
|
| 1586 |
"step": 935
|
| 1587 |
},
|
| 1588 |
{
|
| 1589 |
-
"epoch":
|
| 1590 |
-
"grad_norm": 0.
|
| 1591 |
"learning_rate": 2e-05,
|
| 1592 |
-
"loss": 0.
|
| 1593 |
-
"mean_token_accuracy": 0.
|
| 1594 |
"step": 940
|
| 1595 |
},
|
| 1596 |
{
|
| 1597 |
-
"epoch":
|
| 1598 |
-
"grad_norm": 0.
|
| 1599 |
"learning_rate": 2e-05,
|
| 1600 |
-
"loss": 0.
|
| 1601 |
-
"mean_token_accuracy": 0.
|
| 1602 |
"step": 945
|
| 1603 |
},
|
| 1604 |
{
|
| 1605 |
-
"epoch":
|
| 1606 |
-
"grad_norm": 0.
|
| 1607 |
"learning_rate": 2e-05,
|
| 1608 |
-
"loss": 0.
|
| 1609 |
-
"mean_token_accuracy": 0.
|
| 1610 |
"step": 950
|
| 1611 |
},
|
| 1612 |
{
|
| 1613 |
-
"epoch":
|
| 1614 |
-
"grad_norm": 0.
|
| 1615 |
"learning_rate": 2e-05,
|
| 1616 |
-
"loss": 0.
|
| 1617 |
-
"mean_token_accuracy": 0.
|
| 1618 |
"step": 955
|
| 1619 |
},
|
| 1620 |
{
|
| 1621 |
-
"epoch":
|
| 1622 |
-
"grad_norm": 0.
|
| 1623 |
"learning_rate": 2e-05,
|
| 1624 |
-
"loss": 0.
|
| 1625 |
-
"mean_token_accuracy": 0.
|
| 1626 |
"step": 960
|
| 1627 |
},
|
| 1628 |
{
|
| 1629 |
-
"epoch":
|
| 1630 |
-
"grad_norm": 0.
|
| 1631 |
"learning_rate": 2e-05,
|
| 1632 |
-
"loss":
|
| 1633 |
-
"mean_token_accuracy": 0.
|
| 1634 |
"step": 965
|
| 1635 |
},
|
| 1636 |
{
|
| 1637 |
-
"epoch":
|
| 1638 |
-
"grad_norm": 0.
|
| 1639 |
"learning_rate": 2e-05,
|
| 1640 |
-
"loss": 0.
|
| 1641 |
-
"mean_token_accuracy": 0.
|
| 1642 |
"step": 970
|
| 1643 |
},
|
| 1644 |
{
|
| 1645 |
-
"epoch":
|
| 1646 |
-
"grad_norm": 0.
|
| 1647 |
"learning_rate": 2e-05,
|
| 1648 |
-
"loss": 0.
|
| 1649 |
-
"mean_token_accuracy": 0.
|
| 1650 |
"step": 975
|
| 1651 |
},
|
| 1652 |
{
|
| 1653 |
-
"epoch":
|
| 1654 |
-
"grad_norm": 0.
|
| 1655 |
"learning_rate": 2e-05,
|
| 1656 |
-
"loss": 0.
|
| 1657 |
-
"mean_token_accuracy": 0.
|
| 1658 |
"step": 980
|
| 1659 |
},
|
| 1660 |
{
|
| 1661 |
-
"epoch":
|
| 1662 |
-
"grad_norm": 0.
|
| 1663 |
"learning_rate": 2e-05,
|
| 1664 |
-
"loss": 0.
|
| 1665 |
-
"mean_token_accuracy": 0.
|
| 1666 |
"step": 985
|
| 1667 |
},
|
| 1668 |
{
|
| 1669 |
-
"epoch":
|
| 1670 |
-
"grad_norm": 0.
|
| 1671 |
"learning_rate": 2e-05,
|
| 1672 |
-
"loss":
|
| 1673 |
-
"mean_token_accuracy": 0.
|
| 1674 |
"step": 990
|
| 1675 |
},
|
| 1676 |
{
|
| 1677 |
-
"epoch":
|
| 1678 |
-
"grad_norm": 0.
|
| 1679 |
"learning_rate": 2e-05,
|
| 1680 |
-
"loss":
|
| 1681 |
-
"mean_token_accuracy": 0.
|
| 1682 |
"step": 995
|
| 1683 |
},
|
| 1684 |
{
|
| 1685 |
-
"epoch":
|
| 1686 |
-
"grad_norm": 0.
|
| 1687 |
"learning_rate": 2e-05,
|
| 1688 |
-
"loss": 0.
|
| 1689 |
-
"mean_token_accuracy": 0.
|
| 1690 |
"step": 1000
|
| 1691 |
},
|
| 1692 |
{
|
| 1693 |
-
"epoch":
|
| 1694 |
-
"eval_loss": 0.
|
| 1695 |
-
"eval_mean_token_accuracy": 0.
|
| 1696 |
-
"eval_runtime":
|
| 1697 |
-
"eval_samples_per_second":
|
| 1698 |
-
"eval_steps_per_second":
|
| 1699 |
"step": 1000
|
| 1700 |
},
|
| 1701 |
{
|
| 1702 |
-
"epoch":
|
| 1703 |
-
"grad_norm": 0.
|
| 1704 |
"learning_rate": 2e-05,
|
| 1705 |
-
"loss": 0.
|
| 1706 |
-
"mean_token_accuracy": 0.
|
| 1707 |
"step": 1005
|
| 1708 |
},
|
| 1709 |
{
|
| 1710 |
-
"epoch":
|
| 1711 |
-
"grad_norm": 0.
|
| 1712 |
"learning_rate": 2e-05,
|
| 1713 |
-
"loss": 0.
|
| 1714 |
-
"mean_token_accuracy": 0.
|
| 1715 |
"step": 1010
|
| 1716 |
},
|
| 1717 |
{
|
| 1718 |
-
"epoch":
|
| 1719 |
-
"grad_norm": 0.
|
| 1720 |
"learning_rate": 2e-05,
|
| 1721 |
-
"loss": 0.
|
| 1722 |
-
"mean_token_accuracy": 0.
|
| 1723 |
"step": 1015
|
| 1724 |
},
|
| 1725 |
{
|
| 1726 |
-
"epoch":
|
| 1727 |
-
"grad_norm": 0.
|
| 1728 |
"learning_rate": 2e-05,
|
| 1729 |
-
"loss": 0.
|
| 1730 |
-
"mean_token_accuracy": 0.
|
| 1731 |
"step": 1020
|
| 1732 |
},
|
| 1733 |
{
|
| 1734 |
-
"epoch":
|
| 1735 |
-
"grad_norm": 0.
|
| 1736 |
"learning_rate": 2e-05,
|
| 1737 |
-
"loss": 0.
|
| 1738 |
-
"mean_token_accuracy": 0.
|
| 1739 |
"step": 1025
|
| 1740 |
},
|
| 1741 |
{
|
| 1742 |
-
"epoch":
|
| 1743 |
-
"
|
| 1744 |
-
"
|
| 1745 |
-
"loss": 0.966,
|
| 1746 |
-
"mean_token_accuracy": 0.7264703720808029,
|
| 1747 |
-
"step": 1030
|
| 1748 |
-
},
|
| 1749 |
-
{
|
| 1750 |
-
"epoch": 0.9968697327233326,
|
| 1751 |
-
"grad_norm": 0.45738255977630615,
|
| 1752 |
-
"learning_rate": 2e-05,
|
| 1753 |
-
"loss": 0.9525,
|
| 1754 |
-
"mean_token_accuracy": 0.7287515312433243,
|
| 1755 |
-
"step": 1035
|
| 1756 |
-
},
|
| 1757 |
-
{
|
| 1758 |
-
"epoch": 0.9997592102094871,
|
| 1759 |
-
"mean_token_accuracy": 0.7337231040000916,
|
| 1760 |
-
"step": 1038,
|
| 1761 |
"total_flos": 0.0,
|
| 1762 |
-
"train_loss":
|
| 1763 |
-
"train_runtime":
|
| 1764 |
-
"train_samples_per_second":
|
| 1765 |
-
"train_steps_per_second": 0.
|
| 1766 |
}
|
| 1767 |
],
|
| 1768 |
"logging_steps": 5,
|
| 1769 |
-
"max_steps":
|
| 1770 |
"num_input_tokens_seen": 0,
|
| 1771 |
-
"num_train_epochs":
|
| 1772 |
"save_steps": 500,
|
| 1773 |
"stateful_callbacks": {
|
| 1774 |
"TrainerControl": {
|
|
@@ -1783,7 +1767,7 @@
|
|
| 1783 |
}
|
| 1784 |
},
|
| 1785 |
"total_flos": 0.0,
|
| 1786 |
-
"train_batch_size":
|
| 1787 |
"trial_name": null,
|
| 1788 |
"trial_params": null
|
| 1789 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 2.9989078995267566,
|
| 5 |
"eval_steps": 100,
|
| 6 |
+
"global_step": 1029,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
+
"epoch": 0.014561339643247179,
|
| 13 |
+
"grad_norm": 37.66851806640625,
|
| 14 |
+
"learning_rate": 6.94512199751671e-06,
|
| 15 |
+
"loss": 2.2001,
|
| 16 |
+
"mean_token_accuracy": 0.5431830704212188,
|
| 17 |
"step": 5
|
| 18 |
},
|
| 19 |
{
|
| 20 |
+
"epoch": 0.029122679286494358,
|
| 21 |
+
"grad_norm": 35.93154525756836,
|
| 22 |
+
"learning_rate": 9.936223234807016e-06,
|
| 23 |
+
"loss": 2.0917,
|
| 24 |
+
"mean_token_accuracy": 0.5551130786538124,
|
| 25 |
"step": 10
|
| 26 |
},
|
| 27 |
{
|
| 28 |
+
"epoch": 0.04368401892974154,
|
| 29 |
+
"grad_norm": 11.33205509185791,
|
| 30 |
+
"learning_rate": 1.1685905294482498e-05,
|
| 31 |
+
"loss": 1.6615,
|
| 32 |
+
"mean_token_accuracy": 0.6157955378293991,
|
| 33 |
"step": 15
|
| 34 |
},
|
| 35 |
{
|
| 36 |
+
"epoch": 0.058245358572988716,
|
| 37 |
+
"grad_norm": 0.5927252769470215,
|
| 38 |
+
"learning_rate": 1.292732447209732e-05,
|
| 39 |
+
"loss": 1.4254,
|
| 40 |
+
"mean_token_accuracy": 0.6511392086744309,
|
| 41 |
"step": 20
|
| 42 |
},
|
| 43 |
{
|
| 44 |
+
"epoch": 0.0728066982162359,
|
| 45 |
+
"grad_norm": 0.420287549495697,
|
| 46 |
+
"learning_rate": 1.389024399503342e-05,
|
| 47 |
+
"loss": 1.3534,
|
| 48 |
+
"mean_token_accuracy": 0.6588488951325416,
|
| 49 |
"step": 25
|
| 50 |
},
|
| 51 |
{
|
| 52 |
+
"epoch": 0.08736803785948308,
|
| 53 |
+
"grad_norm": 0.3345254361629486,
|
| 54 |
+
"learning_rate": 1.4677006531772802e-05,
|
| 55 |
+
"loss": 1.3017,
|
| 56 |
+
"mean_token_accuracy": 0.6658114358782768,
|
| 57 |
"step": 30
|
| 58 |
},
|
| 59 |
{
|
| 60 |
+
"epoch": 0.10192937750273025,
|
| 61 |
+
"grad_norm": 0.28586578369140625,
|
| 62 |
+
"learning_rate": 1.5342204778396236e-05,
|
| 63 |
+
"loss": 1.2668,
|
| 64 |
+
"mean_token_accuracy": 0.6703020080924034,
|
| 65 |
"step": 35
|
| 66 |
},
|
| 67 |
{
|
| 68 |
+
"epoch": 0.11649071714597743,
|
| 69 |
+
"grad_norm": 0.2802036702632904,
|
| 70 |
+
"learning_rate": 1.5918425709387625e-05,
|
| 71 |
+
"loss": 1.2384,
|
| 72 |
+
"mean_token_accuracy": 0.6751068279147148,
|
| 73 |
"step": 40
|
| 74 |
},
|
| 75 |
{
|
| 76 |
+
"epoch": 0.1310520567892246,
|
| 77 |
+
"grad_norm": 0.2390078604221344,
|
| 78 |
+
"learning_rate": 1.6426688591448284e-05,
|
| 79 |
+
"loss": 1.2159,
|
| 80 |
+
"mean_token_accuracy": 0.6786533042788505,
|
| 81 |
"step": 45
|
| 82 |
},
|
| 83 |
{
|
| 84 |
+
"epoch": 0.1456133964324718,
|
| 85 |
+
"grad_norm": 0.22002895176410675,
|
| 86 |
+
"learning_rate": 1.6881345232323726e-05,
|
| 87 |
+
"loss": 1.1892,
|
| 88 |
+
"mean_token_accuracy": 0.6840800851583481,
|
| 89 |
"step": 50
|
| 90 |
},
|
| 91 |
{
|
| 92 |
+
"epoch": 0.16017473607571897,
|
| 93 |
+
"grad_norm": 0.24121476709842682,
|
| 94 |
+
"learning_rate": 1.7292632192343935e-05,
|
| 95 |
+
"loss": 1.1866,
|
| 96 |
+
"mean_token_accuracy": 0.6831674978137017,
|
| 97 |
"step": 55
|
| 98 |
},
|
| 99 |
{
|
| 100 |
+
"epoch": 0.17473607571896616,
|
| 101 |
+
"grad_norm": 0.2153487652540207,
|
| 102 |
+
"learning_rate": 1.7668107769063104e-05,
|
| 103 |
+
"loss": 1.1414,
|
| 104 |
+
"mean_token_accuracy": 0.6937515258789062,
|
| 105 |
"step": 60
|
| 106 |
},
|
| 107 |
{
|
| 108 |
+
"epoch": 0.1892974153622133,
|
| 109 |
+
"grad_norm": 0.19544684886932373,
|
| 110 |
+
"learning_rate": 1.8013511816966716e-05,
|
| 111 |
+
"loss": 1.1364,
|
| 112 |
+
"mean_token_accuracy": 0.6935119941830635,
|
| 113 |
"step": 65
|
| 114 |
},
|
| 115 |
{
|
| 116 |
+
"epoch": 0.2038587550054605,
|
| 117 |
+
"grad_norm": 0.20832200348377228,
|
| 118 |
+
"learning_rate": 1.8333306015686545e-05,
|
| 119 |
+
"loss": 1.1522,
|
| 120 |
+
"mean_token_accuracy": 0.6890207245945931,
|
| 121 |
"step": 70
|
| 122 |
},
|
| 123 |
{
|
| 124 |
+
"epoch": 0.21842009464870768,
|
| 125 |
+
"grad_norm": 0.20898529887199402,
|
| 126 |
+
"learning_rate": 1.8631027291999205e-05,
|
| 127 |
+
"loss": 1.1359,
|
| 128 |
+
"mean_token_accuracy": 0.6920820102095604,
|
| 129 |
"step": 75
|
| 130 |
},
|
| 131 |
{
|
| 132 |
+
"epoch": 0.23298143429195486,
|
| 133 |
+
"grad_norm": 0.21433797478675842,
|
| 134 |
+
"learning_rate": 1.890952694667793e-05,
|
| 135 |
+
"loss": 1.1177,
|
| 136 |
+
"mean_token_accuracy": 0.6967326954007149,
|
| 137 |
"step": 80
|
| 138 |
},
|
| 139 |
{
|
| 140 |
+
"epoch": 0.24754277393520205,
|
| 141 |
+
"grad_norm": 0.19631509482860565,
|
| 142 |
+
"learning_rate": 1.9171137159358747e-05,
|
| 143 |
+
"loss": 1.1084,
|
| 144 |
+
"mean_token_accuracy": 0.6984182327985764,
|
| 145 |
"step": 85
|
| 146 |
},
|
| 147 |
{
|
| 148 |
+
"epoch": 0.2621041135784492,
|
| 149 |
+
"grad_norm": 0.18118107318878174,
|
| 150 |
+
"learning_rate": 1.941778982873859e-05,
|
| 151 |
+
"loss": 1.1085,
|
| 152 |
+
"mean_token_accuracy": 0.6977160558104515,
|
| 153 |
"step": 90
|
| 154 |
},
|
| 155 |
{
|
| 156 |
+
"epoch": 0.2766654532216964,
|
| 157 |
+
"grad_norm": 0.20939502120018005,
|
| 158 |
+
"learning_rate": 1.965110323889735e-05,
|
| 159 |
+
"loss": 1.1065,
|
| 160 |
+
"mean_token_accuracy": 0.6980448961257935,
|
| 161 |
"step": 95
|
| 162 |
},
|
| 163 |
{
|
| 164 |
+
"epoch": 0.2912267928649436,
|
| 165 |
+
"grad_norm": 0.1910240799188614,
|
| 166 |
+
"learning_rate": 1.987244646961403e-05,
|
| 167 |
+
"loss": 1.098,
|
| 168 |
+
"mean_token_accuracy": 0.6989446982741356,
|
| 169 |
"step": 100
|
| 170 |
},
|
| 171 |
{
|
| 172 |
+
"epoch": 0.2912267928649436,
|
| 173 |
+
"eval_loss": 1.1112689971923828,
|
| 174 |
+
"eval_mean_token_accuracy": 0.6931903600692749,
|
| 175 |
+
"eval_runtime": 13.1717,
|
| 176 |
+
"eval_samples_per_second": 5.011,
|
| 177 |
+
"eval_steps_per_second": 0.38,
|
| 178 |
"step": 100
|
| 179 |
},
|
| 180 |
{
|
| 181 |
+
"epoch": 0.30578813250819076,
|
| 182 |
+
"grad_norm": 0.19251097738742828,
|
| 183 |
"learning_rate": 2e-05,
|
| 184 |
+
"loss": 1.0873,
|
| 185 |
+
"mean_token_accuracy": 0.7030002444982528,
|
| 186 |
"step": 105
|
| 187 |
},
|
| 188 |
{
|
| 189 |
+
"epoch": 0.32034947215143794,
|
| 190 |
+
"grad_norm": 0.1976984441280365,
|
| 191 |
"learning_rate": 2e-05,
|
| 192 |
+
"loss": 1.0954,
|
| 193 |
+
"mean_token_accuracy": 0.6988905444741249,
|
| 194 |
"step": 110
|
| 195 |
},
|
| 196 |
{
|
| 197 |
+
"epoch": 0.3349108117946851,
|
| 198 |
+
"grad_norm": 0.18475359678268433,
|
| 199 |
"learning_rate": 2e-05,
|
| 200 |
+
"loss": 1.0734,
|
| 201 |
+
"mean_token_accuracy": 0.7050403520464897,
|
| 202 |
"step": 115
|
| 203 |
},
|
| 204 |
{
|
| 205 |
+
"epoch": 0.3494721514379323,
|
| 206 |
+
"grad_norm": 0.21479089558124542,
|
| 207 |
"learning_rate": 2e-05,
|
| 208 |
+
"loss": 1.0792,
|
| 209 |
+
"mean_token_accuracy": 0.7049864172935486,
|
| 210 |
"step": 120
|
| 211 |
},
|
| 212 |
{
|
| 213 |
+
"epoch": 0.3640334910811795,
|
| 214 |
+
"grad_norm": 0.20598657429218292,
|
| 215 |
"learning_rate": 2e-05,
|
| 216 |
+
"loss": 1.0853,
|
| 217 |
+
"mean_token_accuracy": 0.7013437002897263,
|
| 218 |
"step": 125
|
| 219 |
},
|
| 220 |
{
|
| 221 |
+
"epoch": 0.3785948307244266,
|
| 222 |
+
"grad_norm": 0.2035069763660431,
|
| 223 |
"learning_rate": 2e-05,
|
| 224 |
+
"loss": 1.0758,
|
| 225 |
+
"mean_token_accuracy": 0.7027781993150711,
|
| 226 |
"step": 130
|
| 227 |
},
|
| 228 |
{
|
| 229 |
+
"epoch": 0.3931561703676738,
|
| 230 |
+
"grad_norm": 0.20786674320697784,
|
| 231 |
"learning_rate": 2e-05,
|
| 232 |
+
"loss": 1.086,
|
| 233 |
+
"mean_token_accuracy": 0.7012925714254379,
|
| 234 |
"step": 135
|
| 235 |
},
|
| 236 |
{
|
| 237 |
+
"epoch": 0.407717510010921,
|
| 238 |
+
"grad_norm": 0.19930203258991241,
|
| 239 |
"learning_rate": 2e-05,
|
| 240 |
+
"loss": 1.0819,
|
| 241 |
+
"mean_token_accuracy": 0.7024676471948623,
|
| 242 |
"step": 140
|
| 243 |
},
|
| 244 |
{
|
| 245 |
+
"epoch": 0.4222788496541682,
|
| 246 |
+
"grad_norm": 0.2310095727443695,
|
| 247 |
"learning_rate": 2e-05,
|
| 248 |
+
"loss": 1.0658,
|
| 249 |
+
"mean_token_accuracy": 0.7064277902245522,
|
| 250 |
"step": 145
|
| 251 |
},
|
| 252 |
{
|
| 253 |
+
"epoch": 0.43684018929741536,
|
| 254 |
+
"grad_norm": 0.2163601964712143,
|
| 255 |
"learning_rate": 2e-05,
|
| 256 |
+
"loss": 1.0741,
|
| 257 |
+
"mean_token_accuracy": 0.7032444015145302,
|
| 258 |
"step": 150
|
| 259 |
},
|
| 260 |
{
|
| 261 |
+
"epoch": 0.45140152894066254,
|
| 262 |
+
"grad_norm": 0.22852301597595215,
|
| 263 |
"learning_rate": 2e-05,
|
| 264 |
+
"loss": 1.0647,
|
| 265 |
+
"mean_token_accuracy": 0.7054958015680313,
|
| 266 |
"step": 155
|
| 267 |
},
|
| 268 |
{
|
| 269 |
+
"epoch": 0.46596286858390973,
|
| 270 |
+
"grad_norm": 0.21535199880599976,
|
| 271 |
"learning_rate": 2e-05,
|
| 272 |
+
"loss": 1.0753,
|
| 273 |
+
"mean_token_accuracy": 0.7024859562516212,
|
| 274 |
"step": 160
|
| 275 |
},
|
| 276 |
{
|
| 277 |
+
"epoch": 0.4805242082271569,
|
| 278 |
+
"grad_norm": 0.1874823421239853,
|
| 279 |
"learning_rate": 2e-05,
|
| 280 |
+
"loss": 1.0466,
|
| 281 |
+
"mean_token_accuracy": 0.710732813179493,
|
| 282 |
"step": 165
|
| 283 |
},
|
| 284 |
{
|
| 285 |
+
"epoch": 0.4950855478704041,
|
| 286 |
+
"grad_norm": 0.17631468176841736,
|
| 287 |
"learning_rate": 2e-05,
|
| 288 |
+
"loss": 1.0599,
|
| 289 |
+
"mean_token_accuracy": 0.7064499124884606,
|
| 290 |
"step": 170
|
| 291 |
},
|
| 292 |
{
|
| 293 |
+
"epoch": 0.5096468875136513,
|
| 294 |
+
"grad_norm": 0.20174729824066162,
|
| 295 |
"learning_rate": 2e-05,
|
| 296 |
+
"loss": 1.0447,
|
| 297 |
+
"mean_token_accuracy": 0.7100659251213074,
|
| 298 |
"step": 175
|
| 299 |
},
|
| 300 |
{
|
| 301 |
+
"epoch": 0.5242082271568984,
|
| 302 |
+
"grad_norm": 0.1740773767232895,
|
| 303 |
"learning_rate": 2e-05,
|
| 304 |
+
"loss": 1.0538,
|
| 305 |
+
"mean_token_accuracy": 0.7067421570420265,
|
| 306 |
"step": 180
|
| 307 |
},
|
| 308 |
{
|
| 309 |
+
"epoch": 0.5387695668001456,
|
| 310 |
+
"grad_norm": 0.18247714638710022,
|
| 311 |
"learning_rate": 2e-05,
|
| 312 |
+
"loss": 1.0487,
|
| 313 |
+
"mean_token_accuracy": 0.7085062861442566,
|
| 314 |
"step": 185
|
| 315 |
},
|
| 316 |
{
|
| 317 |
+
"epoch": 0.5533309064433928,
|
| 318 |
+
"grad_norm": 0.19148823618888855,
|
| 319 |
"learning_rate": 2e-05,
|
| 320 |
+
"loss": 1.0597,
|
| 321 |
+
"mean_token_accuracy": 0.7060058280825615,
|
| 322 |
"step": 190
|
| 323 |
},
|
| 324 |
{
|
| 325 |
+
"epoch": 0.56789224608664,
|
| 326 |
+
"grad_norm": 0.1849949061870575,
|
| 327 |
"learning_rate": 2e-05,
|
| 328 |
+
"loss": 1.0422,
|
| 329 |
+
"mean_token_accuracy": 0.7095287501811981,
|
| 330 |
"step": 195
|
| 331 |
},
|
| 332 |
{
|
| 333 |
+
"epoch": 0.5824535857298871,
|
| 334 |
+
"grad_norm": 0.18132270872592926,
|
| 335 |
"learning_rate": 2e-05,
|
| 336 |
+
"loss": 1.0468,
|
| 337 |
+
"mean_token_accuracy": 0.7086604192852974,
|
| 338 |
"step": 200
|
| 339 |
},
|
| 340 |
{
|
| 341 |
+
"epoch": 0.5824535857298871,
|
| 342 |
+
"eval_loss": 1.0518466234207153,
|
| 343 |
+
"eval_mean_token_accuracy": 0.7042254090309144,
|
| 344 |
+
"eval_runtime": 9.0776,
|
| 345 |
+
"eval_samples_per_second": 7.271,
|
| 346 |
+
"eval_steps_per_second": 0.551,
|
| 347 |
"step": 200
|
| 348 |
},
|
| 349 |
{
|
| 350 |
+
"epoch": 0.5970149253731343,
|
| 351 |
+
"grad_norm": 0.18448956310749054,
|
| 352 |
"learning_rate": 2e-05,
|
| 353 |
+
"loss": 1.037,
|
| 354 |
+
"mean_token_accuracy": 0.7115538448095322,
|
| 355 |
"step": 205
|
| 356 |
},
|
| 357 |
{
|
| 358 |
+
"epoch": 0.6115762650163815,
|
| 359 |
+
"grad_norm": 0.18751472234725952,
|
| 360 |
"learning_rate": 2e-05,
|
| 361 |
+
"loss": 1.0445,
|
| 362 |
+
"mean_token_accuracy": 0.7096409142017365,
|
| 363 |
"step": 210
|
| 364 |
},
|
| 365 |
{
|
| 366 |
+
"epoch": 0.6261376046596286,
|
| 367 |
+
"grad_norm": 0.1885630488395691,
|
| 368 |
"learning_rate": 2e-05,
|
| 369 |
+
"loss": 1.03,
|
| 370 |
+
"mean_token_accuracy": 0.7122749090194702,
|
| 371 |
"step": 215
|
| 372 |
},
|
| 373 |
{
|
| 374 |
+
"epoch": 0.6406989443028759,
|
| 375 |
+
"grad_norm": 0.16414892673492432,
|
| 376 |
"learning_rate": 2e-05,
|
| 377 |
+
"loss": 1.0203,
|
| 378 |
+
"mean_token_accuracy": 0.7152423396706581,
|
| 379 |
"step": 220
|
| 380 |
},
|
| 381 |
{
|
| 382 |
+
"epoch": 0.655260283946123,
|
| 383 |
+
"grad_norm": 0.19612085819244385,
|
| 384 |
"learning_rate": 2e-05,
|
| 385 |
+
"loss": 1.0325,
|
| 386 |
+
"mean_token_accuracy": 0.7118910998106003,
|
| 387 |
"step": 225
|
| 388 |
},
|
| 389 |
{
|
| 390 |
+
"epoch": 0.6698216235893703,
|
| 391 |
+
"grad_norm": 0.21399515867233276,
|
| 392 |
"learning_rate": 2e-05,
|
| 393 |
+
"loss": 1.0264,
|
| 394 |
+
"mean_token_accuracy": 0.7130562499165535,
|
| 395 |
"step": 230
|
| 396 |
},
|
| 397 |
{
|
| 398 |
+
"epoch": 0.6843829632326174,
|
| 399 |
+
"grad_norm": 0.19661396741867065,
|
| 400 |
"learning_rate": 2e-05,
|
| 401 |
+
"loss": 1.043,
|
| 402 |
+
"mean_token_accuracy": 0.7081087440252304,
|
| 403 |
"step": 235
|
| 404 |
},
|
| 405 |
{
|
| 406 |
+
"epoch": 0.6989443028758646,
|
| 407 |
+
"grad_norm": 0.1965187042951584,
|
| 408 |
"learning_rate": 2e-05,
|
| 409 |
+
"loss": 1.0232,
|
| 410 |
+
"mean_token_accuracy": 0.7128947824239731,
|
| 411 |
"step": 240
|
| 412 |
},
|
| 413 |
{
|
| 414 |
+
"epoch": 0.7135056425191117,
|
| 415 |
+
"grad_norm": 0.19409024715423584,
|
| 416 |
"learning_rate": 2e-05,
|
| 417 |
+
"loss": 1.0328,
|
| 418 |
+
"mean_token_accuracy": 0.7117997944355011,
|
| 419 |
"step": 245
|
| 420 |
},
|
| 421 |
{
|
| 422 |
+
"epoch": 0.728066982162359,
|
| 423 |
+
"grad_norm": 0.19648292660713196,
|
| 424 |
"learning_rate": 2e-05,
|
| 425 |
+
"loss": 1.038,
|
| 426 |
+
"mean_token_accuracy": 0.7104994520545006,
|
| 427 |
"step": 250
|
| 428 |
},
|
| 429 |
{
|
| 430 |
+
"epoch": 0.7426283218056061,
|
| 431 |
+
"grad_norm": 0.16873674094676971,
|
| 432 |
"learning_rate": 2e-05,
|
| 433 |
+
"loss": 1.0204,
|
| 434 |
+
"mean_token_accuracy": 0.7144693836569787,
|
| 435 |
"step": 255
|
| 436 |
},
|
| 437 |
{
|
| 438 |
+
"epoch": 0.7571896614488532,
|
| 439 |
+
"grad_norm": 0.196326345205307,
|
| 440 |
"learning_rate": 2e-05,
|
| 441 |
+
"loss": 1.0213,
|
| 442 |
+
"mean_token_accuracy": 0.7139619678258896,
|
| 443 |
"step": 260
|
| 444 |
},
|
| 445 |
{
|
| 446 |
+
"epoch": 0.7717510010921005,
|
| 447 |
+
"grad_norm": 0.2151579111814499,
|
| 448 |
"learning_rate": 2e-05,
|
| 449 |
+
"loss": 1.0218,
|
| 450 |
+
"mean_token_accuracy": 0.7128418371081352,
|
| 451 |
"step": 265
|
| 452 |
},
|
| 453 |
{
|
| 454 |
+
"epoch": 0.7863123407353476,
|
| 455 |
+
"grad_norm": 0.17913345992565155,
|
| 456 |
"learning_rate": 2e-05,
|
| 457 |
+
"loss": 1.0265,
|
| 458 |
+
"mean_token_accuracy": 0.7119010165333748,
|
| 459 |
"step": 270
|
| 460 |
},
|
| 461 |
{
|
| 462 |
+
"epoch": 0.8008736803785949,
|
| 463 |
+
"grad_norm": 0.1754622459411621,
|
| 464 |
"learning_rate": 2e-05,
|
| 465 |
+
"loss": 1.0154,
|
| 466 |
+
"mean_token_accuracy": 0.7142972201108932,
|
| 467 |
"step": 275
|
| 468 |
},
|
| 469 |
{
|
| 470 |
+
"epoch": 0.815435020021842,
|
| 471 |
+
"grad_norm": 0.19478586316108704,
|
| 472 |
"learning_rate": 2e-05,
|
| 473 |
+
"loss": 1.0302,
|
| 474 |
+
"mean_token_accuracy": 0.711237944662571,
|
| 475 |
"step": 280
|
| 476 |
},
|
| 477 |
{
|
| 478 |
+
"epoch": 0.8299963596650892,
|
| 479 |
+
"grad_norm": 0.18054774403572083,
|
| 480 |
"learning_rate": 2e-05,
|
| 481 |
+
"loss": 1.0102,
|
| 482 |
+
"mean_token_accuracy": 0.7163647577166558,
|
| 483 |
"step": 285
|
| 484 |
},
|
| 485 |
{
|
| 486 |
+
"epoch": 0.8445576993083364,
|
| 487 |
+
"grad_norm": 0.2034561187028885,
|
| 488 |
"learning_rate": 2e-05,
|
| 489 |
+
"loss": 0.9997,
|
| 490 |
+
"mean_token_accuracy": 0.7193443968892097,
|
| 491 |
"step": 290
|
| 492 |
},
|
| 493 |
{
|
| 494 |
+
"epoch": 0.8591190389515836,
|
| 495 |
+
"grad_norm": 0.202008917927742,
|
| 496 |
"learning_rate": 2e-05,
|
| 497 |
+
"loss": 1.0059,
|
| 498 |
+
"mean_token_accuracy": 0.718295231461525,
|
| 499 |
"step": 295
|
| 500 |
},
|
| 501 |
{
|
| 502 |
+
"epoch": 0.8736803785948307,
|
| 503 |
+
"grad_norm": 0.207170769572258,
|
| 504 |
"learning_rate": 2e-05,
|
| 505 |
+
"loss": 1.0078,
|
| 506 |
+
"mean_token_accuracy": 0.7157909572124481,
|
| 507 |
"step": 300
|
| 508 |
},
|
| 509 |
{
|
| 510 |
+
"epoch": 0.8736803785948307,
|
| 511 |
+
"eval_loss": 1.019649624824524,
|
| 512 |
+
"eval_mean_token_accuracy": 0.7100205540657043,
|
| 513 |
+
"eval_runtime": 9.0111,
|
| 514 |
+
"eval_samples_per_second": 7.324,
|
| 515 |
+
"eval_steps_per_second": 0.555,
|
| 516 |
"step": 300
|
| 517 |
},
|
| 518 |
{
|
| 519 |
+
"epoch": 0.8882417182380778,
|
| 520 |
+
"grad_norm": 0.21266800165176392,
|
| 521 |
"learning_rate": 2e-05,
|
| 522 |
+
"loss": 1.0074,
|
| 523 |
+
"mean_token_accuracy": 0.7171560257673264,
|
| 524 |
"step": 305
|
| 525 |
},
|
| 526 |
{
|
| 527 |
+
"epoch": 0.9028030578813251,
|
| 528 |
+
"grad_norm": 0.19650128483772278,
|
| 529 |
"learning_rate": 2e-05,
|
| 530 |
+
"loss": 1.0142,
|
| 531 |
+
"mean_token_accuracy": 0.7148646369576455,
|
| 532 |
"step": 310
|
| 533 |
},
|
| 534 |
{
|
| 535 |
+
"epoch": 0.9173643975245722,
|
| 536 |
+
"grad_norm": 0.22513243556022644,
|
| 537 |
"learning_rate": 2e-05,
|
| 538 |
+
"loss": 0.9986,
|
| 539 |
+
"mean_token_accuracy": 0.7187789976596832,
|
| 540 |
"step": 315
|
| 541 |
},
|
| 542 |
{
|
| 543 |
+
"epoch": 0.9319257371678195,
|
| 544 |
+
"grad_norm": 0.21226869523525238,
|
| 545 |
"learning_rate": 2e-05,
|
| 546 |
+
"loss": 1.0095,
|
| 547 |
+
"mean_token_accuracy": 0.7166241943836212,
|
| 548 |
"step": 320
|
| 549 |
},
|
| 550 |
{
|
| 551 |
+
"epoch": 0.9464870768110666,
|
| 552 |
+
"grad_norm": 0.20326650142669678,
|
| 553 |
"learning_rate": 2e-05,
|
| 554 |
+
"loss": 1.0193,
|
| 555 |
+
"mean_token_accuracy": 0.7132607430219651,
|
| 556 |
"step": 325
|
| 557 |
},
|
| 558 |
{
|
| 559 |
+
"epoch": 0.9610484164543138,
|
| 560 |
+
"grad_norm": 0.2005230039358139,
|
| 561 |
"learning_rate": 2e-05,
|
| 562 |
+
"loss": 0.9969,
|
| 563 |
+
"mean_token_accuracy": 0.7200944602489472,
|
| 564 |
"step": 330
|
| 565 |
},
|
| 566 |
{
|
| 567 |
+
"epoch": 0.975609756097561,
|
| 568 |
+
"grad_norm": 0.18904553353786469,
|
| 569 |
"learning_rate": 2e-05,
|
| 570 |
+
"loss": 1.0021,
|
| 571 |
+
"mean_token_accuracy": 0.7175276219844818,
|
| 572 |
"step": 335
|
| 573 |
},
|
| 574 |
{
|
| 575 |
+
"epoch": 0.9901710957408082,
|
| 576 |
+
"grad_norm": 0.19811247289180756,
|
| 577 |
"learning_rate": 2e-05,
|
| 578 |
+
"loss": 0.9993,
|
| 579 |
+
"mean_token_accuracy": 0.7185588896274566,
|
| 580 |
"step": 340
|
| 581 |
},
|
| 582 |
{
|
| 583 |
+
"epoch": 1.0058245358572988,
|
| 584 |
+
"grad_norm": 0.1984570026397705,
|
| 585 |
"learning_rate": 2e-05,
|
| 586 |
+
"loss": 1.0395,
|
| 587 |
+
"mean_token_accuracy": 0.719094998779751,
|
| 588 |
"step": 345
|
| 589 |
},
|
| 590 |
{
|
| 591 |
+
"epoch": 1.0203858755005462,
|
| 592 |
+
"grad_norm": 0.19051611423492432,
|
| 593 |
"learning_rate": 2e-05,
|
| 594 |
+
"loss": 0.9733,
|
| 595 |
+
"mean_token_accuracy": 0.724665792286396,
|
| 596 |
"step": 350
|
| 597 |
},
|
| 598 |
{
|
| 599 |
+
"epoch": 1.0349472151437933,
|
| 600 |
+
"grad_norm": 0.17580822110176086,
|
| 601 |
"learning_rate": 2e-05,
|
| 602 |
+
"loss": 0.9811,
|
| 603 |
+
"mean_token_accuracy": 0.7220699548721313,
|
| 604 |
"step": 355
|
| 605 |
},
|
| 606 |
{
|
| 607 |
+
"epoch": 1.0495085547870404,
|
| 608 |
+
"grad_norm": 0.18404695391654968,
|
| 609 |
"learning_rate": 2e-05,
|
| 610 |
+
"loss": 0.9757,
|
| 611 |
+
"mean_token_accuracy": 0.7235838070511817,
|
| 612 |
"step": 360
|
| 613 |
},
|
| 614 |
{
|
| 615 |
+
"epoch": 1.0640698944302875,
|
| 616 |
+
"grad_norm": 0.19152525067329407,
|
| 617 |
"learning_rate": 2e-05,
|
| 618 |
+
"loss": 0.9828,
|
| 619 |
+
"mean_token_accuracy": 0.7204630091786385,
|
| 620 |
"step": 365
|
| 621 |
},
|
| 622 |
{
|
| 623 |
+
"epoch": 1.0786312340735347,
|
| 624 |
+
"grad_norm": 0.19752167165279388,
|
| 625 |
"learning_rate": 2e-05,
|
| 626 |
+
"loss": 0.9974,
|
| 627 |
+
"mean_token_accuracy": 0.7173234552145005,
|
| 628 |
"step": 370
|
| 629 |
},
|
| 630 |
{
|
| 631 |
+
"epoch": 1.093192573716782,
|
| 632 |
+
"grad_norm": 0.1899857074022293,
|
| 633 |
"learning_rate": 2e-05,
|
| 634 |
+
"loss": 0.979,
|
| 635 |
+
"mean_token_accuracy": 0.7223904326558113,
|
| 636 |
"step": 375
|
| 637 |
},
|
| 638 |
{
|
| 639 |
+
"epoch": 1.1077539133600292,
|
| 640 |
+
"grad_norm": 0.19329330325126648,
|
| 641 |
"learning_rate": 2e-05,
|
| 642 |
+
"loss": 0.9859,
|
| 643 |
+
"mean_token_accuracy": 0.7198724135756492,
|
| 644 |
"step": 380
|
| 645 |
},
|
| 646 |
{
|
| 647 |
+
"epoch": 1.1223152530032763,
|
| 648 |
+
"grad_norm": 0.18269023299217224,
|
| 649 |
"learning_rate": 2e-05,
|
| 650 |
+
"loss": 0.9642,
|
| 651 |
+
"mean_token_accuracy": 0.727630938589573,
|
| 652 |
"step": 385
|
| 653 |
},
|
| 654 |
{
|
| 655 |
+
"epoch": 1.1368765926465234,
|
| 656 |
+
"grad_norm": 0.1874740570783615,
|
| 657 |
"learning_rate": 2e-05,
|
| 658 |
+
"loss": 0.9778,
|
| 659 |
+
"mean_token_accuracy": 0.722543029487133,
|
| 660 |
"step": 390
|
| 661 |
},
|
| 662 |
{
|
| 663 |
+
"epoch": 1.1514379322897708,
|
| 664 |
+
"grad_norm": 0.174119770526886,
|
| 665 |
"learning_rate": 2e-05,
|
| 666 |
+
"loss": 0.9568,
|
| 667 |
+
"mean_token_accuracy": 0.7279941365122795,
|
| 668 |
"step": 395
|
| 669 |
},
|
| 670 |
{
|
| 671 |
+
"epoch": 1.1659992719330179,
|
| 672 |
+
"grad_norm": 0.1855311542749405,
|
| 673 |
"learning_rate": 2e-05,
|
| 674 |
+
"loss": 0.9688,
|
| 675 |
+
"mean_token_accuracy": 0.7247176736593246,
|
| 676 |
"step": 400
|
| 677 |
},
|
| 678 |
{
|
| 679 |
+
"epoch": 1.1659992719330179,
|
| 680 |
+
"eval_loss": 0.9988163113594055,
|
| 681 |
+
"eval_mean_token_accuracy": 0.7142685413360595,
|
| 682 |
+
"eval_runtime": 9.0627,
|
| 683 |
+
"eval_samples_per_second": 7.283,
|
| 684 |
+
"eval_steps_per_second": 0.552,
|
| 685 |
"step": 400
|
| 686 |
},
|
| 687 |
{
|
| 688 |
+
"epoch": 1.180560611576265,
|
| 689 |
+
"grad_norm": 0.2153940200805664,
|
| 690 |
"learning_rate": 2e-05,
|
| 691 |
+
"loss": 0.9744,
|
| 692 |
+
"mean_token_accuracy": 0.723403736948967,
|
| 693 |
"step": 405
|
| 694 |
},
|
| 695 |
{
|
| 696 |
+
"epoch": 1.1951219512195121,
|
| 697 |
+
"grad_norm": 0.20522184669971466,
|
| 698 |
"learning_rate": 2e-05,
|
| 699 |
+
"loss": 0.9725,
|
| 700 |
+
"mean_token_accuracy": 0.7236453905701637,
|
| 701 |
"step": 410
|
| 702 |
},
|
| 703 |
{
|
| 704 |
+
"epoch": 1.2096832908627593,
|
| 705 |
+
"grad_norm": 0.1979447901248932,
|
| 706 |
"learning_rate": 2e-05,
|
| 707 |
+
"loss": 0.978,
|
| 708 |
+
"mean_token_accuracy": 0.7225941598415375,
|
| 709 |
"step": 415
|
| 710 |
},
|
| 711 |
{
|
| 712 |
+
"epoch": 1.2242446305060066,
|
| 713 |
+
"grad_norm": 0.19050033390522003,
|
| 714 |
"learning_rate": 2e-05,
|
| 715 |
+
"loss": 0.9645,
|
| 716 |
+
"mean_token_accuracy": 0.7253967747092247,
|
| 717 |
"step": 420
|
| 718 |
},
|
| 719 |
{
|
| 720 |
+
"epoch": 1.2388059701492538,
|
| 721 |
+
"grad_norm": 0.19461773335933685,
|
| 722 |
"learning_rate": 2e-05,
|
| 723 |
+
"loss": 0.9469,
|
| 724 |
+
"mean_token_accuracy": 0.7304514393210411,
|
| 725 |
"step": 425
|
| 726 |
},
|
| 727 |
{
|
| 728 |
+
"epoch": 1.2533673097925009,
|
| 729 |
+
"grad_norm": 0.19699646532535553,
|
| 730 |
"learning_rate": 2e-05,
|
| 731 |
+
"loss": 0.9661,
|
| 732 |
+
"mean_token_accuracy": 0.7242850378155709,
|
| 733 |
"step": 430
|
| 734 |
},
|
| 735 |
{
|
| 736 |
+
"epoch": 1.267928649435748,
|
| 737 |
+
"grad_norm": 0.1707252860069275,
|
| 738 |
"learning_rate": 2e-05,
|
| 739 |
+
"loss": 0.9677,
|
| 740 |
+
"mean_token_accuracy": 0.7248802036046982,
|
| 741 |
"step": 435
|
| 742 |
},
|
| 743 |
{
|
| 744 |
+
"epoch": 1.2824899890789951,
|
| 745 |
+
"grad_norm": 0.191995769739151,
|
| 746 |
"learning_rate": 2e-05,
|
| 747 |
+
"loss": 0.9526,
|
| 748 |
+
"mean_token_accuracy": 0.7288147822022438,
|
| 749 |
"step": 440
|
| 750 |
},
|
| 751 |
{
|
| 752 |
+
"epoch": 1.2970513287222425,
|
| 753 |
+
"grad_norm": 0.18631628155708313,
|
| 754 |
"learning_rate": 2e-05,
|
| 755 |
+
"loss": 0.9705,
|
| 756 |
+
"mean_token_accuracy": 0.723578467965126,
|
| 757 |
"step": 445
|
| 758 |
},
|
| 759 |
{
|
| 760 |
+
"epoch": 1.3116126683654896,
|
| 761 |
+
"grad_norm": 0.18883132934570312,
|
| 762 |
"learning_rate": 2e-05,
|
| 763 |
+
"loss": 0.963,
|
| 764 |
+
"mean_token_accuracy": 0.7254890978336335,
|
| 765 |
"step": 450
|
| 766 |
},
|
| 767 |
{
|
| 768 |
+
"epoch": 1.3261740080087367,
|
| 769 |
+
"grad_norm": 0.19409437477588654,
|
| 770 |
"learning_rate": 2e-05,
|
| 771 |
+
"loss": 0.9664,
|
| 772 |
+
"mean_token_accuracy": 0.7247894033789635,
|
| 773 |
"step": 455
|
| 774 |
},
|
| 775 |
{
|
| 776 |
+
"epoch": 1.340735347651984,
|
| 777 |
+
"grad_norm": 0.17440907657146454,
|
| 778 |
"learning_rate": 2e-05,
|
| 779 |
+
"loss": 0.9539,
|
| 780 |
+
"mean_token_accuracy": 0.7279758274555206,
|
| 781 |
"step": 460
|
| 782 |
},
|
| 783 |
{
|
| 784 |
+
"epoch": 1.3552966872952312,
|
| 785 |
+
"grad_norm": 0.19074754416942596,
|
| 786 |
"learning_rate": 2e-05,
|
| 787 |
+
"loss": 0.9577,
|
| 788 |
+
"mean_token_accuracy": 0.726808387041092,
|
| 789 |
"step": 465
|
| 790 |
},
|
| 791 |
{
|
| 792 |
+
"epoch": 1.3698580269384784,
|
| 793 |
+
"grad_norm": 0.17158101499080658,
|
| 794 |
"learning_rate": 2e-05,
|
| 795 |
+
"loss": 0.9624,
|
| 796 |
+
"mean_token_accuracy": 0.7265920951962471,
|
| 797 |
"step": 470
|
| 798 |
},
|
| 799 |
{
|
| 800 |
+
"epoch": 1.3844193665817255,
|
| 801 |
+
"grad_norm": 0.18509717285633087,
|
| 802 |
"learning_rate": 2e-05,
|
| 803 |
+
"loss": 0.9537,
|
| 804 |
+
"mean_token_accuracy": 0.7284008353948593,
|
| 805 |
"step": 475
|
| 806 |
},
|
| 807 |
{
|
| 808 |
+
"epoch": 1.3989807062249726,
|
| 809 |
+
"grad_norm": 0.18281057476997375,
|
| 810 |
"learning_rate": 2e-05,
|
| 811 |
+
"loss": 0.9676,
|
| 812 |
+
"mean_token_accuracy": 0.7247360810637474,
|
| 813 |
"step": 480
|
| 814 |
},
|
| 815 |
{
|
| 816 |
+
"epoch": 1.41354204586822,
|
| 817 |
+
"grad_norm": 0.17993324995040894,
|
| 818 |
"learning_rate": 2e-05,
|
| 819 |
+
"loss": 0.9662,
|
| 820 |
+
"mean_token_accuracy": 0.7256203427910805,
|
| 821 |
"step": 485
|
| 822 |
},
|
| 823 |
{
|
| 824 |
+
"epoch": 1.428103385511467,
|
| 825 |
+
"grad_norm": 0.21310538053512573,
|
| 826 |
"learning_rate": 2e-05,
|
| 827 |
+
"loss": 0.9601,
|
| 828 |
+
"mean_token_accuracy": 0.7264375537633896,
|
| 829 |
"step": 490
|
| 830 |
},
|
| 831 |
{
|
| 832 |
+
"epoch": 1.4426647251547142,
|
| 833 |
+
"grad_norm": 0.18315577507019043,
|
| 834 |
"learning_rate": 2e-05,
|
| 835 |
+
"loss": 0.9434,
|
| 836 |
+
"mean_token_accuracy": 0.73150485008955,
|
| 837 |
"step": 495
|
| 838 |
},
|
| 839 |
{
|
| 840 |
+
"epoch": 1.4572260647979614,
|
| 841 |
+
"grad_norm": 0.19212491810321808,
|
| 842 |
"learning_rate": 2e-05,
|
| 843 |
+
"loss": 0.949,
|
| 844 |
+
"mean_token_accuracy": 0.7304915532469749,
|
| 845 |
"step": 500
|
| 846 |
},
|
| 847 |
{
|
| 848 |
+
"epoch": 1.4572260647979614,
|
| 849 |
+
"eval_loss": 0.9828361868858337,
|
| 850 |
+
"eval_mean_token_accuracy": 0.7173765182495118,
|
| 851 |
+
"eval_runtime": 9.0411,
|
| 852 |
+
"eval_samples_per_second": 7.3,
|
| 853 |
+
"eval_steps_per_second": 0.553,
|
| 854 |
"step": 500
|
| 855 |
},
|
| 856 |
{
|
| 857 |
+
"epoch": 1.4717874044412085,
|
| 858 |
+
"grad_norm": 0.17383301258087158,
|
| 859 |
"learning_rate": 2e-05,
|
| 860 |
+
"loss": 0.9673,
|
| 861 |
+
"mean_token_accuracy": 0.7242163941264153,
|
| 862 |
"step": 505
|
| 863 |
},
|
| 864 |
{
|
| 865 |
+
"epoch": 1.4863487440844558,
|
| 866 |
+
"grad_norm": 0.1923714131116867,
|
| 867 |
"learning_rate": 2e-05,
|
| 868 |
+
"loss": 0.9525,
|
| 869 |
+
"mean_token_accuracy": 0.7281391903758049,
|
| 870 |
"step": 510
|
| 871 |
},
|
| 872 |
{
|
| 873 |
+
"epoch": 1.500910083727703,
|
| 874 |
+
"grad_norm": 0.20028460025787354,
|
| 875 |
"learning_rate": 2e-05,
|
| 876 |
+
"loss": 0.9684,
|
| 877 |
+
"mean_token_accuracy": 0.723628830909729,
|
| 878 |
"step": 515
|
| 879 |
},
|
| 880 |
{
|
| 881 |
+
"epoch": 1.51547142337095,
|
| 882 |
+
"grad_norm": 0.20677019655704498,
|
| 883 |
"learning_rate": 2e-05,
|
| 884 |
+
"loss": 0.9802,
|
| 885 |
+
"mean_token_accuracy": 0.7201669454574585,
|
| 886 |
"step": 520
|
| 887 |
},
|
| 888 |
{
|
| 889 |
+
"epoch": 1.5300327630141974,
|
| 890 |
+
"grad_norm": 0.17966614663600922,
|
| 891 |
"learning_rate": 2e-05,
|
| 892 |
+
"loss": 0.9509,
|
| 893 |
+
"mean_token_accuracy": 0.7284548431634903,
|
| 894 |
"step": 525
|
| 895 |
},
|
| 896 |
{
|
| 897 |
+
"epoch": 1.5445941026574443,
|
| 898 |
+
"grad_norm": 0.19962792098522186,
|
| 899 |
"learning_rate": 2e-05,
|
| 900 |
+
"loss": 0.9556,
|
| 901 |
+
"mean_token_accuracy": 0.727879686653614,
|
| 902 |
"step": 530
|
| 903 |
},
|
| 904 |
{
|
| 905 |
+
"epoch": 1.5591554423006917,
|
| 906 |
+
"grad_norm": 0.1987488865852356,
|
| 907 |
"learning_rate": 2e-05,
|
| 908 |
+
"loss": 0.9441,
|
| 909 |
+
"mean_token_accuracy": 0.7310003876686096,
|
| 910 |
"step": 535
|
| 911 |
},
|
| 912 |
{
|
| 913 |
+
"epoch": 1.5737167819439388,
|
| 914 |
+
"grad_norm": 0.19578705728054047,
|
| 915 |
"learning_rate": 2e-05,
|
| 916 |
+
"loss": 0.9646,
|
| 917 |
+
"mean_token_accuracy": 0.7239905044436454,
|
| 918 |
"step": 540
|
| 919 |
},
|
| 920 |
{
|
| 921 |
+
"epoch": 1.588278121587186,
|
| 922 |
+
"grad_norm": 0.18124082684516907,
|
| 923 |
"learning_rate": 2e-05,
|
| 924 |
+
"loss": 0.9648,
|
| 925 |
+
"mean_token_accuracy": 0.7232480734586716,
|
| 926 |
"step": 545
|
| 927 |
},
|
| 928 |
{
|
| 929 |
+
"epoch": 1.6028394612304333,
|
| 930 |
+
"grad_norm": 0.18559084832668304,
|
| 931 |
"learning_rate": 2e-05,
|
| 932 |
+
"loss": 0.9374,
|
| 933 |
+
"mean_token_accuracy": 0.7308333814144135,
|
| 934 |
"step": 550
|
| 935 |
},
|
| 936 |
{
|
| 937 |
+
"epoch": 1.6174008008736804,
|
| 938 |
+
"grad_norm": 0.20631375908851624,
|
| 939 |
"learning_rate": 2e-05,
|
| 940 |
+
"loss": 0.9618,
|
| 941 |
+
"mean_token_accuracy": 0.7258706241846085,
|
| 942 |
"step": 555
|
| 943 |
},
|
| 944 |
{
|
| 945 |
+
"epoch": 1.6319621405169276,
|
| 946 |
+
"grad_norm": 0.1777167022228241,
|
| 947 |
"learning_rate": 2e-05,
|
| 948 |
+
"loss": 0.9447,
|
| 949 |
+
"mean_token_accuracy": 0.7300238028168679,
|
| 950 |
"step": 560
|
| 951 |
},
|
| 952 |
{
|
| 953 |
+
"epoch": 1.6465234801601747,
|
| 954 |
+
"grad_norm": 0.17086541652679443,
|
| 955 |
"learning_rate": 2e-05,
|
| 956 |
+
"loss": 0.9396,
|
| 957 |
+
"mean_token_accuracy": 0.7313713192939758,
|
| 958 |
"step": 565
|
| 959 |
},
|
| 960 |
{
|
| 961 |
+
"epoch": 1.6610848198034218,
|
| 962 |
+
"grad_norm": 0.21147526800632477,
|
| 963 |
"learning_rate": 2e-05,
|
| 964 |
+
"loss": 0.9512,
|
| 965 |
+
"mean_token_accuracy": 0.7282726511359214,
|
| 966 |
"step": 570
|
| 967 |
},
|
| 968 |
{
|
| 969 |
+
"epoch": 1.6756461594466692,
|
| 970 |
+
"grad_norm": 0.1869203746318817,
|
| 971 |
"learning_rate": 2e-05,
|
| 972 |
+
"loss": 0.9442,
|
| 973 |
+
"mean_token_accuracy": 0.7296415269374847,
|
| 974 |
"step": 575
|
| 975 |
},
|
| 976 |
{
|
| 977 |
+
"epoch": 1.6902074990899163,
|
| 978 |
+
"grad_norm": 0.19998286664485931,
|
| 979 |
"learning_rate": 2e-05,
|
| 980 |
+
"loss": 0.9495,
|
| 981 |
+
"mean_token_accuracy": 0.7277965158224106,
|
| 982 |
"step": 580
|
| 983 |
},
|
| 984 |
{
|
| 985 |
+
"epoch": 1.7047688387331634,
|
| 986 |
+
"grad_norm": 0.1844228357076645,
|
| 987 |
"learning_rate": 2e-05,
|
| 988 |
+
"loss": 0.9535,
|
| 989 |
+
"mean_token_accuracy": 0.7272509470582008,
|
| 990 |
"step": 585
|
| 991 |
},
|
| 992 |
{
|
| 993 |
+
"epoch": 1.7193301783764108,
|
| 994 |
+
"grad_norm": 0.22390896081924438,
|
| 995 |
"learning_rate": 2e-05,
|
| 996 |
+
"loss": 0.9526,
|
| 997 |
+
"mean_token_accuracy": 0.7276457890868187,
|
| 998 |
"step": 590
|
| 999 |
},
|
| 1000 |
{
|
| 1001 |
+
"epoch": 1.7338915180196577,
|
| 1002 |
+
"grad_norm": 0.18321609497070312,
|
| 1003 |
"learning_rate": 2e-05,
|
| 1004 |
+
"loss": 0.9534,
|
| 1005 |
+
"mean_token_accuracy": 0.7267900720238686,
|
| 1006 |
"step": 595
|
| 1007 |
},
|
| 1008 |
{
|
| 1009 |
+
"epoch": 1.748452857662905,
|
| 1010 |
+
"grad_norm": 0.1791963130235672,
|
| 1011 |
"learning_rate": 2e-05,
|
| 1012 |
+
"loss": 0.9656,
|
| 1013 |
+
"mean_token_accuracy": 0.7226544409990311,
|
| 1014 |
"step": 600
|
| 1015 |
},
|
| 1016 |
{
|
| 1017 |
+
"epoch": 1.748452857662905,
|
| 1018 |
+
"eval_loss": 0.9683948755264282,
|
| 1019 |
+
"eval_mean_token_accuracy": 0.719802176952362,
|
| 1020 |
+
"eval_runtime": 9.0575,
|
| 1021 |
+
"eval_samples_per_second": 7.287,
|
| 1022 |
+
"eval_steps_per_second": 0.552,
|
| 1023 |
"step": 600
|
| 1024 |
},
|
| 1025 |
{
|
| 1026 |
+
"epoch": 1.7630141973061522,
|
| 1027 |
+
"grad_norm": 0.2111322432756424,
|
| 1028 |
"learning_rate": 2e-05,
|
| 1029 |
+
"loss": 0.9594,
|
| 1030 |
+
"mean_token_accuracy": 0.7252815589308739,
|
| 1031 |
"step": 605
|
| 1032 |
},
|
| 1033 |
{
|
| 1034 |
+
"epoch": 1.7775755369493993,
|
| 1035 |
+
"grad_norm": 0.1824427843093872,
|
| 1036 |
"learning_rate": 2e-05,
|
| 1037 |
+
"loss": 0.9448,
|
| 1038 |
+
"mean_token_accuracy": 0.730655600130558,
|
| 1039 |
"step": 610
|
| 1040 |
},
|
| 1041 |
{
|
| 1042 |
+
"epoch": 1.7921368765926466,
|
| 1043 |
+
"grad_norm": 0.1896345168352127,
|
| 1044 |
"learning_rate": 2e-05,
|
| 1045 |
+
"loss": 0.9429,
|
| 1046 |
+
"mean_token_accuracy": 0.7295209676027298,
|
| 1047 |
"step": 615
|
| 1048 |
},
|
| 1049 |
{
|
| 1050 |
+
"epoch": 1.8066982162358936,
|
| 1051 |
+
"grad_norm": 0.19917869567871094,
|
| 1052 |
"learning_rate": 2e-05,
|
| 1053 |
+
"loss": 0.9481,
|
| 1054 |
+
"mean_token_accuracy": 0.72839834690094,
|
| 1055 |
"step": 620
|
| 1056 |
},
|
| 1057 |
{
|
| 1058 |
+
"epoch": 1.821259555879141,
|
| 1059 |
+
"grad_norm": 0.18886856734752655,
|
| 1060 |
"learning_rate": 2e-05,
|
| 1061 |
+
"loss": 0.9494,
|
| 1062 |
+
"mean_token_accuracy": 0.7279918506741524,
|
| 1063 |
"step": 625
|
| 1064 |
},
|
| 1065 |
{
|
| 1066 |
+
"epoch": 1.835820895522388,
|
| 1067 |
+
"grad_norm": 0.18665249645709991,
|
| 1068 |
"learning_rate": 2e-05,
|
| 1069 |
+
"loss": 0.941,
|
| 1070 |
+
"mean_token_accuracy": 0.7308425426483154,
|
| 1071 |
"step": 630
|
| 1072 |
},
|
| 1073 |
{
|
| 1074 |
+
"epoch": 1.8503822351656352,
|
| 1075 |
+
"grad_norm": 0.1790022999048233,
|
| 1076 |
"learning_rate": 2e-05,
|
| 1077 |
+
"loss": 0.9459,
|
| 1078 |
+
"mean_token_accuracy": 0.7287953227758408,
|
| 1079 |
"step": 635
|
| 1080 |
},
|
| 1081 |
{
|
| 1082 |
+
"epoch": 1.8649435748088825,
|
| 1083 |
+
"grad_norm": 0.18598653376102448,
|
| 1084 |
"learning_rate": 2e-05,
|
| 1085 |
+
"loss": 0.9344,
|
| 1086 |
+
"mean_token_accuracy": 0.7317764893174171,
|
| 1087 |
"step": 640
|
| 1088 |
},
|
| 1089 |
{
|
| 1090 |
+
"epoch": 1.8795049144521296,
|
| 1091 |
+
"grad_norm": 0.18939712643623352,
|
| 1092 |
"learning_rate": 2e-05,
|
| 1093 |
+
"loss": 0.9473,
|
| 1094 |
+
"mean_token_accuracy": 0.7287281811237335,
|
| 1095 |
"step": 645
|
| 1096 |
},
|
| 1097 |
{
|
| 1098 |
+
"epoch": 1.8940662540953768,
|
| 1099 |
+
"grad_norm": 0.1971713751554489,
|
| 1100 |
"learning_rate": 2e-05,
|
| 1101 |
+
"loss": 0.9363,
|
| 1102 |
+
"mean_token_accuracy": 0.731384290754795,
|
| 1103 |
"step": 650
|
| 1104 |
},
|
| 1105 |
{
|
| 1106 |
+
"epoch": 1.9086275937386241,
|
| 1107 |
+
"grad_norm": 0.19053973257541656,
|
| 1108 |
"learning_rate": 2e-05,
|
| 1109 |
+
"loss": 0.9419,
|
| 1110 |
+
"mean_token_accuracy": 0.7299284294247628,
|
| 1111 |
"step": 655
|
| 1112 |
},
|
| 1113 |
{
|
| 1114 |
+
"epoch": 1.923188933381871,
|
| 1115 |
+
"grad_norm": 0.1968354731798172,
|
| 1116 |
"learning_rate": 2e-05,
|
| 1117 |
+
"loss": 0.9427,
|
| 1118 |
+
"mean_token_accuracy": 0.7299772590398789,
|
| 1119 |
"step": 660
|
| 1120 |
},
|
| 1121 |
{
|
| 1122 |
+
"epoch": 1.9377502730251184,
|
| 1123 |
+
"grad_norm": 0.19762156903743744,
|
| 1124 |
"learning_rate": 2e-05,
|
| 1125 |
+
"loss": 0.9517,
|
| 1126 |
+
"mean_token_accuracy": 0.7256975680589676,
|
| 1127 |
"step": 665
|
| 1128 |
},
|
| 1129 |
{
|
| 1130 |
+
"epoch": 1.9523116126683655,
|
| 1131 |
+
"grad_norm": 0.18599362671375275,
|
| 1132 |
"learning_rate": 2e-05,
|
| 1133 |
+
"loss": 0.9346,
|
| 1134 |
+
"mean_token_accuracy": 0.7319954812526703,
|
| 1135 |
"step": 670
|
| 1136 |
},
|
| 1137 |
{
|
| 1138 |
+
"epoch": 1.9668729523116126,
|
| 1139 |
+
"grad_norm": 0.18457342684268951,
|
| 1140 |
"learning_rate": 2e-05,
|
| 1141 |
+
"loss": 0.9555,
|
| 1142 |
+
"mean_token_accuracy": 0.726130048930645,
|
| 1143 |
"step": 675
|
| 1144 |
},
|
| 1145 |
{
|
| 1146 |
+
"epoch": 1.98143429195486,
|
| 1147 |
+
"grad_norm": 0.1952061802148819,
|
| 1148 |
"learning_rate": 2e-05,
|
| 1149 |
+
"loss": 0.9396,
|
| 1150 |
+
"mean_token_accuracy": 0.7293583780527115,
|
| 1151 |
"step": 680
|
| 1152 |
},
|
| 1153 |
{
|
| 1154 |
+
"epoch": 1.995995631598107,
|
| 1155 |
+
"grad_norm": 0.24200007319450378,
|
| 1156 |
"learning_rate": 2e-05,
|
| 1157 |
+
"loss": 0.9461,
|
| 1158 |
+
"mean_token_accuracy": 0.7285397097468376,
|
| 1159 |
"step": 685
|
| 1160 |
},
|
| 1161 |
{
|
| 1162 |
+
"epoch": 2.0116490717145976,
|
| 1163 |
+
"grad_norm": 0.23615935444831848,
|
| 1164 |
"learning_rate": 2e-05,
|
| 1165 |
+
"loss": 0.9642,
|
| 1166 |
+
"mean_token_accuracy": 0.735066328729902,
|
| 1167 |
"step": 690
|
| 1168 |
},
|
| 1169 |
{
|
| 1170 |
+
"epoch": 2.026210411357845,
|
| 1171 |
+
"grad_norm": 0.20251308381557465,
|
| 1172 |
"learning_rate": 2e-05,
|
| 1173 |
+
"loss": 0.9104,
|
| 1174 |
+
"mean_token_accuracy": 0.7364416778087616,
|
| 1175 |
"step": 695
|
| 1176 |
},
|
| 1177 |
{
|
| 1178 |
+
"epoch": 2.0407717510010923,
|
| 1179 |
+
"grad_norm": 0.17524279654026031,
|
| 1180 |
"learning_rate": 2e-05,
|
| 1181 |
+
"loss": 0.9201,
|
| 1182 |
+
"mean_token_accuracy": 0.7344257399439812,
|
| 1183 |
"step": 700
|
| 1184 |
},
|
| 1185 |
{
|
| 1186 |
+
"epoch": 2.0407717510010923,
|
| 1187 |
+
"eval_loss": 0.9588068127632141,
|
| 1188 |
+
"eval_mean_token_accuracy": 0.7223332643508911,
|
| 1189 |
+
"eval_runtime": 9.0725,
|
| 1190 |
+
"eval_samples_per_second": 7.275,
|
| 1191 |
+
"eval_steps_per_second": 0.551,
|
| 1192 |
"step": 700
|
| 1193 |
},
|
| 1194 |
{
|
| 1195 |
+
"epoch": 2.0553330906443392,
|
| 1196 |
+
"grad_norm": 0.177406445145607,
|
| 1197 |
"learning_rate": 2e-05,
|
| 1198 |
+
"loss": 0.8955,
|
| 1199 |
+
"mean_token_accuracy": 0.741851630806923,
|
| 1200 |
"step": 705
|
| 1201 |
},
|
| 1202 |
{
|
| 1203 |
+
"epoch": 2.0698944302875866,
|
| 1204 |
+
"grad_norm": 0.17523570358753204,
|
| 1205 |
"learning_rate": 2e-05,
|
| 1206 |
+
"loss": 0.9214,
|
| 1207 |
+
"mean_token_accuracy": 0.7350994989275932,
|
| 1208 |
"step": 710
|
| 1209 |
},
|
| 1210 |
{
|
| 1211 |
+
"epoch": 2.0844557699308335,
|
| 1212 |
+
"grad_norm": 0.18342241644859314,
|
| 1213 |
"learning_rate": 2e-05,
|
| 1214 |
+
"loss": 0.9071,
|
| 1215 |
+
"mean_token_accuracy": 0.7384255573153495,
|
| 1216 |
"step": 715
|
| 1217 |
},
|
| 1218 |
{
|
| 1219 |
+
"epoch": 2.099017109574081,
|
| 1220 |
+
"grad_norm": 0.17445097863674164,
|
| 1221 |
"learning_rate": 2e-05,
|
| 1222 |
+
"loss": 0.901,
|
| 1223 |
+
"mean_token_accuracy": 0.7399073630571366,
|
| 1224 |
"step": 720
|
| 1225 |
},
|
| 1226 |
{
|
| 1227 |
+
"epoch": 2.113578449217328,
|
| 1228 |
+
"grad_norm": 0.18421486020088196,
|
| 1229 |
"learning_rate": 2e-05,
|
| 1230 |
+
"loss": 0.9381,
|
| 1231 |
+
"mean_token_accuracy": 0.7290249973535537,
|
| 1232 |
"step": 725
|
| 1233 |
},
|
| 1234 |
{
|
| 1235 |
+
"epoch": 2.128139788860575,
|
| 1236 |
+
"grad_norm": 0.18407249450683594,
|
| 1237 |
"learning_rate": 2e-05,
|
| 1238 |
+
"loss": 0.9094,
|
| 1239 |
+
"mean_token_accuracy": 0.7381876617670059,
|
| 1240 |
"step": 730
|
| 1241 |
},
|
| 1242 |
{
|
| 1243 |
+
"epoch": 2.1427011285038224,
|
| 1244 |
+
"grad_norm": 0.19551889598369598,
|
| 1245 |
"learning_rate": 2e-05,
|
| 1246 |
+
"loss": 0.902,
|
| 1247 |
+
"mean_token_accuracy": 0.7397311061620713,
|
| 1248 |
"step": 735
|
| 1249 |
},
|
| 1250 |
{
|
| 1251 |
+
"epoch": 2.1572624681470693,
|
| 1252 |
+
"grad_norm": 0.19646836817264557,
|
| 1253 |
"learning_rate": 2e-05,
|
| 1254 |
+
"loss": 0.9189,
|
| 1255 |
+
"mean_token_accuracy": 0.7339320570230484,
|
| 1256 |
"step": 740
|
| 1257 |
},
|
| 1258 |
{
|
| 1259 |
+
"epoch": 2.1718238077903167,
|
| 1260 |
+
"grad_norm": 0.19153904914855957,
|
| 1261 |
"learning_rate": 2e-05,
|
| 1262 |
+
"loss": 0.9209,
|
| 1263 |
+
"mean_token_accuracy": 0.7338812783360481,
|
| 1264 |
"step": 745
|
| 1265 |
},
|
| 1266 |
{
|
| 1267 |
+
"epoch": 2.186385147433564,
|
| 1268 |
+
"grad_norm": 0.18896165490150452,
|
| 1269 |
"learning_rate": 2e-05,
|
| 1270 |
+
"loss": 0.9115,
|
| 1271 |
+
"mean_token_accuracy": 0.7361234918236732,
|
| 1272 |
"step": 750
|
| 1273 |
},
|
| 1274 |
{
|
| 1275 |
+
"epoch": 2.200946487076811,
|
| 1276 |
+
"grad_norm": 0.18773645162582397,
|
| 1277 |
"learning_rate": 2e-05,
|
| 1278 |
+
"loss": 0.9167,
|
| 1279 |
+
"mean_token_accuracy": 0.7351216241717339,
|
| 1280 |
"step": 755
|
| 1281 |
},
|
| 1282 |
{
|
| 1283 |
+
"epoch": 2.2155078267200583,
|
| 1284 |
+
"grad_norm": 0.16787868738174438,
|
| 1285 |
"learning_rate": 2e-05,
|
| 1286 |
+
"loss": 0.9127,
|
| 1287 |
+
"mean_token_accuracy": 0.7366888895630836,
|
| 1288 |
"step": 760
|
| 1289 |
},
|
| 1290 |
{
|
| 1291 |
+
"epoch": 2.2300691663633057,
|
| 1292 |
+
"grad_norm": 0.1758783459663391,
|
| 1293 |
"learning_rate": 2e-05,
|
| 1294 |
+
"loss": 0.9042,
|
| 1295 |
+
"mean_token_accuracy": 0.738301183283329,
|
| 1296 |
"step": 765
|
| 1297 |
},
|
| 1298 |
{
|
| 1299 |
+
"epoch": 2.2446305060065526,
|
| 1300 |
+
"grad_norm": 0.17144909501075745,
|
| 1301 |
"learning_rate": 2e-05,
|
| 1302 |
+
"loss": 0.8967,
|
| 1303 |
+
"mean_token_accuracy": 0.7403302609920501,
|
| 1304 |
"step": 770
|
| 1305 |
},
|
| 1306 |
{
|
| 1307 |
+
"epoch": 2.2591918456498,
|
| 1308 |
+
"grad_norm": 0.16952501237392426,
|
| 1309 |
"learning_rate": 2e-05,
|
| 1310 |
+
"loss": 0.9238,
|
| 1311 |
+
"mean_token_accuracy": 0.7333216354250908,
|
| 1312 |
"step": 775
|
| 1313 |
},
|
| 1314 |
{
|
| 1315 |
+
"epoch": 2.273753185293047,
|
| 1316 |
+
"grad_norm": 0.19771642982959747,
|
| 1317 |
"learning_rate": 2e-05,
|
| 1318 |
+
"loss": 0.9211,
|
| 1319 |
+
"mean_token_accuracy": 0.7324410900473595,
|
| 1320 |
"step": 780
|
| 1321 |
},
|
| 1322 |
{
|
| 1323 |
+
"epoch": 2.288314524936294,
|
| 1324 |
+
"grad_norm": 0.20366059243679047,
|
| 1325 |
"learning_rate": 2e-05,
|
| 1326 |
+
"loss": 0.9051,
|
| 1327 |
+
"mean_token_accuracy": 0.7371811017394065,
|
| 1328 |
"step": 785
|
| 1329 |
},
|
| 1330 |
{
|
| 1331 |
+
"epoch": 2.3028758645795415,
|
| 1332 |
+
"grad_norm": 0.18298108875751495,
|
| 1333 |
"learning_rate": 2e-05,
|
| 1334 |
+
"loss": 0.915,
|
| 1335 |
+
"mean_token_accuracy": 0.7352513417601585,
|
| 1336 |
"step": 790
|
| 1337 |
},
|
| 1338 |
{
|
| 1339 |
+
"epoch": 2.3174372042227884,
|
| 1340 |
+
"grad_norm": 0.17126043140888214,
|
| 1341 |
"learning_rate": 2e-05,
|
| 1342 |
+
"loss": 0.9078,
|
| 1343 |
+
"mean_token_accuracy": 0.7370902448892593,
|
| 1344 |
"step": 795
|
| 1345 |
},
|
| 1346 |
{
|
| 1347 |
+
"epoch": 2.3319985438660358,
|
| 1348 |
+
"grad_norm": 0.17268440127372742,
|
| 1349 |
"learning_rate": 2e-05,
|
| 1350 |
+
"loss": 0.9083,
|
| 1351 |
+
"mean_token_accuracy": 0.7362028434872627,
|
| 1352 |
"step": 800
|
| 1353 |
},
|
| 1354 |
{
|
| 1355 |
+
"epoch": 2.3319985438660358,
|
| 1356 |
+
"eval_loss": 0.9501657485961914,
|
| 1357 |
+
"eval_mean_token_accuracy": 0.7240748167037964,
|
| 1358 |
+
"eval_runtime": 9.038,
|
| 1359 |
+
"eval_samples_per_second": 7.303,
|
| 1360 |
+
"eval_steps_per_second": 0.553,
|
| 1361 |
"step": 800
|
| 1362 |
},
|
| 1363 |
{
|
| 1364 |
+
"epoch": 2.3465598835092827,
|
| 1365 |
+
"grad_norm": 0.18702249228954315,
|
| 1366 |
"learning_rate": 2e-05,
|
| 1367 |
+
"loss": 0.912,
|
| 1368 |
+
"mean_token_accuracy": 0.7360319286584854,
|
| 1369 |
"step": 805
|
| 1370 |
},
|
| 1371 |
{
|
| 1372 |
+
"epoch": 2.36112122315253,
|
| 1373 |
+
"grad_norm": 0.18535007536411285,
|
| 1374 |
"learning_rate": 2e-05,
|
| 1375 |
+
"loss": 0.9181,
|
| 1376 |
+
"mean_token_accuracy": 0.7348553270101548,
|
| 1377 |
"step": 810
|
| 1378 |
},
|
| 1379 |
{
|
| 1380 |
+
"epoch": 2.3756825627957774,
|
| 1381 |
+
"grad_norm": 0.19746533036231995,
|
| 1382 |
"learning_rate": 2e-05,
|
| 1383 |
+
"loss": 0.9119,
|
| 1384 |
+
"mean_token_accuracy": 0.7346351534128189,
|
| 1385 |
"step": 815
|
| 1386 |
},
|
| 1387 |
{
|
| 1388 |
+
"epoch": 2.3902439024390243,
|
| 1389 |
+
"grad_norm": 0.1812228113412857,
|
| 1390 |
"learning_rate": 2e-05,
|
| 1391 |
+
"loss": 0.9166,
|
| 1392 |
+
"mean_token_accuracy": 0.7355435863137245,
|
| 1393 |
"step": 820
|
| 1394 |
},
|
| 1395 |
{
|
| 1396 |
+
"epoch": 2.4048052420822716,
|
| 1397 |
+
"grad_norm": 0.1950427144765854,
|
| 1398 |
"learning_rate": 2e-05,
|
| 1399 |
+
"loss": 0.9186,
|
| 1400 |
+
"mean_token_accuracy": 0.7341731756925582,
|
| 1401 |
"step": 825
|
| 1402 |
},
|
| 1403 |
{
|
| 1404 |
+
"epoch": 2.4193665817255186,
|
| 1405 |
+
"grad_norm": 0.19460786879062653,
|
| 1406 |
"learning_rate": 2e-05,
|
| 1407 |
+
"loss": 0.9136,
|
| 1408 |
+
"mean_token_accuracy": 0.7349476546049118,
|
| 1409 |
"step": 830
|
| 1410 |
},
|
| 1411 |
{
|
| 1412 |
+
"epoch": 2.433927921368766,
|
| 1413 |
+
"grad_norm": 0.1719018816947937,
|
| 1414 |
"learning_rate": 2e-05,
|
| 1415 |
+
"loss": 0.9104,
|
| 1416 |
+
"mean_token_accuracy": 0.7361870780587196,
|
| 1417 |
"step": 835
|
| 1418 |
},
|
| 1419 |
{
|
| 1420 |
+
"epoch": 2.4484892610120133,
|
| 1421 |
+
"grad_norm": 0.16659170389175415,
|
| 1422 |
"learning_rate": 2e-05,
|
| 1423 |
+
"loss": 0.8925,
|
| 1424 |
+
"mean_token_accuracy": 0.740250737965107,
|
| 1425 |
"step": 840
|
| 1426 |
},
|
| 1427 |
{
|
| 1428 |
+
"epoch": 2.46305060065526,
|
| 1429 |
+
"grad_norm": 0.20289891958236694,
|
| 1430 |
"learning_rate": 2e-05,
|
| 1431 |
+
"loss": 0.8955,
|
| 1432 |
+
"mean_token_accuracy": 0.7406818434596062,
|
| 1433 |
"step": 845
|
| 1434 |
},
|
| 1435 |
{
|
| 1436 |
+
"epoch": 2.4776119402985075,
|
| 1437 |
+
"grad_norm": 0.19275344908237457,
|
| 1438 |
"learning_rate": 2e-05,
|
| 1439 |
+
"loss": 0.9111,
|
| 1440 |
+
"mean_token_accuracy": 0.7354253143072128,
|
| 1441 |
"step": 850
|
| 1442 |
},
|
| 1443 |
{
|
| 1444 |
+
"epoch": 2.4921732799417544,
|
| 1445 |
+
"grad_norm": 0.23027832806110382,
|
| 1446 |
"learning_rate": 2e-05,
|
| 1447 |
+
"loss": 0.903,
|
| 1448 |
+
"mean_token_accuracy": 0.7381577342748642,
|
| 1449 |
"step": 855
|
| 1450 |
},
|
| 1451 |
{
|
| 1452 |
+
"epoch": 2.5067346195850018,
|
| 1453 |
+
"grad_norm": 0.1945352405309677,
|
| 1454 |
"learning_rate": 2e-05,
|
| 1455 |
+
"loss": 0.9107,
|
| 1456 |
+
"mean_token_accuracy": 0.7365095824003219,
|
| 1457 |
"step": 860
|
| 1458 |
},
|
| 1459 |
{
|
| 1460 |
+
"epoch": 2.521295959228249,
|
| 1461 |
+
"grad_norm": 0.18727873265743256,
|
| 1462 |
"learning_rate": 2e-05,
|
| 1463 |
+
"loss": 0.9044,
|
| 1464 |
+
"mean_token_accuracy": 0.7381272122263909,
|
| 1465 |
"step": 865
|
| 1466 |
},
|
| 1467 |
{
|
| 1468 |
+
"epoch": 2.535857298871496,
|
| 1469 |
+
"grad_norm": 0.1687106490135193,
|
| 1470 |
"learning_rate": 2e-05,
|
| 1471 |
+
"loss": 0.9016,
|
| 1472 |
+
"mean_token_accuracy": 0.7383408591151237,
|
| 1473 |
"step": 870
|
| 1474 |
},
|
| 1475 |
{
|
| 1476 |
+
"epoch": 2.5504186385147434,
|
| 1477 |
+
"grad_norm": 0.18417419493198395,
|
| 1478 |
"learning_rate": 2e-05,
|
| 1479 |
+
"loss": 0.9023,
|
| 1480 |
+
"mean_token_accuracy": 0.7390033379197121,
|
| 1481 |
"step": 875
|
| 1482 |
},
|
| 1483 |
{
|
| 1484 |
+
"epoch": 2.5649799781579903,
|
| 1485 |
+
"grad_norm": 0.18553201854228973,
|
| 1486 |
"learning_rate": 2e-05,
|
| 1487 |
+
"loss": 0.8995,
|
| 1488 |
+
"mean_token_accuracy": 0.7391414895653725,
|
| 1489 |
"step": 880
|
| 1490 |
},
|
| 1491 |
{
|
| 1492 |
+
"epoch": 2.5795413178012376,
|
| 1493 |
+
"grad_norm": 0.17184361815452576,
|
| 1494 |
"learning_rate": 2e-05,
|
| 1495 |
+
"loss": 0.8977,
|
| 1496 |
+
"mean_token_accuracy": 0.7395716354250907,
|
| 1497 |
"step": 885
|
| 1498 |
},
|
| 1499 |
{
|
| 1500 |
+
"epoch": 2.594102657444485,
|
| 1501 |
+
"grad_norm": 0.19258299469947815,
|
| 1502 |
"learning_rate": 2e-05,
|
| 1503 |
+
"loss": 0.9003,
|
| 1504 |
+
"mean_token_accuracy": 0.7391077131032944,
|
| 1505 |
"step": 890
|
| 1506 |
},
|
| 1507 |
{
|
| 1508 |
+
"epoch": 2.6086639970877323,
|
| 1509 |
+
"grad_norm": 0.18599550426006317,
|
| 1510 |
"learning_rate": 2e-05,
|
| 1511 |
+
"loss": 0.9007,
|
| 1512 |
+
"mean_token_accuracy": 0.7381867274641991,
|
| 1513 |
"step": 895
|
| 1514 |
},
|
| 1515 |
{
|
| 1516 |
+
"epoch": 2.6232253367309792,
|
| 1517 |
+
"grad_norm": 0.1717829406261444,
|
| 1518 |
"learning_rate": 2e-05,
|
| 1519 |
+
"loss": 0.9141,
|
| 1520 |
+
"mean_token_accuracy": 0.7353848740458488,
|
| 1521 |
"step": 900
|
| 1522 |
},
|
| 1523 |
{
|
| 1524 |
+
"epoch": 2.6232253367309792,
|
| 1525 |
+
"eval_loss": 0.9417613744735718,
|
| 1526 |
+
"eval_mean_token_accuracy": 0.7259013175964355,
|
| 1527 |
+
"eval_runtime": 9.0308,
|
| 1528 |
+
"eval_samples_per_second": 7.308,
|
| 1529 |
+
"eval_steps_per_second": 0.554,
|
| 1530 |
"step": 900
|
| 1531 |
},
|
| 1532 |
{
|
| 1533 |
+
"epoch": 2.6377866763742266,
|
| 1534 |
+
"grad_norm": 0.1715007722377777,
|
| 1535 |
"learning_rate": 2e-05,
|
| 1536 |
+
"loss": 0.8983,
|
| 1537 |
+
"mean_token_accuracy": 0.7399233922362327,
|
| 1538 |
"step": 905
|
| 1539 |
},
|
| 1540 |
{
|
| 1541 |
+
"epoch": 2.6523480160174735,
|
| 1542 |
+
"grad_norm": 0.20033277571201324,
|
| 1543 |
"learning_rate": 2e-05,
|
| 1544 |
+
"loss": 0.8892,
|
| 1545 |
+
"mean_token_accuracy": 0.7421796754002571,
|
| 1546 |
"step": 910
|
| 1547 |
},
|
| 1548 |
{
|
| 1549 |
+
"epoch": 2.666909355660721,
|
| 1550 |
+
"grad_norm": 0.19205278158187866,
|
| 1551 |
"learning_rate": 2e-05,
|
| 1552 |
+
"loss": 0.9151,
|
| 1553 |
+
"mean_token_accuracy": 0.7349491819739342,
|
| 1554 |
"step": 915
|
| 1555 |
},
|
| 1556 |
{
|
| 1557 |
+
"epoch": 2.681470695303968,
|
| 1558 |
+
"grad_norm": 0.1759193241596222,
|
| 1559 |
"learning_rate": 2e-05,
|
| 1560 |
+
"loss": 0.8897,
|
| 1561 |
+
"mean_token_accuracy": 0.7411266922950744,
|
| 1562 |
"step": 920
|
| 1563 |
},
|
| 1564 |
{
|
| 1565 |
+
"epoch": 2.696032034947215,
|
| 1566 |
+
"grad_norm": 0.18072772026062012,
|
| 1567 |
"learning_rate": 2e-05,
|
| 1568 |
+
"loss": 0.9085,
|
| 1569 |
+
"mean_token_accuracy": 0.737930352985859,
|
| 1570 |
"step": 925
|
| 1571 |
},
|
| 1572 |
{
|
| 1573 |
+
"epoch": 2.7105933745904625,
|
| 1574 |
+
"grad_norm": 0.18214493989944458,
|
| 1575 |
"learning_rate": 2e-05,
|
| 1576 |
+
"loss": 0.9032,
|
| 1577 |
+
"mean_token_accuracy": 0.737660813331604,
|
| 1578 |
"step": 930
|
| 1579 |
},
|
| 1580 |
{
|
| 1581 |
+
"epoch": 2.7251547142337094,
|
| 1582 |
+
"grad_norm": 0.1743292659521103,
|
| 1583 |
"learning_rate": 2e-05,
|
| 1584 |
+
"loss": 0.8834,
|
| 1585 |
+
"mean_token_accuracy": 0.7436462283134461,
|
| 1586 |
"step": 935
|
| 1587 |
},
|
| 1588 |
{
|
| 1589 |
+
"epoch": 2.7397160538769567,
|
| 1590 |
+
"grad_norm": 0.21689608693122864,
|
| 1591 |
"learning_rate": 2e-05,
|
| 1592 |
+
"loss": 0.8943,
|
| 1593 |
+
"mean_token_accuracy": 0.7397776529192924,
|
| 1594 |
"step": 940
|
| 1595 |
},
|
| 1596 |
{
|
| 1597 |
+
"epoch": 2.754277393520204,
|
| 1598 |
+
"grad_norm": 0.19008708000183105,
|
| 1599 |
"learning_rate": 2e-05,
|
| 1600 |
+
"loss": 0.9243,
|
| 1601 |
+
"mean_token_accuracy": 0.7323350265622139,
|
| 1602 |
"step": 945
|
| 1603 |
},
|
| 1604 |
{
|
| 1605 |
+
"epoch": 2.768838733163451,
|
| 1606 |
+
"grad_norm": 0.20445488393306732,
|
| 1607 |
"learning_rate": 2e-05,
|
| 1608 |
+
"loss": 0.8923,
|
| 1609 |
+
"mean_token_accuracy": 0.7400558426976204,
|
| 1610 |
"step": 950
|
| 1611 |
},
|
| 1612 |
{
|
| 1613 |
+
"epoch": 2.7834000728066983,
|
| 1614 |
+
"grad_norm": 0.18380148708820343,
|
| 1615 |
"learning_rate": 2e-05,
|
| 1616 |
+
"loss": 0.8938,
|
| 1617 |
+
"mean_token_accuracy": 0.7402835443615914,
|
| 1618 |
"step": 955
|
| 1619 |
},
|
| 1620 |
{
|
| 1621 |
+
"epoch": 2.7979614124499452,
|
| 1622 |
+
"grad_norm": 0.17851661145687103,
|
| 1623 |
"learning_rate": 2e-05,
|
| 1624 |
+
"loss": 0.8992,
|
| 1625 |
+
"mean_token_accuracy": 0.7384199738502503,
|
| 1626 |
"step": 960
|
| 1627 |
},
|
| 1628 |
{
|
| 1629 |
+
"epoch": 2.8125227520931926,
|
| 1630 |
+
"grad_norm": 0.20929712057113647,
|
| 1631 |
"learning_rate": 2e-05,
|
| 1632 |
+
"loss": 0.895,
|
| 1633 |
+
"mean_token_accuracy": 0.740375104546547,
|
| 1634 |
"step": 965
|
| 1635 |
},
|
| 1636 |
{
|
| 1637 |
+
"epoch": 2.82708409173644,
|
| 1638 |
+
"grad_norm": 0.19595912098884583,
|
| 1639 |
"learning_rate": 2e-05,
|
| 1640 |
+
"loss": 0.9008,
|
| 1641 |
+
"mean_token_accuracy": 0.7384095326066017,
|
| 1642 |
"step": 970
|
| 1643 |
},
|
| 1644 |
{
|
| 1645 |
+
"epoch": 2.841645431379687,
|
| 1646 |
+
"grad_norm": 0.17353329062461853,
|
| 1647 |
"learning_rate": 2e-05,
|
| 1648 |
+
"loss": 0.9021,
|
| 1649 |
+
"mean_token_accuracy": 0.7383004203438759,
|
| 1650 |
"step": 975
|
| 1651 |
},
|
| 1652 |
{
|
| 1653 |
+
"epoch": 2.856206771022934,
|
| 1654 |
+
"grad_norm": 0.1858338862657547,
|
| 1655 |
"learning_rate": 2e-05,
|
| 1656 |
+
"loss": 0.9004,
|
| 1657 |
+
"mean_token_accuracy": 0.7389308467507363,
|
| 1658 |
"step": 980
|
| 1659 |
},
|
| 1660 |
{
|
| 1661 |
+
"epoch": 2.870768110666181,
|
| 1662 |
+
"grad_norm": 0.18422286212444305,
|
| 1663 |
"learning_rate": 2e-05,
|
| 1664 |
+
"loss": 0.9109,
|
| 1665 |
+
"mean_token_accuracy": 0.7359930142760277,
|
| 1666 |
"step": 985
|
| 1667 |
},
|
| 1668 |
{
|
| 1669 |
+
"epoch": 2.8853294503094284,
|
| 1670 |
+
"grad_norm": 0.17848879098892212,
|
| 1671 |
"learning_rate": 2e-05,
|
| 1672 |
+
"loss": 0.8953,
|
| 1673 |
+
"mean_token_accuracy": 0.7397311091423034,
|
| 1674 |
"step": 990
|
| 1675 |
},
|
| 1676 |
{
|
| 1677 |
+
"epoch": 2.899890789952676,
|
| 1678 |
+
"grad_norm": 0.1943521797657013,
|
| 1679 |
"learning_rate": 2e-05,
|
| 1680 |
+
"loss": 0.8961,
|
| 1681 |
+
"mean_token_accuracy": 0.7394258737564087,
|
| 1682 |
"step": 995
|
| 1683 |
},
|
| 1684 |
{
|
| 1685 |
+
"epoch": 2.9144521295959227,
|
| 1686 |
+
"grad_norm": 0.19629698991775513,
|
| 1687 |
"learning_rate": 2e-05,
|
| 1688 |
+
"loss": 0.9044,
|
| 1689 |
+
"mean_token_accuracy": 0.7370201960206032,
|
| 1690 |
"step": 1000
|
| 1691 |
},
|
| 1692 |
{
|
| 1693 |
+
"epoch": 2.9144521295959227,
|
| 1694 |
+
"eval_loss": 0.9340672492980957,
|
| 1695 |
+
"eval_mean_token_accuracy": 0.7274513006210327,
|
| 1696 |
+
"eval_runtime": 9.0532,
|
| 1697 |
+
"eval_samples_per_second": 7.29,
|
| 1698 |
+
"eval_steps_per_second": 0.552,
|
| 1699 |
"step": 1000
|
| 1700 |
},
|
| 1701 |
{
|
| 1702 |
+
"epoch": 2.92901346923917,
|
| 1703 |
+
"grad_norm": 0.19953298568725586,
|
| 1704 |
"learning_rate": 2e-05,
|
| 1705 |
+
"loss": 0.9018,
|
| 1706 |
+
"mean_token_accuracy": 0.7379929170012474,
|
| 1707 |
"step": 1005
|
| 1708 |
},
|
| 1709 |
{
|
| 1710 |
+
"epoch": 2.943574808882417,
|
| 1711 |
+
"grad_norm": 0.19467906653881073,
|
| 1712 |
"learning_rate": 2e-05,
|
| 1713 |
+
"loss": 0.9054,
|
| 1714 |
+
"mean_token_accuracy": 0.7371795266866684,
|
| 1715 |
"step": 1010
|
| 1716 |
},
|
| 1717 |
{
|
| 1718 |
+
"epoch": 2.9581361485256643,
|
| 1719 |
+
"grad_norm": 0.18607822060585022,
|
| 1720 |
"learning_rate": 2e-05,
|
| 1721 |
+
"loss": 0.9015,
|
| 1722 |
+
"mean_token_accuracy": 0.7378426045179367,
|
| 1723 |
"step": 1015
|
| 1724 |
},
|
| 1725 |
{
|
| 1726 |
+
"epoch": 2.9726974881689117,
|
| 1727 |
+
"grad_norm": 0.18633662164211273,
|
| 1728 |
"learning_rate": 2e-05,
|
| 1729 |
+
"loss": 0.9047,
|
| 1730 |
+
"mean_token_accuracy": 0.737396989762783,
|
| 1731 |
"step": 1020
|
| 1732 |
},
|
| 1733 |
{
|
| 1734 |
+
"epoch": 2.9872588278121586,
|
| 1735 |
+
"grad_norm": 0.17410708963871002,
|
| 1736 |
"learning_rate": 2e-05,
|
| 1737 |
+
"loss": 0.8927,
|
| 1738 |
+
"mean_token_accuracy": 0.7408069744706154,
|
| 1739 |
"step": 1025
|
| 1740 |
},
|
| 1741 |
{
|
| 1742 |
+
"epoch": 2.9989078995267566,
|
| 1743 |
+
"mean_token_accuracy": 0.7377504613250494,
|
| 1744 |
+
"step": 1029,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1745 |
"total_flos": 0.0,
|
| 1746 |
+
"train_loss": 0.9946218774200528,
|
| 1747 |
+
"train_runtime": 21041.6427,
|
| 1748 |
+
"train_samples_per_second": 1.566,
|
| 1749 |
+
"train_steps_per_second": 0.049
|
| 1750 |
}
|
| 1751 |
],
|
| 1752 |
"logging_steps": 5,
|
| 1753 |
+
"max_steps": 1029,
|
| 1754 |
"num_input_tokens_seen": 0,
|
| 1755 |
+
"num_train_epochs": 3,
|
| 1756 |
"save_steps": 500,
|
| 1757 |
"stateful_callbacks": {
|
| 1758 |
"TrainerControl": {
|
|
|
|
| 1767 |
}
|
| 1768 |
},
|
| 1769 |
"total_flos": 0.0,
|
| 1770 |
+
"train_batch_size": 1,
|
| 1771 |
"trial_name": null,
|
| 1772 |
"trial_params": null
|
| 1773 |
}
|