Model save
Browse files- README.md +8 -14
- all_results.json +6 -11
- model.safetensors +1 -1
- runs/Sep22_21-06-18_nova.cs.ucla.edu/events.out.tfevents.1727064531.nova.cs.ucla.edu.485295.0 +2 -2
- train_results.json +6 -6
- trainer_state.json +744 -1010
README.md
CHANGED
|
@@ -2,15 +2,9 @@
|
|
| 2 |
license: apache-2.0
|
| 3 |
base_model: TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T
|
| 4 |
tags:
|
| 5 |
-
- alignment-handbook
|
| 6 |
- trl
|
| 7 |
- sft
|
| 8 |
- generated_from_trainer
|
| 9 |
-
- trl
|
| 10 |
-
- sft
|
| 11 |
-
- generated_from_trainer
|
| 12 |
-
datasets:
|
| 13 |
-
- yihanwang617/WizardLM_70k_processed_indicator
|
| 14 |
model-index:
|
| 15 |
- name: tinyllama-sft-wizard-processed-indicator-0.6-full
|
| 16 |
results: []
|
|
@@ -21,9 +15,9 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 21 |
|
| 22 |
# tinyllama-sft-wizard-processed-indicator-0.6-full
|
| 23 |
|
| 24 |
-
This model is a fine-tuned version of [TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T](https://huggingface.co/TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T) on
|
| 25 |
It achieves the following results on the evaluation set:
|
| 26 |
-
- Loss: 0.
|
| 27 |
|
| 28 |
## Model description
|
| 29 |
|
|
@@ -58,15 +52,15 @@ The following hyperparameters were used during training:
|
|
| 58 |
|
| 59 |
### Training results
|
| 60 |
|
| 61 |
-
| Training Loss | Epoch
|
| 62 |
-
|
| 63 |
-
| 0.
|
| 64 |
-
| 0.
|
| 65 |
|
| 66 |
|
| 67 |
### Framework versions
|
| 68 |
|
| 69 |
- Transformers 4.40.1
|
| 70 |
-
- Pytorch 2.
|
| 71 |
-
- Datasets 2.
|
| 72 |
- Tokenizers 0.19.1
|
|
|
|
| 2 |
license: apache-2.0
|
| 3 |
base_model: TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T
|
| 4 |
tags:
|
|
|
|
| 5 |
- trl
|
| 6 |
- sft
|
| 7 |
- generated_from_trainer
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 8 |
model-index:
|
| 9 |
- name: tinyllama-sft-wizard-processed-indicator-0.6-full
|
| 10 |
results: []
|
|
|
|
| 15 |
|
| 16 |
# tinyllama-sft-wizard-processed-indicator-0.6-full
|
| 17 |
|
| 18 |
+
This model is a fine-tuned version of [TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T](https://huggingface.co/TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T) on an unknown dataset.
|
| 19 |
It achieves the following results on the evaluation set:
|
| 20 |
+
- Loss: 0.7533
|
| 21 |
|
| 22 |
## Model description
|
| 23 |
|
|
|
|
| 52 |
|
| 53 |
### Training results
|
| 54 |
|
| 55 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
| 56 |
+
|:-------------:|:------:|:----:|:---------------:|
|
| 57 |
+
| 0.7385 | 0.9989 | 449 | 0.7580 |
|
| 58 |
+
| 0.616 | 1.9978 | 898 | 0.7533 |
|
| 59 |
|
| 60 |
|
| 61 |
### Framework versions
|
| 62 |
|
| 63 |
- Transformers 4.40.1
|
| 64 |
+
- Pytorch 2.4.0+cu121
|
| 65 |
+
- Datasets 2.20.0
|
| 66 |
- Tokenizers 0.19.1
|
all_results.json
CHANGED
|
@@ -1,14 +1,9 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"
|
| 4 |
-
"
|
| 5 |
-
"
|
| 6 |
-
"
|
| 7 |
-
"
|
| 8 |
-
"total_flos": 98609754931200.0,
|
| 9 |
-
"train_loss": 0.7310274319095507,
|
| 10 |
-
"train_runtime": 7114.9679,
|
| 11 |
-
"train_samples": 69499,
|
| 12 |
-
"train_samples_per_second": 19.536,
|
| 13 |
"train_steps_per_second": 0.153
|
| 14 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 1.9977753058954395,
|
| 3 |
+
"total_flos": 81428314521600.0,
|
| 4 |
+
"train_loss": 0.7065894536026868,
|
| 5 |
+
"train_runtime": 5871.8562,
|
| 6 |
+
"train_samples": 57523,
|
| 7 |
+
"train_samples_per_second": 19.593,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 8 |
"train_steps_per_second": 0.153
|
| 9 |
}
|
model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 2200136248
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e524bb680938336a0d0caa6671337b7777d6be5159e33786782ea430cc5a64d8
|
| 3 |
size 2200136248
|
runs/Sep22_21-06-18_nova.cs.ucla.edu/events.out.tfevents.1727064531.nova.cs.ucla.edu.485295.0
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e0c85bf361daa1a64e1a2941d7f98e2cbb02b3fa8eb1e1a5552d2bb28d00e668
|
| 3 |
+
size 43754
|
train_results.json
CHANGED
|
@@ -1,9 +1,9 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"total_flos":
|
| 4 |
-
"train_loss": 0.
|
| 5 |
-
"train_runtime":
|
| 6 |
-
"train_samples":
|
| 7 |
-
"train_samples_per_second": 19.
|
| 8 |
"train_steps_per_second": 0.153
|
| 9 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 1.9977753058954395,
|
| 3 |
+
"total_flos": 81428314521600.0,
|
| 4 |
+
"train_loss": 0.7065894536026868,
|
| 5 |
+
"train_runtime": 5871.8562,
|
| 6 |
+
"train_samples": 57523,
|
| 7 |
+
"train_samples_per_second": 19.593,
|
| 8 |
"train_steps_per_second": 0.153
|
| 9 |
}
|
trainer_state.json
CHANGED
|
@@ -1,1571 +1,1305 @@
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
-
"epoch":
|
| 5 |
"eval_steps": 500,
|
| 6 |
-
"global_step":
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
-
"epoch": 0.
|
| 13 |
-
"grad_norm":
|
| 14 |
-
"learning_rate":
|
| 15 |
-
"loss": 1.
|
| 16 |
"step": 1
|
| 17 |
},
|
| 18 |
{
|
| 19 |
-
"epoch": 0.
|
| 20 |
-
"grad_norm":
|
| 21 |
-
"learning_rate":
|
| 22 |
-
"loss": 1.
|
| 23 |
"step": 5
|
| 24 |
},
|
| 25 |
{
|
| 26 |
-
"epoch": 0.
|
| 27 |
-
"grad_norm": 7.
|
| 28 |
-
"learning_rate":
|
| 29 |
-
"loss": 1.
|
| 30 |
"step": 10
|
| 31 |
},
|
| 32 |
{
|
| 33 |
-
"epoch": 0.
|
| 34 |
-
"grad_norm": 5.
|
| 35 |
-
"learning_rate":
|
| 36 |
-
"loss": 1.
|
| 37 |
"step": 15
|
| 38 |
},
|
| 39 |
{
|
| 40 |
-
"epoch": 0.
|
| 41 |
-
"grad_norm":
|
| 42 |
-
"learning_rate":
|
| 43 |
-
"loss":
|
| 44 |
"step": 20
|
| 45 |
},
|
| 46 |
{
|
| 47 |
-
"epoch": 0.
|
| 48 |
-
"grad_norm": 1.
|
| 49 |
-
"learning_rate":
|
| 50 |
-
"loss": 0.
|
| 51 |
"step": 25
|
| 52 |
},
|
| 53 |
{
|
| 54 |
-
"epoch": 0.
|
| 55 |
-
"grad_norm": 1.
|
| 56 |
-
"learning_rate":
|
| 57 |
-
"loss": 0.
|
| 58 |
"step": 30
|
| 59 |
},
|
| 60 |
{
|
| 61 |
-
"epoch": 0.
|
| 62 |
-
"grad_norm":
|
| 63 |
-
"learning_rate":
|
| 64 |
-
"loss": 0.
|
| 65 |
"step": 35
|
| 66 |
},
|
| 67 |
{
|
| 68 |
-
"epoch": 0.
|
| 69 |
-
"grad_norm": 0.
|
| 70 |
-
"learning_rate":
|
| 71 |
-
"loss": 0.
|
| 72 |
"step": 40
|
| 73 |
},
|
| 74 |
{
|
| 75 |
-
"epoch": 0.
|
| 76 |
-
"grad_norm": 0.
|
| 77 |
-
"learning_rate":
|
| 78 |
-
"loss": 0.
|
| 79 |
"step": 45
|
| 80 |
},
|
| 81 |
{
|
| 82 |
-
"epoch": 0.
|
| 83 |
-
"grad_norm": 0.
|
| 84 |
-
"learning_rate":
|
| 85 |
-
"loss": 0.
|
| 86 |
"step": 50
|
| 87 |
},
|
| 88 |
{
|
| 89 |
-
"epoch": 0.
|
| 90 |
-
"grad_norm": 0.
|
| 91 |
-
"learning_rate": 1.
|
| 92 |
-
"loss": 0.
|
| 93 |
"step": 55
|
| 94 |
},
|
| 95 |
{
|
| 96 |
-
"epoch": 0.
|
| 97 |
-
"grad_norm": 0.
|
| 98 |
-
"learning_rate": 1.
|
| 99 |
-
"loss": 0.
|
| 100 |
"step": 60
|
| 101 |
},
|
| 102 |
{
|
| 103 |
-
"epoch": 0.
|
| 104 |
-
"grad_norm": 0.
|
| 105 |
-
"learning_rate": 1.
|
| 106 |
-
"loss": 0.
|
| 107 |
"step": 65
|
| 108 |
},
|
| 109 |
{
|
| 110 |
-
"epoch": 0.
|
| 111 |
-
"grad_norm":
|
| 112 |
-
"learning_rate": 1.
|
| 113 |
-
"loss": 0.
|
| 114 |
"step": 70
|
| 115 |
},
|
| 116 |
{
|
| 117 |
-
"epoch": 0.
|
| 118 |
-
"grad_norm": 0.
|
| 119 |
-
"learning_rate": 1.
|
| 120 |
-
"loss": 0.
|
| 121 |
"step": 75
|
| 122 |
},
|
| 123 |
{
|
| 124 |
-
"epoch": 0.
|
| 125 |
-
"grad_norm": 0.
|
| 126 |
-
"learning_rate": 1.
|
| 127 |
-
"loss": 0.
|
| 128 |
"step": 80
|
| 129 |
},
|
| 130 |
{
|
| 131 |
-
"epoch": 0.
|
| 132 |
-
"grad_norm": 0.
|
| 133 |
-
"learning_rate": 1.
|
| 134 |
-
"loss": 0.
|
| 135 |
"step": 85
|
| 136 |
},
|
| 137 |
{
|
| 138 |
-
"epoch": 0.
|
| 139 |
-
"grad_norm": 0.
|
| 140 |
-
"learning_rate":
|
| 141 |
-
"loss": 0.
|
| 142 |
"step": 90
|
| 143 |
},
|
| 144 |
{
|
| 145 |
-
"epoch": 0.
|
| 146 |
-
"grad_norm": 0.
|
| 147 |
-
"learning_rate": 1.
|
| 148 |
-
"loss": 0.
|
| 149 |
"step": 95
|
| 150 |
},
|
| 151 |
{
|
| 152 |
-
"epoch": 0.
|
| 153 |
-
"grad_norm": 0.
|
| 154 |
-
"learning_rate": 1.
|
| 155 |
-
"loss": 0.
|
| 156 |
"step": 100
|
| 157 |
},
|
| 158 |
{
|
| 159 |
-
"epoch": 0.
|
| 160 |
-
"grad_norm": 0.
|
| 161 |
-
"learning_rate": 1.
|
| 162 |
-
"loss": 0.
|
| 163 |
"step": 105
|
| 164 |
},
|
| 165 |
{
|
| 166 |
-
"epoch": 0.
|
| 167 |
-
"grad_norm": 0.
|
| 168 |
-
"learning_rate": 1.
|
| 169 |
-
"loss": 0.
|
| 170 |
"step": 110
|
| 171 |
},
|
| 172 |
{
|
| 173 |
-
"epoch": 0.
|
| 174 |
-
"grad_norm": 0.
|
| 175 |
-
"learning_rate": 1.
|
| 176 |
-
"loss": 0.
|
| 177 |
"step": 115
|
| 178 |
},
|
| 179 |
{
|
| 180 |
-
"epoch": 0.
|
| 181 |
-
"grad_norm": 0.
|
| 182 |
-
"learning_rate": 1.
|
| 183 |
-
"loss": 0.
|
| 184 |
"step": 120
|
| 185 |
},
|
| 186 |
{
|
| 187 |
-
"epoch": 0.
|
| 188 |
-
"grad_norm": 0.
|
| 189 |
-
"learning_rate": 1.
|
| 190 |
-
"loss": 0.
|
| 191 |
"step": 125
|
| 192 |
},
|
| 193 |
{
|
| 194 |
-
"epoch": 0.
|
| 195 |
-
"grad_norm": 0.
|
| 196 |
-
"learning_rate": 1.
|
| 197 |
-
"loss": 0.
|
| 198 |
"step": 130
|
| 199 |
},
|
| 200 |
{
|
| 201 |
-
"epoch": 0.
|
| 202 |
-
"grad_norm": 0.
|
| 203 |
-
"learning_rate": 1.
|
| 204 |
-
"loss": 0.
|
| 205 |
"step": 135
|
| 206 |
},
|
| 207 |
{
|
| 208 |
-
"epoch": 0.
|
| 209 |
-
"grad_norm": 0.
|
| 210 |
-
"learning_rate": 1.
|
| 211 |
-
"loss": 0.
|
| 212 |
"step": 140
|
| 213 |
},
|
| 214 |
{
|
| 215 |
-
"epoch": 0.
|
| 216 |
-
"grad_norm": 0.
|
| 217 |
-
"learning_rate": 1.
|
| 218 |
-
"loss": 0.
|
| 219 |
"step": 145
|
| 220 |
},
|
| 221 |
{
|
| 222 |
-
"epoch": 0.
|
| 223 |
-
"grad_norm": 0.
|
| 224 |
-
"learning_rate": 1.
|
| 225 |
-
"loss": 0.
|
| 226 |
"step": 150
|
| 227 |
},
|
| 228 |
{
|
| 229 |
-
"epoch": 0.
|
| 230 |
-
"grad_norm": 0.
|
| 231 |
-
"learning_rate": 1.
|
| 232 |
-
"loss": 0.
|
| 233 |
"step": 155
|
| 234 |
},
|
| 235 |
{
|
| 236 |
-
"epoch": 0.
|
| 237 |
-
"grad_norm": 0.
|
| 238 |
-
"learning_rate": 1.
|
| 239 |
-
"loss": 0.
|
| 240 |
"step": 160
|
| 241 |
},
|
| 242 |
{
|
| 243 |
-
"epoch": 0.
|
| 244 |
-
"grad_norm": 0.
|
| 245 |
-
"learning_rate": 1.
|
| 246 |
-
"loss": 0.
|
| 247 |
"step": 165
|
| 248 |
},
|
| 249 |
{
|
| 250 |
-
"epoch": 0.
|
| 251 |
-
"grad_norm": 0.
|
| 252 |
-
"learning_rate": 1.
|
| 253 |
-
"loss": 0.
|
| 254 |
"step": 170
|
| 255 |
},
|
| 256 |
{
|
| 257 |
-
"epoch": 0.
|
| 258 |
-
"grad_norm": 0.
|
| 259 |
-
"learning_rate": 1.
|
| 260 |
-
"loss": 0.
|
| 261 |
"step": 175
|
| 262 |
},
|
| 263 |
{
|
| 264 |
-
"epoch": 0.
|
| 265 |
-
"grad_norm": 0.
|
| 266 |
-
"learning_rate": 1.
|
| 267 |
-
"loss": 0.
|
| 268 |
"step": 180
|
| 269 |
},
|
| 270 |
{
|
| 271 |
-
"epoch": 0.
|
| 272 |
-
"grad_norm": 0.
|
| 273 |
-
"learning_rate": 1.
|
| 274 |
-
"loss": 0.
|
| 275 |
"step": 185
|
| 276 |
},
|
| 277 |
{
|
| 278 |
-
"epoch": 0.
|
| 279 |
-
"grad_norm": 0.
|
| 280 |
-
"learning_rate": 1.
|
| 281 |
-
"loss": 0.
|
| 282 |
"step": 190
|
| 283 |
},
|
| 284 |
{
|
| 285 |
-
"epoch": 0.
|
| 286 |
-
"grad_norm": 0.
|
| 287 |
-
"learning_rate": 1.
|
| 288 |
-
"loss": 0.
|
| 289 |
"step": 195
|
| 290 |
},
|
| 291 |
{
|
| 292 |
-
"epoch": 0.
|
| 293 |
-
"grad_norm": 0.
|
| 294 |
-
"learning_rate": 1.
|
| 295 |
-
"loss": 0.
|
| 296 |
"step": 200
|
| 297 |
},
|
| 298 |
{
|
| 299 |
-
"epoch": 0.
|
| 300 |
-
"grad_norm": 0.
|
| 301 |
-
"learning_rate": 1.
|
| 302 |
-
"loss": 0.
|
| 303 |
"step": 205
|
| 304 |
},
|
| 305 |
{
|
| 306 |
-
"epoch": 0.
|
| 307 |
-
"grad_norm": 0.
|
| 308 |
-
"learning_rate": 1.
|
| 309 |
-
"loss": 0.
|
| 310 |
"step": 210
|
| 311 |
},
|
| 312 |
{
|
| 313 |
-
"epoch": 0.
|
| 314 |
-
"grad_norm": 0.
|
| 315 |
-
"learning_rate": 1.
|
| 316 |
-
"loss": 0.
|
| 317 |
"step": 215
|
| 318 |
},
|
| 319 |
{
|
| 320 |
-
"epoch": 0.
|
| 321 |
-
"grad_norm": 0.
|
| 322 |
-
"learning_rate": 1.
|
| 323 |
-
"loss": 0.
|
| 324 |
"step": 220
|
| 325 |
},
|
| 326 |
{
|
| 327 |
-
"epoch": 0.
|
| 328 |
-
"grad_norm": 0.
|
| 329 |
-
"learning_rate": 1.
|
| 330 |
-
"loss": 0.
|
| 331 |
"step": 225
|
| 332 |
},
|
| 333 |
{
|
| 334 |
-
"epoch": 0.
|
| 335 |
-
"grad_norm": 0.
|
| 336 |
-
"learning_rate": 1.
|
| 337 |
-
"loss": 0.
|
| 338 |
"step": 230
|
| 339 |
},
|
| 340 |
{
|
| 341 |
-
"epoch": 0.
|
| 342 |
-
"grad_norm": 0.
|
| 343 |
-
"learning_rate": 1.
|
| 344 |
-
"loss": 0.
|
| 345 |
"step": 235
|
| 346 |
},
|
| 347 |
{
|
| 348 |
-
"epoch": 0.
|
| 349 |
-
"grad_norm": 0.
|
| 350 |
-
"learning_rate": 1.
|
| 351 |
-
"loss": 0.
|
| 352 |
"step": 240
|
| 353 |
},
|
| 354 |
{
|
| 355 |
-
"epoch": 0.
|
| 356 |
-
"grad_norm": 0.
|
| 357 |
-
"learning_rate": 1.
|
| 358 |
-
"loss": 0.
|
| 359 |
"step": 245
|
| 360 |
},
|
| 361 |
{
|
| 362 |
-
"epoch": 0.
|
| 363 |
-
"grad_norm": 0.
|
| 364 |
-
"learning_rate": 1.
|
| 365 |
-
"loss": 0.
|
| 366 |
"step": 250
|
| 367 |
},
|
| 368 |
{
|
| 369 |
-
"epoch": 0.
|
| 370 |
-
"grad_norm": 0.
|
| 371 |
-
"learning_rate": 1.
|
| 372 |
-
"loss": 0.
|
| 373 |
"step": 255
|
| 374 |
},
|
| 375 |
{
|
| 376 |
-
"epoch": 0.
|
| 377 |
-
"grad_norm": 0.
|
| 378 |
-
"learning_rate": 1.
|
| 379 |
-
"loss": 0.
|
| 380 |
"step": 260
|
| 381 |
},
|
| 382 |
{
|
| 383 |
-
"epoch": 0.
|
| 384 |
-
"grad_norm": 0.
|
| 385 |
-
"learning_rate": 1.
|
| 386 |
-
"loss": 0.
|
| 387 |
"step": 265
|
| 388 |
},
|
| 389 |
{
|
| 390 |
-
"epoch": 0.
|
| 391 |
-
"grad_norm": 0.
|
| 392 |
-
"learning_rate": 1.
|
| 393 |
-
"loss": 0.
|
| 394 |
"step": 270
|
| 395 |
},
|
| 396 |
{
|
| 397 |
-
"epoch": 0.
|
| 398 |
-
"grad_norm": 0.
|
| 399 |
-
"learning_rate": 1.
|
| 400 |
-
"loss": 0.
|
| 401 |
"step": 275
|
| 402 |
},
|
| 403 |
{
|
| 404 |
-
"epoch": 0.
|
| 405 |
-
"grad_norm": 0.
|
| 406 |
-
"learning_rate": 1.
|
| 407 |
-
"loss": 0.
|
| 408 |
"step": 280
|
| 409 |
},
|
| 410 |
{
|
| 411 |
-
"epoch": 0.
|
| 412 |
-
"grad_norm": 0.
|
| 413 |
-
"learning_rate": 1.
|
| 414 |
-
"loss": 0.
|
| 415 |
"step": 285
|
| 416 |
},
|
| 417 |
{
|
| 418 |
-
"epoch": 0.
|
| 419 |
-
"grad_norm": 0.
|
| 420 |
-
"learning_rate": 1.
|
| 421 |
-
"loss": 0.
|
| 422 |
"step": 290
|
| 423 |
},
|
| 424 |
{
|
| 425 |
-
"epoch": 0.
|
| 426 |
-
"grad_norm": 0.
|
| 427 |
-
"learning_rate": 1.
|
| 428 |
-
"loss": 0.
|
| 429 |
"step": 295
|
| 430 |
},
|
| 431 |
{
|
| 432 |
-
"epoch": 0.
|
| 433 |
-
"grad_norm": 0.
|
| 434 |
-
"learning_rate": 1.
|
| 435 |
-
"loss": 0.
|
| 436 |
"step": 300
|
| 437 |
},
|
| 438 |
{
|
| 439 |
-
"epoch": 0.
|
| 440 |
-
"grad_norm": 0.
|
| 441 |
-
"learning_rate": 1.
|
| 442 |
-
"loss": 0.
|
| 443 |
"step": 305
|
| 444 |
},
|
| 445 |
{
|
| 446 |
-
"epoch": 0.
|
| 447 |
-
"grad_norm": 0.
|
| 448 |
-
"learning_rate": 1.
|
| 449 |
-
"loss": 0.
|
| 450 |
"step": 310
|
| 451 |
},
|
| 452 |
{
|
| 453 |
-
"epoch": 0.
|
| 454 |
-
"grad_norm": 0.
|
| 455 |
-
"learning_rate": 1.
|
| 456 |
-
"loss": 0.
|
| 457 |
"step": 315
|
| 458 |
},
|
| 459 |
{
|
| 460 |
-
"epoch": 0.
|
| 461 |
-
"grad_norm": 0.
|
| 462 |
-
"learning_rate": 1.
|
| 463 |
-
"loss": 0.
|
| 464 |
"step": 320
|
| 465 |
},
|
| 466 |
{
|
| 467 |
-
"epoch": 0.
|
| 468 |
-
"grad_norm": 0.
|
| 469 |
-
"learning_rate": 1.
|
| 470 |
-
"loss": 0.
|
| 471 |
"step": 325
|
| 472 |
},
|
| 473 |
{
|
| 474 |
-
"epoch": 0.
|
| 475 |
-
"grad_norm": 0.
|
| 476 |
-
"learning_rate": 1.
|
| 477 |
-
"loss": 0.
|
| 478 |
"step": 330
|
| 479 |
},
|
| 480 |
{
|
| 481 |
-
"epoch": 0.
|
| 482 |
-
"grad_norm": 0.
|
| 483 |
-
"learning_rate": 1.
|
| 484 |
-
"loss": 0.
|
| 485 |
"step": 335
|
| 486 |
},
|
| 487 |
{
|
| 488 |
-
"epoch": 0.
|
| 489 |
-
"grad_norm": 0.
|
| 490 |
-
"learning_rate": 1.
|
| 491 |
-
"loss": 0.
|
| 492 |
"step": 340
|
| 493 |
},
|
| 494 |
{
|
| 495 |
-
"epoch": 0.
|
| 496 |
-
"grad_norm": 0.
|
| 497 |
-
"learning_rate": 1.
|
| 498 |
-
"loss": 0.
|
| 499 |
"step": 345
|
| 500 |
},
|
| 501 |
{
|
| 502 |
-
"epoch": 0.
|
| 503 |
-
"grad_norm": 0.
|
| 504 |
-
"learning_rate": 1.
|
| 505 |
-
"loss": 0.
|
| 506 |
"step": 350
|
| 507 |
},
|
| 508 |
{
|
| 509 |
-
"epoch": 0.
|
| 510 |
-
"grad_norm": 0.
|
| 511 |
-
"learning_rate": 1.
|
| 512 |
-
"loss": 0.
|
| 513 |
"step": 355
|
| 514 |
},
|
| 515 |
{
|
| 516 |
-
"epoch": 0.
|
| 517 |
-
"grad_norm": 0.
|
| 518 |
-
"learning_rate": 1.
|
| 519 |
-
"loss": 0.
|
| 520 |
"step": 360
|
| 521 |
},
|
| 522 |
{
|
| 523 |
-
"epoch": 0.
|
| 524 |
-
"grad_norm": 0.
|
| 525 |
-
"learning_rate": 1.
|
| 526 |
-
"loss": 0.
|
| 527 |
"step": 365
|
| 528 |
},
|
| 529 |
{
|
| 530 |
-
"epoch": 0.
|
| 531 |
-
"grad_norm": 0.
|
| 532 |
-
"learning_rate": 1.
|
| 533 |
-
"loss": 0.
|
| 534 |
"step": 370
|
| 535 |
},
|
| 536 |
{
|
| 537 |
-
"epoch": 0.
|
| 538 |
-
"grad_norm": 0.
|
| 539 |
-
"learning_rate": 1.
|
| 540 |
-
"loss": 0.
|
| 541 |
"step": 375
|
| 542 |
},
|
| 543 |
{
|
| 544 |
-
"epoch": 0.
|
| 545 |
-
"grad_norm": 0.
|
| 546 |
-
"learning_rate": 1.
|
| 547 |
-
"loss": 0.
|
| 548 |
"step": 380
|
| 549 |
},
|
| 550 |
{
|
| 551 |
-
"epoch": 0.
|
| 552 |
-
"grad_norm": 0.
|
| 553 |
-
"learning_rate": 1.
|
| 554 |
-
"loss": 0.
|
| 555 |
"step": 385
|
| 556 |
},
|
| 557 |
{
|
| 558 |
-
"epoch": 0.
|
| 559 |
-
"grad_norm": 0.
|
| 560 |
-
"learning_rate": 1.
|
| 561 |
-
"loss": 0.
|
| 562 |
"step": 390
|
| 563 |
},
|
| 564 |
{
|
| 565 |
-
"epoch": 0.
|
| 566 |
-
"grad_norm": 0.
|
| 567 |
-
"learning_rate": 1.
|
| 568 |
-
"loss": 0.
|
| 569 |
"step": 395
|
| 570 |
},
|
| 571 |
{
|
| 572 |
-
"epoch": 0.
|
| 573 |
-
"grad_norm": 0.
|
| 574 |
-
"learning_rate": 1.
|
| 575 |
-
"loss": 0.
|
| 576 |
"step": 400
|
| 577 |
},
|
| 578 |
{
|
| 579 |
-
"epoch": 0.
|
| 580 |
-
"grad_norm": 0.
|
| 581 |
-
"learning_rate": 1.
|
| 582 |
-
"loss": 0.
|
| 583 |
"step": 405
|
| 584 |
},
|
| 585 |
{
|
| 586 |
-
"epoch": 0.
|
| 587 |
-
"grad_norm": 0.
|
| 588 |
-
"learning_rate": 1.
|
| 589 |
-
"loss": 0.
|
| 590 |
"step": 410
|
| 591 |
},
|
| 592 |
{
|
| 593 |
-
"epoch": 0.
|
| 594 |
-
"grad_norm": 0.
|
| 595 |
-
"learning_rate": 1.
|
| 596 |
-
"loss": 0.
|
| 597 |
"step": 415
|
| 598 |
},
|
| 599 |
{
|
| 600 |
-
"epoch": 0.
|
| 601 |
-
"grad_norm": 0.
|
| 602 |
-
"learning_rate": 1.
|
| 603 |
-
"loss": 0.
|
| 604 |
"step": 420
|
| 605 |
},
|
| 606 |
{
|
| 607 |
-
"epoch": 0.
|
| 608 |
-
"grad_norm": 0.
|
| 609 |
-
"learning_rate": 1.
|
| 610 |
-
"loss": 0.
|
| 611 |
"step": 425
|
| 612 |
},
|
| 613 |
{
|
| 614 |
-
"epoch": 0.
|
| 615 |
-
"grad_norm": 0.
|
| 616 |
-
"learning_rate": 1.
|
| 617 |
-
"loss": 0.
|
| 618 |
"step": 430
|
| 619 |
},
|
| 620 |
{
|
| 621 |
-
"epoch": 0.
|
| 622 |
-
"grad_norm": 0.
|
| 623 |
-
"learning_rate": 1.
|
| 624 |
-
"loss": 0.
|
| 625 |
"step": 435
|
| 626 |
},
|
| 627 |
{
|
| 628 |
-
"epoch": 0.
|
| 629 |
-
"grad_norm": 0.
|
| 630 |
-
"learning_rate": 1.
|
| 631 |
-
"loss": 0.
|
| 632 |
"step": 440
|
| 633 |
},
|
| 634 |
{
|
| 635 |
-
"epoch": 0.
|
| 636 |
-
"grad_norm": 0.
|
| 637 |
-
"learning_rate": 1.
|
| 638 |
-
"loss": 0.
|
| 639 |
"step": 445
|
| 640 |
},
|
| 641 |
{
|
| 642 |
-
"epoch": 0.
|
| 643 |
-
"
|
| 644 |
-
"
|
| 645 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 646 |
"step": 450
|
| 647 |
},
|
| 648 |
{
|
| 649 |
-
"epoch":
|
| 650 |
-
"grad_norm": 0.
|
| 651 |
-
"learning_rate": 1.
|
| 652 |
-
"loss": 0.
|
| 653 |
"step": 455
|
| 654 |
},
|
| 655 |
{
|
| 656 |
-
"epoch":
|
| 657 |
-
"grad_norm": 0.
|
| 658 |
-
"learning_rate": 1.
|
| 659 |
-
"loss": 0.
|
| 660 |
"step": 460
|
| 661 |
},
|
| 662 |
{
|
| 663 |
-
"epoch":
|
| 664 |
-
"grad_norm": 0.
|
| 665 |
-
"learning_rate": 1.
|
| 666 |
-
"loss": 0.
|
| 667 |
"step": 465
|
| 668 |
},
|
| 669 |
{
|
| 670 |
-
"epoch":
|
| 671 |
-
"grad_norm": 0.
|
| 672 |
-
"learning_rate": 1.
|
| 673 |
-
"loss": 0.
|
| 674 |
"step": 470
|
| 675 |
},
|
| 676 |
{
|
| 677 |
-
"epoch":
|
| 678 |
-
"grad_norm": 0.
|
| 679 |
-
"learning_rate": 1.
|
| 680 |
-
"loss": 0.
|
| 681 |
"step": 475
|
| 682 |
},
|
| 683 |
{
|
| 684 |
-
"epoch":
|
| 685 |
-
"grad_norm": 0.
|
| 686 |
-
"learning_rate": 1.
|
| 687 |
-
"loss": 0.
|
| 688 |
"step": 480
|
| 689 |
},
|
| 690 |
{
|
| 691 |
-
"epoch":
|
| 692 |
-
"grad_norm": 0.
|
| 693 |
-
"learning_rate": 1.
|
| 694 |
-
"loss": 0.
|
| 695 |
"step": 485
|
| 696 |
},
|
| 697 |
{
|
| 698 |
-
"epoch":
|
| 699 |
-
"grad_norm": 0.
|
| 700 |
-
"learning_rate": 1.
|
| 701 |
-
"loss": 0.
|
| 702 |
"step": 490
|
| 703 |
},
|
| 704 |
{
|
| 705 |
-
"epoch":
|
| 706 |
-
"grad_norm": 0.
|
| 707 |
-
"learning_rate":
|
| 708 |
-
"loss": 0.
|
| 709 |
"step": 495
|
| 710 |
},
|
| 711 |
{
|
| 712 |
-
"epoch":
|
| 713 |
-
"grad_norm": 0.
|
| 714 |
-
"learning_rate":
|
| 715 |
-
"loss": 0.
|
| 716 |
"step": 500
|
| 717 |
},
|
| 718 |
{
|
| 719 |
-
"epoch":
|
| 720 |
-
"grad_norm": 0.
|
| 721 |
-
"learning_rate":
|
| 722 |
-
"loss": 0.
|
| 723 |
"step": 505
|
| 724 |
},
|
| 725 |
{
|
| 726 |
-
"epoch":
|
| 727 |
-
"grad_norm": 0.
|
| 728 |
-
"learning_rate":
|
| 729 |
-
"loss": 0.
|
| 730 |
"step": 510
|
| 731 |
},
|
| 732 |
{
|
| 733 |
-
"epoch":
|
| 734 |
-
"grad_norm": 0.
|
| 735 |
-
"learning_rate":
|
| 736 |
-
"loss": 0.
|
| 737 |
"step": 515
|
| 738 |
},
|
| 739 |
{
|
| 740 |
-
"epoch":
|
| 741 |
-
"grad_norm": 0.
|
| 742 |
-
"learning_rate":
|
| 743 |
-
"loss": 0.
|
| 744 |
"step": 520
|
| 745 |
},
|
| 746 |
{
|
| 747 |
-
"epoch":
|
| 748 |
-
"grad_norm": 0.
|
| 749 |
-
"learning_rate":
|
| 750 |
-
"loss": 0.
|
| 751 |
"step": 525
|
| 752 |
},
|
| 753 |
{
|
| 754 |
-
"epoch":
|
| 755 |
-
"grad_norm": 0.
|
| 756 |
-
"learning_rate":
|
| 757 |
-
"loss": 0.
|
| 758 |
"step": 530
|
| 759 |
},
|
| 760 |
{
|
| 761 |
-
"epoch":
|
| 762 |
-
"grad_norm": 0.
|
| 763 |
-
"learning_rate":
|
| 764 |
-
"loss": 0.
|
| 765 |
"step": 535
|
| 766 |
},
|
| 767 |
{
|
| 768 |
-
"epoch":
|
| 769 |
-
"grad_norm": 0.
|
| 770 |
-
"learning_rate":
|
| 771 |
-
"loss": 0.
|
| 772 |
"step": 540
|
| 773 |
},
|
| 774 |
{
|
| 775 |
-
"epoch": 1.
|
| 776 |
-
"
|
| 777 |
-
"
|
| 778 |
-
"
|
| 779 |
-
"eval_steps_per_second": 2.189,
|
| 780 |
-
"step": 543
|
| 781 |
-
},
|
| 782 |
-
{
|
| 783 |
-
"epoch": 1.003683241252302,
|
| 784 |
-
"grad_norm": 0.8608149523736537,
|
| 785 |
-
"learning_rate": 1.1680156841418911e-05,
|
| 786 |
-
"loss": 0.725,
|
| 787 |
"step": 545
|
| 788 |
},
|
| 789 |
{
|
| 790 |
-
"epoch": 1.
|
| 791 |
-
"grad_norm": 0.
|
| 792 |
-
"learning_rate":
|
| 793 |
-
"loss": 0.
|
| 794 |
"step": 550
|
| 795 |
},
|
| 796 |
{
|
| 797 |
-
"epoch": 1.
|
| 798 |
-
"grad_norm": 0.
|
| 799 |
-
"learning_rate":
|
| 800 |
-
"loss": 0.
|
| 801 |
"step": 555
|
| 802 |
},
|
| 803 |
{
|
| 804 |
-
"epoch": 1.
|
| 805 |
-
"grad_norm": 0.
|
| 806 |
-
"learning_rate":
|
| 807 |
-
"loss": 0.
|
| 808 |
"step": 560
|
| 809 |
},
|
| 810 |
{
|
| 811 |
-
"epoch": 1.
|
| 812 |
-
"grad_norm": 0.
|
| 813 |
-
"learning_rate":
|
| 814 |
-
"loss": 0.
|
| 815 |
"step": 565
|
| 816 |
},
|
| 817 |
{
|
| 818 |
-
"epoch": 1.
|
| 819 |
-
"grad_norm": 0.
|
| 820 |
-
"learning_rate":
|
| 821 |
-
"loss": 0.
|
| 822 |
"step": 570
|
| 823 |
},
|
| 824 |
{
|
| 825 |
-
"epoch": 1.
|
| 826 |
-
"grad_norm":
|
| 827 |
-
"learning_rate":
|
| 828 |
-
"loss": 0.
|
| 829 |
"step": 575
|
| 830 |
},
|
| 831 |
{
|
| 832 |
-
"epoch": 1.
|
| 833 |
-
"grad_norm": 0.
|
| 834 |
-
"learning_rate":
|
| 835 |
-
"loss": 0.
|
| 836 |
"step": 580
|
| 837 |
},
|
| 838 |
{
|
| 839 |
-
"epoch": 1.
|
| 840 |
-
"grad_norm": 0.
|
| 841 |
-
"learning_rate":
|
| 842 |
-
"loss": 0.
|
| 843 |
"step": 585
|
| 844 |
},
|
| 845 |
{
|
| 846 |
-
"epoch": 1.
|
| 847 |
-
"grad_norm": 0.
|
| 848 |
-
"learning_rate":
|
| 849 |
-
"loss": 0.
|
| 850 |
"step": 590
|
| 851 |
},
|
| 852 |
{
|
| 853 |
-
"epoch": 1.
|
| 854 |
-
"grad_norm": 0.
|
| 855 |
-
"learning_rate":
|
| 856 |
-
"loss": 0.
|
| 857 |
"step": 595
|
| 858 |
},
|
| 859 |
{
|
| 860 |
-
"epoch": 1.
|
| 861 |
-
"grad_norm": 0.
|
| 862 |
-
"learning_rate":
|
| 863 |
-
"loss": 0.
|
| 864 |
"step": 600
|
| 865 |
},
|
| 866 |
{
|
| 867 |
-
"epoch": 1.
|
| 868 |
-
"grad_norm": 0.
|
| 869 |
-
"learning_rate":
|
| 870 |
-
"loss": 0.
|
| 871 |
"step": 605
|
| 872 |
},
|
| 873 |
{
|
| 874 |
-
"epoch": 1.
|
| 875 |
-
"grad_norm": 0.
|
| 876 |
-
"learning_rate":
|
| 877 |
-
"loss": 0.
|
| 878 |
"step": 610
|
| 879 |
},
|
| 880 |
{
|
| 881 |
-
"epoch": 1.
|
| 882 |
-
"grad_norm": 0.
|
| 883 |
-
"learning_rate":
|
| 884 |
-
"loss": 0.
|
| 885 |
"step": 615
|
| 886 |
},
|
| 887 |
{
|
| 888 |
-
"epoch": 1.
|
| 889 |
-
"grad_norm": 0.
|
| 890 |
-
"learning_rate":
|
| 891 |
-
"loss": 0.
|
| 892 |
"step": 620
|
| 893 |
},
|
| 894 |
{
|
| 895 |
-
"epoch": 1.
|
| 896 |
-
"grad_norm": 0.
|
| 897 |
-
"learning_rate":
|
| 898 |
-
"loss": 0.
|
| 899 |
"step": 625
|
| 900 |
},
|
| 901 |
{
|
| 902 |
-
"epoch": 1.
|
| 903 |
-
"grad_norm": 0.
|
| 904 |
-
"learning_rate":
|
| 905 |
-
"loss": 0.
|
| 906 |
"step": 630
|
| 907 |
},
|
| 908 |
{
|
| 909 |
-
"epoch": 1.
|
| 910 |
-
"grad_norm": 0.
|
| 911 |
-
"learning_rate":
|
| 912 |
-
"loss": 0.
|
| 913 |
"step": 635
|
| 914 |
},
|
| 915 |
{
|
| 916 |
-
"epoch": 1.
|
| 917 |
-
"grad_norm": 0.
|
| 918 |
-
"learning_rate":
|
| 919 |
-
"loss": 0.
|
| 920 |
"step": 640
|
| 921 |
},
|
| 922 |
{
|
| 923 |
-
"epoch": 1.
|
| 924 |
-
"grad_norm": 0.
|
| 925 |
-
"learning_rate":
|
| 926 |
-
"loss": 0.
|
| 927 |
"step": 645
|
| 928 |
},
|
| 929 |
{
|
| 930 |
-
"epoch": 1.
|
| 931 |
-
"grad_norm": 0.
|
| 932 |
-
"learning_rate":
|
| 933 |
-
"loss": 0.
|
| 934 |
"step": 650
|
| 935 |
},
|
| 936 |
{
|
| 937 |
-
"epoch": 1.
|
| 938 |
-
"grad_norm": 0.
|
| 939 |
-
"learning_rate":
|
| 940 |
-
"loss": 0.
|
| 941 |
"step": 655
|
| 942 |
},
|
| 943 |
{
|
| 944 |
-
"epoch": 1.
|
| 945 |
-
"grad_norm": 0.
|
| 946 |
-
"learning_rate":
|
| 947 |
-
"loss": 0.
|
| 948 |
"step": 660
|
| 949 |
},
|
| 950 |
{
|
| 951 |
-
"epoch": 1.
|
| 952 |
-
"grad_norm": 0.
|
| 953 |
-
"learning_rate":
|
| 954 |
-
"loss": 0.
|
| 955 |
"step": 665
|
| 956 |
},
|
| 957 |
{
|
| 958 |
-
"epoch": 1.
|
| 959 |
-
"grad_norm": 0.
|
| 960 |
-
"learning_rate":
|
| 961 |
-
"loss": 0.
|
| 962 |
"step": 670
|
| 963 |
},
|
| 964 |
{
|
| 965 |
-
"epoch": 1.
|
| 966 |
-
"grad_norm": 0.
|
| 967 |
-
"learning_rate":
|
| 968 |
-
"loss": 0.
|
| 969 |
"step": 675
|
| 970 |
},
|
| 971 |
{
|
| 972 |
-
"epoch": 1.
|
| 973 |
-
"grad_norm": 0.
|
| 974 |
-
"learning_rate":
|
| 975 |
-
"loss": 0.
|
| 976 |
"step": 680
|
| 977 |
},
|
| 978 |
{
|
| 979 |
-
"epoch": 1.
|
| 980 |
-
"grad_norm": 0.
|
| 981 |
-
"learning_rate":
|
| 982 |
-
"loss": 0.
|
| 983 |
"step": 685
|
| 984 |
},
|
| 985 |
{
|
| 986 |
-
"epoch": 1.
|
| 987 |
-
"grad_norm": 0.
|
| 988 |
-
"learning_rate":
|
| 989 |
-
"loss": 0.
|
| 990 |
"step": 690
|
| 991 |
},
|
| 992 |
{
|
| 993 |
-
"epoch": 1.
|
| 994 |
-
"grad_norm": 0.
|
| 995 |
-
"learning_rate":
|
| 996 |
-
"loss": 0.
|
| 997 |
"step": 695
|
| 998 |
},
|
| 999 |
{
|
| 1000 |
-
"epoch": 1.
|
| 1001 |
-
"grad_norm": 0.
|
| 1002 |
-
"learning_rate":
|
| 1003 |
-
"loss": 0.
|
| 1004 |
"step": 700
|
| 1005 |
},
|
| 1006 |
{
|
| 1007 |
-
"epoch": 1.
|
| 1008 |
-
"grad_norm": 0.
|
| 1009 |
-
"learning_rate":
|
| 1010 |
-
"loss": 0.
|
| 1011 |
"step": 705
|
| 1012 |
},
|
| 1013 |
{
|
| 1014 |
-
"epoch": 1.
|
| 1015 |
-
"grad_norm": 0.
|
| 1016 |
-
"learning_rate":
|
| 1017 |
-
"loss": 0.
|
| 1018 |
"step": 710
|
| 1019 |
},
|
| 1020 |
{
|
| 1021 |
-
"epoch": 1.
|
| 1022 |
-
"grad_norm": 0.
|
| 1023 |
-
"learning_rate":
|
| 1024 |
-
"loss": 0.
|
| 1025 |
"step": 715
|
| 1026 |
},
|
| 1027 |
{
|
| 1028 |
-
"epoch": 1.
|
| 1029 |
-
"grad_norm": 0.
|
| 1030 |
-
"learning_rate":
|
| 1031 |
-
"loss": 0.
|
| 1032 |
"step": 720
|
| 1033 |
},
|
| 1034 |
{
|
| 1035 |
-
"epoch": 1.
|
| 1036 |
-
"grad_norm": 0.
|
| 1037 |
-
"learning_rate":
|
| 1038 |
-
"loss": 0.
|
| 1039 |
"step": 725
|
| 1040 |
},
|
| 1041 |
{
|
| 1042 |
-
"epoch": 1.
|
| 1043 |
-
"grad_norm": 0.
|
| 1044 |
-
"learning_rate":
|
| 1045 |
-
"loss": 0.
|
| 1046 |
"step": 730
|
| 1047 |
},
|
| 1048 |
{
|
| 1049 |
-
"epoch": 1.
|
| 1050 |
-
"grad_norm": 0.
|
| 1051 |
-
"learning_rate":
|
| 1052 |
-
"loss": 0.
|
| 1053 |
"step": 735
|
| 1054 |
},
|
| 1055 |
{
|
| 1056 |
-
"epoch": 1.
|
| 1057 |
-
"grad_norm": 0.
|
| 1058 |
-
"learning_rate":
|
| 1059 |
-
"loss": 0.
|
| 1060 |
"step": 740
|
| 1061 |
},
|
| 1062 |
{
|
| 1063 |
-
"epoch": 1.
|
| 1064 |
-
"grad_norm": 0.
|
| 1065 |
-
"learning_rate":
|
| 1066 |
-
"loss": 0.
|
| 1067 |
"step": 745
|
| 1068 |
},
|
| 1069 |
{
|
| 1070 |
-
"epoch": 1.
|
| 1071 |
-
"grad_norm": 0.
|
| 1072 |
-
"learning_rate":
|
| 1073 |
-
"loss": 0.
|
| 1074 |
"step": 750
|
| 1075 |
},
|
| 1076 |
{
|
| 1077 |
-
"epoch": 1.
|
| 1078 |
-
"grad_norm": 0.
|
| 1079 |
-
"learning_rate":
|
| 1080 |
-
"loss": 0.
|
| 1081 |
"step": 755
|
| 1082 |
},
|
| 1083 |
{
|
| 1084 |
-
"epoch": 1.
|
| 1085 |
-
"grad_norm": 0.
|
| 1086 |
-
"learning_rate":
|
| 1087 |
-
"loss": 0.
|
| 1088 |
"step": 760
|
| 1089 |
},
|
| 1090 |
{
|
| 1091 |
-
"epoch": 1.
|
| 1092 |
-
"grad_norm": 0.
|
| 1093 |
-
"learning_rate":
|
| 1094 |
-
"loss": 0.
|
| 1095 |
"step": 765
|
| 1096 |
},
|
| 1097 |
{
|
| 1098 |
-
"epoch": 1.
|
| 1099 |
-
"grad_norm": 0.
|
| 1100 |
-
"learning_rate":
|
| 1101 |
-
"loss": 0.
|
| 1102 |
"step": 770
|
| 1103 |
},
|
| 1104 |
{
|
| 1105 |
-
"epoch": 1.
|
| 1106 |
-
"grad_norm": 0.
|
| 1107 |
-
"learning_rate":
|
| 1108 |
-
"loss": 0.
|
| 1109 |
"step": 775
|
| 1110 |
},
|
| 1111 |
{
|
| 1112 |
-
"epoch": 1.
|
| 1113 |
-
"grad_norm": 0.
|
| 1114 |
-
"learning_rate":
|
| 1115 |
-
"loss": 0.
|
| 1116 |
"step": 780
|
| 1117 |
},
|
| 1118 |
{
|
| 1119 |
-
"epoch": 1.
|
| 1120 |
-
"grad_norm": 0.
|
| 1121 |
-
"learning_rate":
|
| 1122 |
-
"loss": 0.
|
| 1123 |
"step": 785
|
| 1124 |
},
|
| 1125 |
{
|
| 1126 |
-
"epoch": 1.
|
| 1127 |
-
"grad_norm": 0.
|
| 1128 |
-
"learning_rate":
|
| 1129 |
-
"loss": 0.
|
| 1130 |
"step": 790
|
| 1131 |
},
|
| 1132 |
{
|
| 1133 |
-
"epoch": 1.
|
| 1134 |
-
"grad_norm": 0.
|
| 1135 |
-
"learning_rate":
|
| 1136 |
-
"loss": 0.
|
| 1137 |
"step": 795
|
| 1138 |
},
|
| 1139 |
{
|
| 1140 |
-
"epoch": 1.
|
| 1141 |
-
"grad_norm": 0.
|
| 1142 |
-
"learning_rate":
|
| 1143 |
-
"loss": 0.
|
| 1144 |
"step": 800
|
| 1145 |
},
|
| 1146 |
{
|
| 1147 |
-
"epoch": 1.
|
| 1148 |
-
"grad_norm": 0.
|
| 1149 |
-
"learning_rate":
|
| 1150 |
-
"loss": 0.
|
| 1151 |
"step": 805
|
| 1152 |
},
|
| 1153 |
{
|
| 1154 |
-
"epoch": 1.
|
| 1155 |
-
"grad_norm": 0.
|
| 1156 |
-
"learning_rate":
|
| 1157 |
-
"loss": 0.
|
| 1158 |
"step": 810
|
| 1159 |
},
|
| 1160 |
{
|
| 1161 |
-
"epoch": 1.
|
| 1162 |
-
"grad_norm": 0.
|
| 1163 |
-
"learning_rate":
|
| 1164 |
-
"loss": 0.
|
| 1165 |
"step": 815
|
| 1166 |
},
|
| 1167 |
{
|
| 1168 |
-
"epoch": 1.
|
| 1169 |
-
"grad_norm": 0.
|
| 1170 |
-
"learning_rate":
|
| 1171 |
-
"loss": 0.
|
| 1172 |
"step": 820
|
| 1173 |
},
|
| 1174 |
{
|
| 1175 |
-
"epoch": 1.
|
| 1176 |
-
"grad_norm": 0.
|
| 1177 |
-
"learning_rate":
|
| 1178 |
-
"loss": 0.
|
| 1179 |
"step": 825
|
| 1180 |
},
|
| 1181 |
{
|
| 1182 |
-
"epoch": 1.
|
| 1183 |
-
"grad_norm": 0.
|
| 1184 |
-
"learning_rate": 3.
|
| 1185 |
-
"loss": 0.
|
| 1186 |
"step": 830
|
| 1187 |
},
|
| 1188 |
{
|
| 1189 |
-
"epoch": 1.
|
| 1190 |
-
"grad_norm": 0.
|
| 1191 |
-
"learning_rate":
|
| 1192 |
-
"loss": 0.
|
| 1193 |
"step": 835
|
| 1194 |
},
|
| 1195 |
{
|
| 1196 |
-
"epoch": 1.
|
| 1197 |
-
"grad_norm": 0.
|
| 1198 |
-
"learning_rate": 2.
|
| 1199 |
-
"loss": 0.
|
| 1200 |
"step": 840
|
| 1201 |
},
|
| 1202 |
{
|
| 1203 |
-
"epoch": 1.
|
| 1204 |
-
"grad_norm": 0.
|
| 1205 |
-
"learning_rate": 2.
|
| 1206 |
-
"loss": 0.
|
| 1207 |
"step": 845
|
| 1208 |
},
|
| 1209 |
{
|
| 1210 |
-
"epoch": 1.
|
| 1211 |
-
"grad_norm": 0.
|
| 1212 |
-
"learning_rate":
|
| 1213 |
-
"loss": 0.
|
| 1214 |
"step": 850
|
| 1215 |
},
|
| 1216 |
{
|
| 1217 |
-
"epoch": 1.
|
| 1218 |
-
"grad_norm": 0.
|
| 1219 |
-
"learning_rate":
|
| 1220 |
-
"loss": 0.
|
| 1221 |
"step": 855
|
| 1222 |
},
|
| 1223 |
{
|
| 1224 |
-
"epoch": 1.
|
| 1225 |
-
"grad_norm": 0.
|
| 1226 |
-
"learning_rate":
|
| 1227 |
-
"loss": 0.
|
| 1228 |
"step": 860
|
| 1229 |
},
|
| 1230 |
{
|
| 1231 |
-
"epoch": 1.
|
| 1232 |
-
"grad_norm": 0.
|
| 1233 |
-
"learning_rate":
|
| 1234 |
-
"loss": 0.
|
| 1235 |
"step": 865
|
| 1236 |
},
|
| 1237 |
{
|
| 1238 |
-
"epoch": 1.
|
| 1239 |
-
"grad_norm": 0.
|
| 1240 |
-
"learning_rate":
|
| 1241 |
-
"loss": 0.
|
| 1242 |
"step": 870
|
| 1243 |
},
|
| 1244 |
{
|
| 1245 |
-
"epoch": 1.
|
| 1246 |
-
"grad_norm": 0.
|
| 1247 |
-
"learning_rate":
|
| 1248 |
-
"loss": 0.
|
| 1249 |
"step": 875
|
| 1250 |
},
|
| 1251 |
{
|
| 1252 |
-
"epoch": 1.
|
| 1253 |
-
"grad_norm": 0.
|
| 1254 |
-
"learning_rate": 2.
|
| 1255 |
-
"loss": 0.
|
| 1256 |
"step": 880
|
| 1257 |
},
|
| 1258 |
{
|
| 1259 |
-
"epoch": 1.
|
| 1260 |
-
"grad_norm": 0.
|
| 1261 |
-
"learning_rate":
|
| 1262 |
-
"loss": 0.
|
| 1263 |
"step": 885
|
| 1264 |
},
|
| 1265 |
{
|
| 1266 |
-
"epoch": 1.
|
| 1267 |
-
"grad_norm": 0.
|
| 1268 |
-
"learning_rate":
|
| 1269 |
-
"loss": 0.
|
| 1270 |
"step": 890
|
| 1271 |
},
|
| 1272 |
{
|
| 1273 |
-
"epoch": 1.
|
| 1274 |
-
"grad_norm": 0.
|
| 1275 |
-
"learning_rate":
|
| 1276 |
-
"loss": 0.
|
| 1277 |
"step": 895
|
| 1278 |
},
|
| 1279 |
{
|
| 1280 |
-
"epoch": 1.
|
| 1281 |
-
"
|
| 1282 |
-
"
|
| 1283 |
-
"
|
| 1284 |
-
"
|
| 1285 |
-
|
| 1286 |
-
{
|
| 1287 |
-
"epoch": 1.6666666666666665,
|
| 1288 |
-
"grad_norm": 0.8050853357467429,
|
| 1289 |
-
"learning_rate": 1.6464309923166033e-06,
|
| 1290 |
-
"loss": 0.6601,
|
| 1291 |
-
"step": 905
|
| 1292 |
-
},
|
| 1293 |
-
{
|
| 1294 |
-
"epoch": 1.6758747697974217,
|
| 1295 |
-
"grad_norm": 0.7632083820451964,
|
| 1296 |
-
"learning_rate": 1.5591337696771247e-06,
|
| 1297 |
-
"loss": 0.6511,
|
| 1298 |
-
"step": 910
|
| 1299 |
-
},
|
| 1300 |
-
{
|
| 1301 |
-
"epoch": 1.6850828729281768,
|
| 1302 |
-
"grad_norm": 0.8160981703519061,
|
| 1303 |
-
"learning_rate": 1.4740184141144664e-06,
|
| 1304 |
-
"loss": 0.6528,
|
| 1305 |
-
"step": 915
|
| 1306 |
-
},
|
| 1307 |
-
{
|
| 1308 |
-
"epoch": 1.694290976058932,
|
| 1309 |
-
"grad_norm": 0.7909791196212368,
|
| 1310 |
-
"learning_rate": 1.391106926971496e-06,
|
| 1311 |
-
"loss": 0.6676,
|
| 1312 |
-
"step": 920
|
| 1313 |
-
},
|
| 1314 |
-
{
|
| 1315 |
-
"epoch": 1.703499079189687,
|
| 1316 |
-
"grad_norm": 0.7627803057441984,
|
| 1317 |
-
"learning_rate": 1.3104207399163816e-06,
|
| 1318 |
-
"loss": 0.6437,
|
| 1319 |
-
"step": 925
|
| 1320 |
-
},
|
| 1321 |
-
{
|
| 1322 |
-
"epoch": 1.7127071823204418,
|
| 1323 |
-
"grad_norm": 0.8629435634729274,
|
| 1324 |
-
"learning_rate": 1.2319807094027492e-06,
|
| 1325 |
-
"loss": 0.6744,
|
| 1326 |
-
"step": 930
|
| 1327 |
-
},
|
| 1328 |
-
{
|
| 1329 |
-
"epoch": 1.721915285451197,
|
| 1330 |
-
"grad_norm": 0.8022493311275769,
|
| 1331 |
-
"learning_rate": 1.1558071112785297e-06,
|
| 1332 |
-
"loss": 0.6405,
|
| 1333 |
-
"step": 935
|
| 1334 |
-
},
|
| 1335 |
-
{
|
| 1336 |
-
"epoch": 1.7311233885819521,
|
| 1337 |
-
"grad_norm": 0.8085541255802431,
|
| 1338 |
-
"learning_rate": 1.0819196355448801e-06,
|
| 1339 |
-
"loss": 0.6316,
|
| 1340 |
-
"step": 940
|
| 1341 |
-
},
|
| 1342 |
-
{
|
| 1343 |
-
"epoch": 1.7403314917127073,
|
| 1344 |
-
"grad_norm": 0.7700288703813564,
|
| 1345 |
-
"learning_rate": 1.0103373812665552e-06,
|
| 1346 |
-
"loss": 0.6616,
|
| 1347 |
-
"step": 945
|
| 1348 |
-
},
|
| 1349 |
-
{
|
| 1350 |
-
"epoch": 1.7495395948434622,
|
| 1351 |
-
"grad_norm": 0.7911373139738717,
|
| 1352 |
-
"learning_rate": 9.410788516350078e-07,
|
| 1353 |
-
"loss": 0.6412,
|
| 1354 |
-
"step": 950
|
| 1355 |
-
},
|
| 1356 |
-
{
|
| 1357 |
-
"epoch": 1.7587476979742172,
|
| 1358 |
-
"grad_norm": 0.7678764765370287,
|
| 1359 |
-
"learning_rate": 8.741619491855291e-07,
|
| 1360 |
-
"loss": 0.6426,
|
| 1361 |
-
"step": 955
|
| 1362 |
-
},
|
| 1363 |
-
{
|
| 1364 |
-
"epoch": 1.7679558011049723,
|
| 1365 |
-
"grad_norm": 0.7815749787856374,
|
| 1366 |
-
"learning_rate": 8.096039711696546e-07,
|
| 1367 |
-
"loss": 0.6515,
|
| 1368 |
-
"step": 960
|
| 1369 |
-
},
|
| 1370 |
-
{
|
| 1371 |
-
"epoch": 1.7771639042357275,
|
| 1372 |
-
"grad_norm": 0.798843530004998,
|
| 1373 |
-
"learning_rate": 7.474216050840122e-07,
|
| 1374 |
-
"loss": 0.6459,
|
| 1375 |
-
"step": 965
|
| 1376 |
-
},
|
| 1377 |
-
{
|
| 1378 |
-
"epoch": 1.7863720073664826,
|
| 1379 |
-
"grad_norm": 0.8229004013795932,
|
| 1380 |
-
"learning_rate": 6.876309243568058e-07,
|
| 1381 |
-
"loss": 0.6581,
|
| 1382 |
-
"step": 970
|
| 1383 |
-
},
|
| 1384 |
-
{
|
| 1385 |
-
"epoch": 1.7955801104972375,
|
| 1386 |
-
"grad_norm": 0.801312315357016,
|
| 1387 |
-
"learning_rate": 6.302473841930135e-07,
|
| 1388 |
-
"loss": 0.6376,
|
| 1389 |
-
"step": 975
|
| 1390 |
-
},
|
| 1391 |
-
{
|
| 1392 |
-
"epoch": 1.8047882136279927,
|
| 1393 |
-
"grad_norm": 0.8305624262846181,
|
| 1394 |
-
"learning_rate": 5.752858175793951e-07,
|
| 1395 |
-
"loss": 0.6544,
|
| 1396 |
-
"step": 980
|
| 1397 |
-
},
|
| 1398 |
-
{
|
| 1399 |
-
"epoch": 1.8139963167587476,
|
| 1400 |
-
"grad_norm": 0.8131955374085587,
|
| 1401 |
-
"learning_rate": 5.227604314503343e-07,
|
| 1402 |
-
"loss": 0.6723,
|
| 1403 |
-
"step": 985
|
| 1404 |
-
},
|
| 1405 |
-
{
|
| 1406 |
-
"epoch": 1.8232044198895028,
|
| 1407 |
-
"grad_norm": 0.7947035393070572,
|
| 1408 |
-
"learning_rate": 4.726848030155129e-07,
|
| 1409 |
-
"loss": 0.6487,
|
| 1410 |
-
"step": 990
|
| 1411 |
-
},
|
| 1412 |
-
{
|
| 1413 |
-
"epoch": 1.832412523020258,
|
| 1414 |
-
"grad_norm": 0.854919068098428,
|
| 1415 |
-
"learning_rate": 4.250718762503514e-07,
|
| 1416 |
-
"loss": 0.6545,
|
| 1417 |
-
"step": 995
|
| 1418 |
-
},
|
| 1419 |
-
{
|
| 1420 |
-
"epoch": 1.8416206261510129,
|
| 1421 |
-
"grad_norm": 0.8021121270473258,
|
| 1422 |
-
"learning_rate": 3.799339585501494e-07,
|
| 1423 |
-
"loss": 0.6446,
|
| 1424 |
-
"step": 1000
|
| 1425 |
-
},
|
| 1426 |
-
{
|
| 1427 |
-
"epoch": 1.850828729281768,
|
| 1428 |
-
"grad_norm": 0.8523268755119086,
|
| 1429 |
-
"learning_rate": 3.3728271754875365e-07,
|
| 1430 |
-
"loss": 0.6508,
|
| 1431 |
-
"step": 1005
|
| 1432 |
-
},
|
| 1433 |
-
{
|
| 1434 |
-
"epoch": 1.860036832412523,
|
| 1435 |
-
"grad_norm": 0.7951948802052549,
|
| 1436 |
-
"learning_rate": 2.9712917810261645e-07,
|
| 1437 |
-
"loss": 0.6515,
|
| 1438 |
-
"step": 1010
|
| 1439 |
-
},
|
| 1440 |
-
{
|
| 1441 |
-
"epoch": 1.869244935543278,
|
| 1442 |
-
"grad_norm": 0.7861608829681604,
|
| 1443 |
-
"learning_rate": 2.5948371944099715e-07,
|
| 1444 |
-
"loss": 0.6585,
|
| 1445 |
-
"step": 1015
|
| 1446 |
-
},
|
| 1447 |
-
{
|
| 1448 |
-
"epoch": 1.8784530386740332,
|
| 1449 |
-
"grad_norm": 0.8978633111243858,
|
| 1450 |
-
"learning_rate": 2.2435607248304624e-07,
|
| 1451 |
-
"loss": 0.6581,
|
| 1452 |
-
"step": 1020
|
| 1453 |
-
},
|
| 1454 |
-
{
|
| 1455 |
-
"epoch": 1.8876611418047882,
|
| 1456 |
-
"grad_norm": 0.8270088384179148,
|
| 1457 |
-
"learning_rate": 1.91755317322474e-07,
|
| 1458 |
-
"loss": 0.676,
|
| 1459 |
-
"step": 1025
|
| 1460 |
-
},
|
| 1461 |
-
{
|
| 1462 |
-
"epoch": 1.8968692449355433,
|
| 1463 |
-
"grad_norm": 0.7911364129059182,
|
| 1464 |
-
"learning_rate": 1.616898808804568e-07,
|
| 1465 |
-
"loss": 0.6551,
|
| 1466 |
-
"step": 1030
|
| 1467 |
-
},
|
| 1468 |
-
{
|
| 1469 |
-
"epoch": 1.9060773480662982,
|
| 1470 |
-
"grad_norm": 0.7914078659350331,
|
| 1471 |
-
"learning_rate": 1.341675347273652e-07,
|
| 1472 |
-
"loss": 0.6542,
|
| 1473 |
-
"step": 1035
|
| 1474 |
-
},
|
| 1475 |
-
{
|
| 1476 |
-
"epoch": 1.9152854511970534,
|
| 1477 |
-
"grad_norm": 0.803263967646776,
|
| 1478 |
-
"learning_rate": 1.0919539307391314e-07,
|
| 1479 |
-
"loss": 0.6649,
|
| 1480 |
-
"step": 1040
|
| 1481 |
-
},
|
| 1482 |
-
{
|
| 1483 |
-
"epoch": 1.9244935543278086,
|
| 1484 |
-
"grad_norm": 0.9249827462440064,
|
| 1485 |
-
"learning_rate": 8.677991093220761e-08,
|
| 1486 |
-
"loss": 0.654,
|
| 1487 |
-
"step": 1045
|
| 1488 |
-
},
|
| 1489 |
-
{
|
| 1490 |
-
"epoch": 1.9337016574585635,
|
| 1491 |
-
"grad_norm": 0.8226059624242837,
|
| 1492 |
-
"learning_rate": 6.692688244720357e-08,
|
| 1493 |
-
"loss": 0.6414,
|
| 1494 |
-
"step": 1050
|
| 1495 |
-
},
|
| 1496 |
-
{
|
| 1497 |
-
"epoch": 1.9429097605893186,
|
| 1498 |
-
"grad_norm": 0.8086200955718532,
|
| 1499 |
-
"learning_rate": 4.964143939898747e-08,
|
| 1500 |
-
"loss": 0.6503,
|
| 1501 |
-
"step": 1055
|
| 1502 |
-
},
|
| 1503 |
-
{
|
| 1504 |
-
"epoch": 1.9521178637200736,
|
| 1505 |
-
"grad_norm": 0.7620700285094846,
|
| 1506 |
-
"learning_rate": 3.49280498762683e-08,
|
| 1507 |
-
"loss": 0.6538,
|
| 1508 |
-
"step": 1060
|
| 1509 |
-
},
|
| 1510 |
-
{
|
| 1511 |
-
"epoch": 1.9613259668508287,
|
| 1512 |
-
"grad_norm": 0.8452874961667022,
|
| 1513 |
-
"learning_rate": 2.2790517121428212e-08,
|
| 1514 |
-
"loss": 0.6534,
|
| 1515 |
-
"step": 1065
|
| 1516 |
-
},
|
| 1517 |
-
{
|
| 1518 |
-
"epoch": 1.9705340699815839,
|
| 1519 |
-
"grad_norm": 0.8148841155131437,
|
| 1520 |
-
"learning_rate": 1.3231978547427749e-08,
|
| 1521 |
-
"loss": 0.646,
|
| 1522 |
-
"step": 1070
|
| 1523 |
-
},
|
| 1524 |
-
{
|
| 1525 |
-
"epoch": 1.979742173112339,
|
| 1526 |
-
"grad_norm": 0.7814438235127492,
|
| 1527 |
-
"learning_rate": 6.254904926820127e-09,
|
| 1528 |
-
"loss": 0.6488,
|
| 1529 |
-
"step": 1075
|
| 1530 |
-
},
|
| 1531 |
-
{
|
| 1532 |
-
"epoch": 1.988950276243094,
|
| 1533 |
-
"grad_norm": 0.8357299862813449,
|
| 1534 |
-
"learning_rate": 1.8610997530876806e-09,
|
| 1535 |
-
"loss": 0.6691,
|
| 1536 |
-
"step": 1080
|
| 1537 |
-
},
|
| 1538 |
-
{
|
| 1539 |
-
"epoch": 1.9981583793738489,
|
| 1540 |
-
"grad_norm": 0.7847164963349244,
|
| 1541 |
-
"learning_rate": 5.169877445809768e-11,
|
| 1542 |
-
"loss": 0.6439,
|
| 1543 |
-
"step": 1085
|
| 1544 |
-
},
|
| 1545 |
-
{
|
| 1546 |
-
"epoch": 2.0,
|
| 1547 |
-
"eval_loss": 0.749311625957489,
|
| 1548 |
-
"eval_runtime": 7.2936,
|
| 1549 |
-
"eval_samples_per_second": 68.553,
|
| 1550 |
-
"eval_steps_per_second": 2.194,
|
| 1551 |
-
"step": 1086
|
| 1552 |
},
|
| 1553 |
{
|
| 1554 |
-
"epoch":
|
| 1555 |
-
"step":
|
| 1556 |
-
"total_flos":
|
| 1557 |
-
"train_loss": 0.
|
| 1558 |
-
"train_runtime":
|
| 1559 |
-
"train_samples_per_second": 19.
|
| 1560 |
"train_steps_per_second": 0.153
|
| 1561 |
}
|
| 1562 |
],
|
| 1563 |
"logging_steps": 5,
|
| 1564 |
-
"max_steps":
|
| 1565 |
"num_input_tokens_seen": 0,
|
| 1566 |
"num_train_epochs": 2,
|
| 1567 |
"save_steps": 100,
|
| 1568 |
-
"total_flos":
|
| 1569 |
"train_batch_size": 16,
|
| 1570 |
"trial_name": null,
|
| 1571 |
"trial_params": null
|
|
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 1.9977753058954395,
|
| 5 |
"eval_steps": 500,
|
| 6 |
+
"global_step": 898,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
+
"epoch": 0.002224694104560623,
|
| 13 |
+
"grad_norm": 10.055098914546134,
|
| 14 |
+
"learning_rate": 2.2222222222222224e-07,
|
| 15 |
+
"loss": 1.3661,
|
| 16 |
"step": 1
|
| 17 |
},
|
| 18 |
{
|
| 19 |
+
"epoch": 0.011123470522803115,
|
| 20 |
+
"grad_norm": 9.126296513090308,
|
| 21 |
+
"learning_rate": 1.111111111111111e-06,
|
| 22 |
+
"loss": 1.2942,
|
| 23 |
"step": 5
|
| 24 |
},
|
| 25 |
{
|
| 26 |
+
"epoch": 0.02224694104560623,
|
| 27 |
+
"grad_norm": 7.84364716096091,
|
| 28 |
+
"learning_rate": 2.222222222222222e-06,
|
| 29 |
+
"loss": 1.313,
|
| 30 |
"step": 10
|
| 31 |
},
|
| 32 |
{
|
| 33 |
+
"epoch": 0.03337041156840934,
|
| 34 |
+
"grad_norm": 5.7358942086951625,
|
| 35 |
+
"learning_rate": 3.3333333333333333e-06,
|
| 36 |
+
"loss": 1.173,
|
| 37 |
"step": 15
|
| 38 |
},
|
| 39 |
{
|
| 40 |
+
"epoch": 0.04449388209121246,
|
| 41 |
+
"grad_norm": 1.8899122666121964,
|
| 42 |
+
"learning_rate": 4.444444444444444e-06,
|
| 43 |
+
"loss": 0.9741,
|
| 44 |
"step": 20
|
| 45 |
},
|
| 46 |
{
|
| 47 |
+
"epoch": 0.05561735261401557,
|
| 48 |
+
"grad_norm": 1.3213559323931832,
|
| 49 |
+
"learning_rate": 5.555555555555557e-06,
|
| 50 |
+
"loss": 0.9238,
|
| 51 |
"step": 25
|
| 52 |
},
|
| 53 |
{
|
| 54 |
+
"epoch": 0.06674082313681869,
|
| 55 |
+
"grad_norm": 1.041365244653977,
|
| 56 |
+
"learning_rate": 6.666666666666667e-06,
|
| 57 |
+
"loss": 0.8728,
|
| 58 |
"step": 30
|
| 59 |
},
|
| 60 |
{
|
| 61 |
+
"epoch": 0.0778642936596218,
|
| 62 |
+
"grad_norm": 0.9488843220134849,
|
| 63 |
+
"learning_rate": 7.77777777777778e-06,
|
| 64 |
+
"loss": 0.9007,
|
| 65 |
"step": 35
|
| 66 |
},
|
| 67 |
{
|
| 68 |
+
"epoch": 0.08898776418242492,
|
| 69 |
+
"grad_norm": 0.8275801836954026,
|
| 70 |
+
"learning_rate": 8.888888888888888e-06,
|
| 71 |
+
"loss": 0.8661,
|
| 72 |
"step": 40
|
| 73 |
},
|
| 74 |
{
|
| 75 |
+
"epoch": 0.10011123470522804,
|
| 76 |
+
"grad_norm": 0.9030018185879571,
|
| 77 |
+
"learning_rate": 1e-05,
|
| 78 |
+
"loss": 0.8484,
|
| 79 |
"step": 45
|
| 80 |
},
|
| 81 |
{
|
| 82 |
+
"epoch": 0.11123470522803114,
|
| 83 |
+
"grad_norm": 0.7848210903418559,
|
| 84 |
+
"learning_rate": 1.1111111111111113e-05,
|
| 85 |
+
"loss": 0.8291,
|
| 86 |
"step": 50
|
| 87 |
},
|
| 88 |
{
|
| 89 |
+
"epoch": 0.12235817575083426,
|
| 90 |
+
"grad_norm": 0.9313858219522357,
|
| 91 |
+
"learning_rate": 1.2222222222222224e-05,
|
| 92 |
+
"loss": 0.8707,
|
| 93 |
"step": 55
|
| 94 |
},
|
| 95 |
{
|
| 96 |
+
"epoch": 0.13348164627363737,
|
| 97 |
+
"grad_norm": 0.8227239131251839,
|
| 98 |
+
"learning_rate": 1.3333333333333333e-05,
|
| 99 |
+
"loss": 0.8055,
|
| 100 |
"step": 60
|
| 101 |
},
|
| 102 |
{
|
| 103 |
+
"epoch": 0.1446051167964405,
|
| 104 |
+
"grad_norm": 0.8199201964167689,
|
| 105 |
+
"learning_rate": 1.4444444444444446e-05,
|
| 106 |
+
"loss": 0.8151,
|
| 107 |
"step": 65
|
| 108 |
},
|
| 109 |
{
|
| 110 |
+
"epoch": 0.1557285873192436,
|
| 111 |
+
"grad_norm": 0.7767041809184307,
|
| 112 |
+
"learning_rate": 1.555555555555556e-05,
|
| 113 |
+
"loss": 0.8384,
|
| 114 |
"step": 70
|
| 115 |
},
|
| 116 |
{
|
| 117 |
+
"epoch": 0.1668520578420467,
|
| 118 |
+
"grad_norm": 0.9088502789460952,
|
| 119 |
+
"learning_rate": 1.6666666666666667e-05,
|
| 120 |
+
"loss": 0.8106,
|
| 121 |
"step": 75
|
| 122 |
},
|
| 123 |
{
|
| 124 |
+
"epoch": 0.17797552836484984,
|
| 125 |
+
"grad_norm": 0.8343739709762875,
|
| 126 |
+
"learning_rate": 1.7777777777777777e-05,
|
| 127 |
+
"loss": 0.7932,
|
| 128 |
"step": 80
|
| 129 |
},
|
| 130 |
{
|
| 131 |
+
"epoch": 0.18909899888765294,
|
| 132 |
+
"grad_norm": 0.8939354128507568,
|
| 133 |
+
"learning_rate": 1.888888888888889e-05,
|
| 134 |
+
"loss": 0.8045,
|
| 135 |
"step": 85
|
| 136 |
},
|
| 137 |
{
|
| 138 |
+
"epoch": 0.20022246941045607,
|
| 139 |
+
"grad_norm": 0.885411772323368,
|
| 140 |
+
"learning_rate": 2e-05,
|
| 141 |
+
"loss": 0.8009,
|
| 142 |
"step": 90
|
| 143 |
},
|
| 144 |
{
|
| 145 |
+
"epoch": 0.21134593993325917,
|
| 146 |
+
"grad_norm": 0.8890660179859062,
|
| 147 |
+
"learning_rate": 1.9998110384864614e-05,
|
| 148 |
+
"loss": 0.8225,
|
| 149 |
"step": 95
|
| 150 |
},
|
| 151 |
{
|
| 152 |
+
"epoch": 0.22246941045606228,
|
| 153 |
+
"grad_norm": 0.8371399480864414,
|
| 154 |
+
"learning_rate": 1.9992442253587533e-05,
|
| 155 |
+
"loss": 0.7893,
|
| 156 |
"step": 100
|
| 157 |
},
|
| 158 |
{
|
| 159 |
+
"epoch": 0.2335928809788654,
|
| 160 |
+
"grad_norm": 0.8656120072046296,
|
| 161 |
+
"learning_rate": 1.998299774828608e-05,
|
| 162 |
+
"loss": 0.812,
|
| 163 |
"step": 105
|
| 164 |
},
|
| 165 |
{
|
| 166 |
+
"epoch": 0.2447163515016685,
|
| 167 |
+
"grad_norm": 0.8486947049775396,
|
| 168 |
+
"learning_rate": 1.9969780438256295e-05,
|
| 169 |
+
"loss": 0.7886,
|
| 170 |
"step": 110
|
| 171 |
},
|
| 172 |
{
|
| 173 |
+
"epoch": 0.25583982202447164,
|
| 174 |
+
"grad_norm": 0.8507931458580409,
|
| 175 |
+
"learning_rate": 1.995279531862399e-05,
|
| 176 |
+
"loss": 0.8078,
|
| 177 |
"step": 115
|
| 178 |
},
|
| 179 |
{
|
| 180 |
+
"epoch": 0.26696329254727474,
|
| 181 |
+
"grad_norm": 0.8150830104906274,
|
| 182 |
+
"learning_rate": 1.993204880845699e-05,
|
| 183 |
+
"loss": 0.7672,
|
| 184 |
"step": 120
|
| 185 |
},
|
| 186 |
{
|
| 187 |
+
"epoch": 0.27808676307007785,
|
| 188 |
+
"grad_norm": 0.8642627038468238,
|
| 189 |
+
"learning_rate": 1.9907548748339223e-05,
|
| 190 |
+
"loss": 0.7929,
|
| 191 |
"step": 125
|
| 192 |
},
|
| 193 |
{
|
| 194 |
+
"epoch": 0.289210233592881,
|
| 195 |
+
"grad_norm": 0.8541146407873894,
|
| 196 |
+
"learning_rate": 1.987930439740757e-05,
|
| 197 |
+
"loss": 0.7873,
|
| 198 |
"step": 130
|
| 199 |
},
|
| 200 |
{
|
| 201 |
+
"epoch": 0.3003337041156841,
|
| 202 |
+
"grad_norm": 0.8995167237276087,
|
| 203 |
+
"learning_rate": 1.9847326429852632e-05,
|
| 204 |
+
"loss": 0.7862,
|
| 205 |
"step": 135
|
| 206 |
},
|
| 207 |
{
|
| 208 |
+
"epoch": 0.3114571746384872,
|
| 209 |
+
"grad_norm": 0.8173771695441595,
|
| 210 |
+
"learning_rate": 1.981162693088471e-05,
|
| 211 |
+
"loss": 0.7983,
|
| 212 |
"step": 140
|
| 213 |
},
|
| 214 |
{
|
| 215 |
+
"epoch": 0.3225806451612903,
|
| 216 |
+
"grad_norm": 0.8404409912157184,
|
| 217 |
+
"learning_rate": 1.977221939216652e-05,
|
| 218 |
+
"loss": 0.8037,
|
| 219 |
"step": 145
|
| 220 |
},
|
| 221 |
{
|
| 222 |
+
"epoch": 0.3337041156840934,
|
| 223 |
+
"grad_norm": 0.8245961889334814,
|
| 224 |
+
"learning_rate": 1.9729118706714377e-05,
|
| 225 |
+
"loss": 0.8027,
|
| 226 |
"step": 150
|
| 227 |
},
|
| 228 |
{
|
| 229 |
+
"epoch": 0.3448275862068966,
|
| 230 |
+
"grad_norm": 0.8314933715558246,
|
| 231 |
+
"learning_rate": 1.96823411632698e-05,
|
| 232 |
+
"loss": 0.7843,
|
| 233 |
"step": 155
|
| 234 |
},
|
| 235 |
{
|
| 236 |
+
"epoch": 0.3559510567296997,
|
| 237 |
+
"grad_norm": 0.8726345334934287,
|
| 238 |
+
"learning_rate": 1.9631904440143614e-05,
|
| 239 |
+
"loss": 0.793,
|
| 240 |
"step": 160
|
| 241 |
},
|
| 242 |
{
|
| 243 |
+
"epoch": 0.3670745272525028,
|
| 244 |
+
"grad_norm": 0.8415196157334568,
|
| 245 |
+
"learning_rate": 1.9577827598534888e-05,
|
| 246 |
+
"loss": 0.7668,
|
| 247 |
"step": 165
|
| 248 |
},
|
| 249 |
{
|
| 250 |
+
"epoch": 0.3781979977753059,
|
| 251 |
+
"grad_norm": 0.8815634680126696,
|
| 252 |
+
"learning_rate": 1.95201310753273e-05,
|
| 253 |
+
"loss": 0.7851,
|
| 254 |
"step": 170
|
| 255 |
},
|
| 256 |
{
|
| 257 |
+
"epoch": 0.389321468298109,
|
| 258 |
+
"grad_norm": 0.8024209332569573,
|
| 259 |
+
"learning_rate": 1.945883667536556e-05,
|
| 260 |
+
"loss": 0.7772,
|
| 261 |
"step": 175
|
| 262 |
},
|
| 263 |
{
|
| 264 |
+
"epoch": 0.40044493882091214,
|
| 265 |
+
"grad_norm": 0.8490296704540178,
|
| 266 |
+
"learning_rate": 1.9393967563214833e-05,
|
| 267 |
+
"loss": 0.7761,
|
| 268 |
"step": 180
|
| 269 |
},
|
| 270 |
{
|
| 271 |
+
"epoch": 0.41156840934371525,
|
| 272 |
+
"grad_norm": 0.8454078852442415,
|
| 273 |
+
"learning_rate": 1.9325548254406354e-05,
|
| 274 |
+
"loss": 0.7624,
|
| 275 |
"step": 185
|
| 276 |
},
|
| 277 |
{
|
| 278 |
+
"epoch": 0.42269187986651835,
|
| 279 |
+
"grad_norm": 0.8511908385629164,
|
| 280 |
+
"learning_rate": 1.925360460617242e-05,
|
| 281 |
+
"loss": 0.7668,
|
| 282 |
"step": 190
|
| 283 |
},
|
| 284 |
{
|
| 285 |
+
"epoch": 0.43381535038932145,
|
| 286 |
+
"grad_norm": 0.7982929700309319,
|
| 287 |
+
"learning_rate": 1.9178163807674343e-05,
|
| 288 |
+
"loss": 0.7634,
|
| 289 |
"step": 195
|
| 290 |
},
|
| 291 |
{
|
| 292 |
+
"epoch": 0.44493882091212456,
|
| 293 |
+
"grad_norm": 0.7816495059977551,
|
| 294 |
+
"learning_rate": 1.9099254369727062e-05,
|
| 295 |
+
"loss": 0.7748,
|
| 296 |
"step": 200
|
| 297 |
},
|
| 298 |
{
|
| 299 |
+
"epoch": 0.4560622914349277,
|
| 300 |
+
"grad_norm": 0.7563436109378239,
|
| 301 |
+
"learning_rate": 1.901690611402423e-05,
|
| 302 |
+
"loss": 0.7722,
|
| 303 |
"step": 205
|
| 304 |
},
|
| 305 |
{
|
| 306 |
+
"epoch": 0.4671857619577308,
|
| 307 |
+
"grad_norm": 0.8466958907764663,
|
| 308 |
+
"learning_rate": 1.8931150161867917e-05,
|
| 309 |
+
"loss": 0.7765,
|
| 310 |
"step": 210
|
| 311 |
},
|
| 312 |
{
|
| 313 |
+
"epoch": 0.4783092324805339,
|
| 314 |
+
"grad_norm": 0.7620831089947199,
|
| 315 |
+
"learning_rate": 1.8842018922407153e-05,
|
| 316 |
+
"loss": 0.7704,
|
| 317 |
"step": 215
|
| 318 |
},
|
| 319 |
{
|
| 320 |
+
"epoch": 0.489432703003337,
|
| 321 |
+
"grad_norm": 0.84897869766796,
|
| 322 |
+
"learning_rate": 1.874954608038976e-05,
|
| 323 |
+
"loss": 0.7729,
|
| 324 |
"step": 220
|
| 325 |
},
|
| 326 |
{
|
| 327 |
+
"epoch": 0.5005561735261401,
|
| 328 |
+
"grad_norm": 0.8210216367491153,
|
| 329 |
+
"learning_rate": 1.8653766583432114e-05,
|
| 330 |
+
"loss": 0.7716,
|
| 331 |
"step": 225
|
| 332 |
},
|
| 333 |
{
|
| 334 |
+
"epoch": 0.5116796440489433,
|
| 335 |
+
"grad_norm": 0.8730864132572128,
|
| 336 |
+
"learning_rate": 1.855471662881164e-05,
|
| 337 |
+
"loss": 0.7882,
|
| 338 |
"step": 230
|
| 339 |
},
|
| 340 |
{
|
| 341 |
+
"epoch": 0.5228031145717463,
|
| 342 |
+
"grad_norm": 0.7987767766986732,
|
| 343 |
+
"learning_rate": 1.845243364978702e-05,
|
| 344 |
+
"loss": 0.7609,
|
| 345 |
"step": 235
|
| 346 |
},
|
| 347 |
{
|
| 348 |
+
"epoch": 0.5339265850945495,
|
| 349 |
+
"grad_norm": 0.7622467591417209,
|
| 350 |
+
"learning_rate": 1.8346956301451303e-05,
|
| 351 |
+
"loss": 0.7551,
|
| 352 |
"step": 240
|
| 353 |
},
|
| 354 |
{
|
| 355 |
+
"epoch": 0.5450500556173526,
|
| 356 |
+
"grad_norm": 0.831755089876671,
|
| 357 |
+
"learning_rate": 1.8238324446123265e-05,
|
| 358 |
+
"loss": 0.7634,
|
| 359 |
"step": 245
|
| 360 |
},
|
| 361 |
{
|
| 362 |
+
"epoch": 0.5561735261401557,
|
| 363 |
+
"grad_norm": 0.7942015525420777,
|
| 364 |
+
"learning_rate": 1.8126579138282502e-05,
|
| 365 |
+
"loss": 0.7541,
|
| 366 |
"step": 250
|
| 367 |
},
|
| 368 |
{
|
| 369 |
+
"epoch": 0.5672969966629589,
|
| 370 |
+
"grad_norm": 0.824338690380162,
|
| 371 |
+
"learning_rate": 1.801176260905402e-05,
|
| 372 |
+
"loss": 0.7641,
|
| 373 |
"step": 255
|
| 374 |
},
|
| 375 |
{
|
| 376 |
+
"epoch": 0.578420467185762,
|
| 377 |
+
"grad_norm": 0.8160737587426502,
|
| 378 |
+
"learning_rate": 1.7893918250248106e-05,
|
| 379 |
+
"loss": 0.747,
|
| 380 |
"step": 260
|
| 381 |
},
|
| 382 |
{
|
| 383 |
+
"epoch": 0.5895439377085651,
|
| 384 |
+
"grad_norm": 0.7619822007899363,
|
| 385 |
+
"learning_rate": 1.7773090597961554e-05,
|
| 386 |
+
"loss": 0.7353,
|
| 387 |
"step": 265
|
| 388 |
},
|
| 389 |
{
|
| 390 |
+
"epoch": 0.6006674082313682,
|
| 391 |
+
"grad_norm": 0.7898915967361627,
|
| 392 |
+
"learning_rate": 1.764932531574648e-05,
|
| 393 |
+
"loss": 0.7588,
|
| 394 |
"step": 270
|
| 395 |
},
|
| 396 |
{
|
| 397 |
+
"epoch": 0.6117908787541713,
|
| 398 |
+
"grad_norm": 0.9091978830168115,
|
| 399 |
+
"learning_rate": 1.7522669177352978e-05,
|
| 400 |
+
"loss": 0.781,
|
| 401 |
"step": 275
|
| 402 |
},
|
| 403 |
{
|
| 404 |
+
"epoch": 0.6229143492769744,
|
| 405 |
+
"grad_norm": 0.8198662250585645,
|
| 406 |
+
"learning_rate": 1.7393170049052274e-05,
|
| 407 |
+
"loss": 0.7545,
|
| 408 |
"step": 280
|
| 409 |
},
|
| 410 |
{
|
| 411 |
+
"epoch": 0.6340378197997776,
|
| 412 |
+
"grad_norm": 0.7880789917007047,
|
| 413 |
+
"learning_rate": 1.7260876871546935e-05,
|
| 414 |
+
"loss": 0.7726,
|
| 415 |
"step": 285
|
| 416 |
},
|
| 417 |
{
|
| 418 |
+
"epoch": 0.6451612903225806,
|
| 419 |
+
"grad_norm": 0.8385501161327127,
|
| 420 |
+
"learning_rate": 1.7125839641475074e-05,
|
| 421 |
+
"loss": 0.7619,
|
| 422 |
"step": 290
|
| 423 |
},
|
| 424 |
{
|
| 425 |
+
"epoch": 0.6562847608453838,
|
| 426 |
+
"grad_norm": 0.8924470377096518,
|
| 427 |
+
"learning_rate": 1.6988109392515432e-05,
|
| 428 |
+
"loss": 0.7346,
|
| 429 |
"step": 295
|
| 430 |
},
|
| 431 |
{
|
| 432 |
+
"epoch": 0.6674082313681868,
|
| 433 |
+
"grad_norm": 0.7890602183226353,
|
| 434 |
+
"learning_rate": 1.6847738176100632e-05,
|
| 435 |
+
"loss": 0.7643,
|
| 436 |
"step": 300
|
| 437 |
},
|
| 438 |
{
|
| 439 |
+
"epoch": 0.67853170189099,
|
| 440 |
+
"grad_norm": 0.8110214434516344,
|
| 441 |
+
"learning_rate": 1.6704779041745686e-05,
|
| 442 |
+
"loss": 0.7603,
|
| 443 |
"step": 305
|
| 444 |
},
|
| 445 |
{
|
| 446 |
+
"epoch": 0.6896551724137931,
|
| 447 |
+
"grad_norm": 0.7873486916181355,
|
| 448 |
+
"learning_rate": 1.65592860169994e-05,
|
| 449 |
+
"loss": 0.7595,
|
| 450 |
"step": 310
|
| 451 |
},
|
| 452 |
{
|
| 453 |
+
"epoch": 0.7007786429365962,
|
| 454 |
+
"grad_norm": 0.7527591581883117,
|
| 455 |
+
"learning_rate": 1.6411314087026108e-05,
|
| 456 |
+
"loss": 0.7508,
|
| 457 |
"step": 315
|
| 458 |
},
|
| 459 |
{
|
| 460 |
+
"epoch": 0.7119021134593994,
|
| 461 |
+
"grad_norm": 0.8283445625547928,
|
| 462 |
+
"learning_rate": 1.6260919173825507e-05,
|
| 463 |
+
"loss": 0.7387,
|
| 464 |
"step": 320
|
| 465 |
},
|
| 466 |
{
|
| 467 |
+
"epoch": 0.7230255839822024,
|
| 468 |
+
"grad_norm": 0.7262591008119376,
|
| 469 |
+
"learning_rate": 1.6108158115098443e-05,
|
| 470 |
+
"loss": 0.7264,
|
| 471 |
"step": 325
|
| 472 |
},
|
| 473 |
{
|
| 474 |
+
"epoch": 0.7341490545050056,
|
| 475 |
+
"grad_norm": 0.7739070575646189,
|
| 476 |
+
"learning_rate": 1.595308864276666e-05,
|
| 477 |
+
"loss": 0.7435,
|
| 478 |
"step": 330
|
| 479 |
},
|
| 480 |
{
|
| 481 |
+
"epoch": 0.7452725250278087,
|
| 482 |
+
"grad_norm": 0.8489625309544235,
|
| 483 |
+
"learning_rate": 1.5795769361154548e-05,
|
| 484 |
+
"loss": 0.7615,
|
| 485 |
"step": 335
|
| 486 |
},
|
| 487 |
{
|
| 488 |
+
"epoch": 0.7563959955506118,
|
| 489 |
+
"grad_norm": 0.769293674851008,
|
| 490 |
+
"learning_rate": 1.5636259724841224e-05,
|
| 491 |
+
"loss": 0.7536,
|
| 492 |
"step": 340
|
| 493 |
},
|
| 494 |
{
|
| 495 |
+
"epoch": 0.7675194660734149,
|
| 496 |
+
"grad_norm": 0.7920492833518509,
|
| 497 |
+
"learning_rate": 1.5474620016191296e-05,
|
| 498 |
+
"loss": 0.7431,
|
| 499 |
"step": 345
|
| 500 |
},
|
| 501 |
{
|
| 502 |
+
"epoch": 0.778642936596218,
|
| 503 |
+
"grad_norm": 0.7468241826638446,
|
| 504 |
+
"learning_rate": 1.531091132257275e-05,
|
| 505 |
+
"loss": 0.732,
|
| 506 |
"step": 350
|
| 507 |
},
|
| 508 |
{
|
| 509 |
+
"epoch": 0.7897664071190211,
|
| 510 |
+
"grad_norm": 0.743401655764991,
|
| 511 |
+
"learning_rate": 1.5145195513270644e-05,
|
| 512 |
+
"loss": 0.7291,
|
| 513 |
"step": 355
|
| 514 |
},
|
| 515 |
{
|
| 516 |
+
"epoch": 0.8008898776418243,
|
| 517 |
+
"grad_norm": 0.8018681967515083,
|
| 518 |
+
"learning_rate": 1.4977535216105258e-05,
|
| 519 |
+
"loss": 0.7257,
|
| 520 |
"step": 360
|
| 521 |
},
|
| 522 |
{
|
| 523 |
+
"epoch": 0.8120133481646273,
|
| 524 |
+
"grad_norm": 0.7600864193920938,
|
| 525 |
+
"learning_rate": 1.480799379376362e-05,
|
| 526 |
+
"loss": 0.741,
|
| 527 |
"step": 365
|
| 528 |
},
|
| 529 |
{
|
| 530 |
+
"epoch": 0.8231368186874305,
|
| 531 |
+
"grad_norm": 0.8389942553789884,
|
| 532 |
+
"learning_rate": 1.4636635319853274e-05,
|
| 533 |
+
"loss": 0.742,
|
| 534 |
"step": 370
|
| 535 |
},
|
| 536 |
{
|
| 537 |
+
"epoch": 0.8342602892102335,
|
| 538 |
+
"grad_norm": 0.7886124496265561,
|
| 539 |
+
"learning_rate": 1.4463524554687398e-05,
|
| 540 |
+
"loss": 0.7545,
|
| 541 |
"step": 375
|
| 542 |
},
|
| 543 |
{
|
| 544 |
+
"epoch": 0.8453837597330367,
|
| 545 |
+
"grad_norm": 0.7344461106509269,
|
| 546 |
+
"learning_rate": 1.4288726920810381e-05,
|
| 547 |
+
"loss": 0.7278,
|
| 548 |
"step": 380
|
| 549 |
},
|
| 550 |
{
|
| 551 |
+
"epoch": 0.8565072302558399,
|
| 552 |
+
"grad_norm": 0.8189552942167496,
|
| 553 |
+
"learning_rate": 1.4112308478273144e-05,
|
| 554 |
+
"loss": 0.7461,
|
| 555 |
"step": 385
|
| 556 |
},
|
| 557 |
{
|
| 558 |
+
"epoch": 0.8676307007786429,
|
| 559 |
+
"grad_norm": 0.7834178214856152,
|
| 560 |
+
"learning_rate": 1.3934335899667526e-05,
|
| 561 |
+
"loss": 0.7378,
|
| 562 |
"step": 390
|
| 563 |
},
|
| 564 |
{
|
| 565 |
+
"epoch": 0.8787541713014461,
|
| 566 |
+
"grad_norm": 0.7714482088214847,
|
| 567 |
+
"learning_rate": 1.3754876444929165e-05,
|
| 568 |
+
"loss": 0.7489,
|
| 569 |
"step": 395
|
| 570 |
},
|
| 571 |
{
|
| 572 |
+
"epoch": 0.8898776418242491,
|
| 573 |
+
"grad_norm": 0.7755466989206458,
|
| 574 |
+
"learning_rate": 1.357399793591844e-05,
|
| 575 |
+
"loss": 0.7469,
|
| 576 |
"step": 400
|
| 577 |
},
|
| 578 |
{
|
| 579 |
+
"epoch": 0.9010011123470523,
|
| 580 |
+
"grad_norm": 0.8377177614112041,
|
| 581 |
+
"learning_rate": 1.3391768730789e-05,
|
| 582 |
+
"loss": 0.739,
|
| 583 |
"step": 405
|
| 584 |
},
|
| 585 |
{
|
| 586 |
+
"epoch": 0.9121245828698554,
|
| 587 |
+
"grad_norm": 0.7908000259612985,
|
| 588 |
+
"learning_rate": 1.3208257698153677e-05,
|
| 589 |
+
"loss": 0.7281,
|
| 590 |
"step": 410
|
| 591 |
},
|
| 592 |
{
|
| 593 |
+
"epoch": 0.9232480533926585,
|
| 594 |
+
"grad_norm": 0.7818338363656034,
|
| 595 |
+
"learning_rate": 1.3023534191057427e-05,
|
| 596 |
+
"loss": 0.753,
|
| 597 |
"step": 415
|
| 598 |
},
|
| 599 |
{
|
| 600 |
+
"epoch": 0.9343715239154616,
|
| 601 |
+
"grad_norm": 0.749284045444348,
|
| 602 |
+
"learning_rate": 1.283766802076722e-05,
|
| 603 |
+
"loss": 0.738,
|
| 604 |
"step": 420
|
| 605 |
},
|
| 606 |
{
|
| 607 |
+
"epoch": 0.9454949944382648,
|
| 608 |
+
"grad_norm": 0.7839632426822802,
|
| 609 |
+
"learning_rate": 1.2650729430388764e-05,
|
| 610 |
+
"loss": 0.7436,
|
| 611 |
"step": 425
|
| 612 |
},
|
| 613 |
{
|
| 614 |
+
"epoch": 0.9566184649610678,
|
| 615 |
+
"grad_norm": 0.7721084207333134,
|
| 616 |
+
"learning_rate": 1.2462789068320016e-05,
|
| 617 |
+
"loss": 0.748,
|
| 618 |
"step": 430
|
| 619 |
},
|
| 620 |
{
|
| 621 |
+
"epoch": 0.967741935483871,
|
| 622 |
+
"grad_norm": 0.7290139666717954,
|
| 623 |
+
"learning_rate": 1.2273917961551513e-05,
|
| 624 |
+
"loss": 0.7239,
|
| 625 |
"step": 435
|
| 626 |
},
|
| 627 |
{
|
| 628 |
+
"epoch": 0.978865406006674,
|
| 629 |
+
"grad_norm": 0.8056619429328024,
|
| 630 |
+
"learning_rate": 1.2084187488823657e-05,
|
| 631 |
+
"loss": 0.738,
|
| 632 |
"step": 440
|
| 633 |
},
|
| 634 |
{
|
| 635 |
+
"epoch": 0.9899888765294772,
|
| 636 |
+
"grad_norm": 0.7770001841474352,
|
| 637 |
+
"learning_rate": 1.1893669353651032e-05,
|
| 638 |
+
"loss": 0.7385,
|
| 639 |
"step": 445
|
| 640 |
},
|
| 641 |
{
|
| 642 |
+
"epoch": 0.9988876529477196,
|
| 643 |
+
"eval_loss": 0.7580433487892151,
|
| 644 |
+
"eval_runtime": 5.7595,
|
| 645 |
+
"eval_samples_per_second": 70.666,
|
| 646 |
+
"eval_steps_per_second": 2.257,
|
| 647 |
+
"step": 449
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 1.0011123470522802,
|
| 651 |
+
"grad_norm": 0.8135422551691482,
|
| 652 |
+
"learning_rate": 1.1702435557223988e-05,
|
| 653 |
+
"loss": 0.7266,
|
| 654 |
"step": 450
|
| 655 |
},
|
| 656 |
{
|
| 657 |
+
"epoch": 1.0122358175750834,
|
| 658 |
+
"grad_norm": 0.8964435701944624,
|
| 659 |
+
"learning_rate": 1.1510558371197754e-05,
|
| 660 |
+
"loss": 0.6412,
|
| 661 |
"step": 455
|
| 662 |
},
|
| 663 |
{
|
| 664 |
+
"epoch": 1.0233592880978866,
|
| 665 |
+
"grad_norm": 0.8600984235769464,
|
| 666 |
+
"learning_rate": 1.1318110310379303e-05,
|
| 667 |
+
"loss": 0.6433,
|
| 668 |
"step": 460
|
| 669 |
},
|
| 670 |
{
|
| 671 |
+
"epoch": 1.0344827586206897,
|
| 672 |
+
"grad_norm": 0.8458880171077358,
|
| 673 |
+
"learning_rate": 1.112516410532233e-05,
|
| 674 |
+
"loss": 0.6292,
|
| 675 |
"step": 465
|
| 676 |
},
|
| 677 |
{
|
| 678 |
+
"epoch": 1.0456062291434929,
|
| 679 |
+
"grad_norm": 0.8284785412200435,
|
| 680 |
+
"learning_rate": 1.0931792674840718e-05,
|
| 681 |
+
"loss": 0.6339,
|
| 682 |
"step": 470
|
| 683 |
},
|
| 684 |
{
|
| 685 |
+
"epoch": 1.0567296996662958,
|
| 686 |
+
"grad_norm": 0.7444359225044646,
|
| 687 |
+
"learning_rate": 1.073806909845082e-05,
|
| 688 |
+
"loss": 0.6355,
|
| 689 |
"step": 475
|
| 690 |
},
|
| 691 |
{
|
| 692 |
+
"epoch": 1.067853170189099,
|
| 693 |
+
"grad_norm": 0.7723149976392786,
|
| 694 |
+
"learning_rate": 1.0544066588753044e-05,
|
| 695 |
+
"loss": 0.6235,
|
| 696 |
"step": 480
|
| 697 |
},
|
| 698 |
{
|
| 699 |
+
"epoch": 1.0789766407119021,
|
| 700 |
+
"grad_norm": 0.7593187602310192,
|
| 701 |
+
"learning_rate": 1.0349858463763114e-05,
|
| 702 |
+
"loss": 0.6105,
|
| 703 |
"step": 485
|
| 704 |
},
|
| 705 |
{
|
| 706 |
+
"epoch": 1.0901001112347053,
|
| 707 |
+
"grad_norm": 0.7855139607799839,
|
| 708 |
+
"learning_rate": 1.0155518119203511e-05,
|
| 709 |
+
"loss": 0.6568,
|
| 710 |
"step": 490
|
| 711 |
},
|
| 712 |
{
|
| 713 |
+
"epoch": 1.1012235817575085,
|
| 714 |
+
"grad_norm": 0.7898700329509037,
|
| 715 |
+
"learning_rate": 9.961119000765532e-06,
|
| 716 |
+
"loss": 0.6225,
|
| 717 |
"step": 495
|
| 718 |
},
|
| 719 |
{
|
| 720 |
+
"epoch": 1.1123470522803114,
|
| 721 |
+
"grad_norm": 0.8669564798886822,
|
| 722 |
+
"learning_rate": 9.766734576352478e-06,
|
| 723 |
+
"loss": 0.6391,
|
| 724 |
"step": 500
|
| 725 |
},
|
| 726 |
{
|
| 727 |
+
"epoch": 1.1234705228031145,
|
| 728 |
+
"grad_norm": 0.7837289412561955,
|
| 729 |
+
"learning_rate": 9.572438308314447e-06,
|
| 730 |
+
"loss": 0.6171,
|
| 731 |
"step": 505
|
| 732 |
},
|
| 733 |
{
|
| 734 |
+
"epoch": 1.1345939933259177,
|
| 735 |
+
"grad_norm": 0.8070851055141667,
|
| 736 |
+
"learning_rate": 9.378303625685196e-06,
|
| 737 |
+
"loss": 0.6282,
|
| 738 |
"step": 510
|
| 739 |
},
|
| 740 |
{
|
| 741 |
+
"epoch": 1.1457174638487209,
|
| 742 |
+
"grad_norm": 0.7979212620110364,
|
| 743 |
+
"learning_rate": 9.184403896431649e-06,
|
| 744 |
+
"loss": 0.6233,
|
| 745 |
"step": 515
|
| 746 |
},
|
| 747 |
{
|
| 748 |
+
"epoch": 1.156840934371524,
|
| 749 |
+
"grad_norm": 0.7704548256097349,
|
| 750 |
+
"learning_rate": 8.990812399726435e-06,
|
| 751 |
+
"loss": 0.5992,
|
| 752 |
"step": 520
|
| 753 |
},
|
| 754 |
{
|
| 755 |
+
"epoch": 1.167964404894327,
|
| 756 |
+
"grad_norm": 0.8961749452380681,
|
| 757 |
+
"learning_rate": 8.797602298254005e-06,
|
| 758 |
+
"loss": 0.6378,
|
| 759 |
"step": 525
|
| 760 |
},
|
| 761 |
{
|
| 762 |
+
"epoch": 1.1790878754171301,
|
| 763 |
+
"grad_norm": 0.8053259116501744,
|
| 764 |
+
"learning_rate": 8.604846610560771e-06,
|
| 765 |
+
"loss": 0.605,
|
| 766 |
"step": 530
|
| 767 |
},
|
| 768 |
{
|
| 769 |
+
"epoch": 1.1902113459399333,
|
| 770 |
+
"grad_norm": 0.7782548725591264,
|
| 771 |
+
"learning_rate": 8.412618183459707e-06,
|
| 772 |
+
"loss": 0.6081,
|
| 773 |
"step": 535
|
| 774 |
},
|
| 775 |
{
|
| 776 |
+
"epoch": 1.2013348164627364,
|
| 777 |
+
"grad_norm": 0.7780589674933976,
|
| 778 |
+
"learning_rate": 8.22098966449988e-06,
|
| 779 |
+
"loss": 0.6251,
|
| 780 |
"step": 540
|
| 781 |
},
|
| 782 |
{
|
| 783 |
+
"epoch": 1.2124582869855396,
|
| 784 |
+
"grad_norm": 0.8213283153654349,
|
| 785 |
+
"learning_rate": 8.030033474511248e-06,
|
| 786 |
+
"loss": 0.6092,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 787 |
"step": 545
|
| 788 |
},
|
| 789 |
{
|
| 790 |
+
"epoch": 1.2235817575083425,
|
| 791 |
+
"grad_norm": 0.7889430474165346,
|
| 792 |
+
"learning_rate": 7.839821780235168e-06,
|
| 793 |
+
"loss": 0.645,
|
| 794 |
"step": 550
|
| 795 |
},
|
| 796 |
{
|
| 797 |
+
"epoch": 1.2347052280311457,
|
| 798 |
+
"grad_norm": 0.8377912406937705,
|
| 799 |
+
"learning_rate": 7.650426467050926e-06,
|
| 800 |
+
"loss": 0.6286,
|
| 801 |
"step": 555
|
| 802 |
},
|
| 803 |
{
|
| 804 |
+
"epoch": 1.2458286985539488,
|
| 805 |
+
"grad_norm": 0.8488324297083317,
|
| 806 |
+
"learning_rate": 7.4619191118085955e-06,
|
| 807 |
+
"loss": 0.6129,
|
| 808 |
"step": 560
|
| 809 |
},
|
| 810 |
{
|
| 811 |
+
"epoch": 1.256952169076752,
|
| 812 |
+
"grad_norm": 0.7608536209939344,
|
| 813 |
+
"learning_rate": 7.274370955778498e-06,
|
| 814 |
+
"loss": 0.6072,
|
| 815 |
"step": 565
|
| 816 |
},
|
| 817 |
{
|
| 818 |
+
"epoch": 1.2680756395995552,
|
| 819 |
+
"grad_norm": 0.7928583676765779,
|
| 820 |
+
"learning_rate": 7.0878528777274814e-06,
|
| 821 |
+
"loss": 0.6042,
|
| 822 |
"step": 570
|
| 823 |
},
|
| 824 |
{
|
| 825 |
+
"epoch": 1.279199110122358,
|
| 826 |
+
"grad_norm": 1.5882806537259504,
|
| 827 |
+
"learning_rate": 6.9024353671322086e-06,
|
| 828 |
+
"loss": 0.647,
|
| 829 |
"step": 575
|
| 830 |
},
|
| 831 |
{
|
| 832 |
+
"epoch": 1.2903225806451613,
|
| 833 |
+
"grad_norm": 0.8447762128943721,
|
| 834 |
+
"learning_rate": 6.718188497539554e-06,
|
| 835 |
+
"loss": 0.6214,
|
| 836 |
"step": 580
|
| 837 |
},
|
| 838 |
{
|
| 839 |
+
"epoch": 1.3014460511679644,
|
| 840 |
+
"grad_norm": 0.8209371297634136,
|
| 841 |
+
"learning_rate": 6.535181900084206e-06,
|
| 842 |
+
"loss": 0.6079,
|
| 843 |
"step": 585
|
| 844 |
},
|
| 845 |
{
|
| 846 |
+
"epoch": 1.3125695216907676,
|
| 847 |
+
"grad_norm": 0.8624820584895021,
|
| 848 |
+
"learning_rate": 6.35348473717345e-06,
|
| 849 |
+
"loss": 0.6221,
|
| 850 |
"step": 590
|
| 851 |
},
|
| 852 |
{
|
| 853 |
+
"epoch": 1.3236929922135707,
|
| 854 |
+
"grad_norm": 0.8056054069589547,
|
| 855 |
+
"learning_rate": 6.173165676349103e-06,
|
| 856 |
+
"loss": 0.6254,
|
| 857 |
"step": 595
|
| 858 |
},
|
| 859 |
{
|
| 860 |
+
"epoch": 1.3348164627363737,
|
| 861 |
+
"grad_norm": 0.8273703875367165,
|
| 862 |
+
"learning_rate": 5.994292864336473e-06,
|
| 863 |
+
"loss": 0.6119,
|
| 864 |
"step": 600
|
| 865 |
},
|
| 866 |
{
|
| 867 |
+
"epoch": 1.3459399332591768,
|
| 868 |
+
"grad_norm": 0.8723777846098392,
|
| 869 |
+
"learning_rate": 5.816933901290136e-06,
|
| 870 |
+
"loss": 0.6395,
|
| 871 |
"step": 605
|
| 872 |
},
|
| 873 |
{
|
| 874 |
+
"epoch": 1.35706340378198,
|
| 875 |
+
"grad_norm": 0.8674540470170442,
|
| 876 |
+
"learning_rate": 5.64115581524629e-06,
|
| 877 |
+
"loss": 0.6163,
|
| 878 |
"step": 610
|
| 879 |
},
|
| 880 |
{
|
| 881 |
+
"epoch": 1.3681868743047831,
|
| 882 |
+
"grad_norm": 0.8762720501618131,
|
| 883 |
+
"learning_rate": 5.4670250367913025e-06,
|
| 884 |
+
"loss": 0.6225,
|
| 885 |
"step": 615
|
| 886 |
},
|
| 887 |
{
|
| 888 |
+
"epoch": 1.3793103448275863,
|
| 889 |
+
"grad_norm": 0.8491305197357123,
|
| 890 |
+
"learning_rate": 5.294607373956071e-06,
|
| 891 |
+
"loss": 0.6093,
|
| 892 |
"step": 620
|
| 893 |
},
|
| 894 |
{
|
| 895 |
+
"epoch": 1.3904338153503892,
|
| 896 |
+
"grad_norm": 0.8166208130830984,
|
| 897 |
+
"learning_rate": 5.1239679873456636e-06,
|
| 898 |
+
"loss": 0.6361,
|
| 899 |
"step": 625
|
| 900 |
},
|
| 901 |
{
|
| 902 |
+
"epoch": 1.4015572858731924,
|
| 903 |
+
"grad_norm": 0.8009131846316857,
|
| 904 |
+
"learning_rate": 4.955171365513603e-06,
|
| 905 |
+
"loss": 0.617,
|
| 906 |
"step": 630
|
| 907 |
},
|
| 908 |
{
|
| 909 |
+
"epoch": 1.4126807563959956,
|
| 910 |
+
"grad_norm": 0.8366294436519559,
|
| 911 |
+
"learning_rate": 4.788281300590169e-06,
|
| 912 |
+
"loss": 0.6118,
|
| 913 |
"step": 635
|
| 914 |
},
|
| 915 |
{
|
| 916 |
+
"epoch": 1.4238042269187987,
|
| 917 |
+
"grad_norm": 0.8808417873595291,
|
| 918 |
+
"learning_rate": 4.623360864173893e-06,
|
| 919 |
+
"loss": 0.6177,
|
| 920 |
"step": 640
|
| 921 |
},
|
| 922 |
{
|
| 923 |
+
"epoch": 1.4349276974416019,
|
| 924 |
+
"grad_norm": 0.8629287010139255,
|
| 925 |
+
"learning_rate": 4.4604723834953315e-06,
|
| 926 |
+
"loss": 0.6251,
|
| 927 |
"step": 645
|
| 928 |
},
|
| 929 |
{
|
| 930 |
+
"epoch": 1.4460511679644048,
|
| 931 |
+
"grad_norm": 0.7875671775994082,
|
| 932 |
+
"learning_rate": 4.299677417862174e-06,
|
| 933 |
+
"loss": 0.6199,
|
| 934 |
"step": 650
|
| 935 |
},
|
| 936 |
{
|
| 937 |
+
"epoch": 1.457174638487208,
|
| 938 |
+
"grad_norm": 0.7828729245421459,
|
| 939 |
+
"learning_rate": 4.141036735394575e-06,
|
| 940 |
+
"loss": 0.6215,
|
| 941 |
"step": 655
|
| 942 |
},
|
| 943 |
{
|
| 944 |
+
"epoch": 1.4682981090100111,
|
| 945 |
+
"grad_norm": 0.8007957527163657,
|
| 946 |
+
"learning_rate": 3.984610290059467e-06,
|
| 947 |
+
"loss": 0.6253,
|
| 948 |
"step": 660
|
| 949 |
},
|
| 950 |
{
|
| 951 |
+
"epoch": 1.4794215795328143,
|
| 952 |
+
"grad_norm": 0.7910153849628225,
|
| 953 |
+
"learning_rate": 3.830457199012585e-06,
|
| 954 |
+
"loss": 0.6157,
|
| 955 |
"step": 665
|
| 956 |
},
|
| 957 |
{
|
| 958 |
+
"epoch": 1.4905450500556174,
|
| 959 |
+
"grad_norm": 0.8255633946537543,
|
| 960 |
+
"learning_rate": 3.6786357202567367e-06,
|
| 961 |
+
"loss": 0.6182,
|
| 962 |
"step": 670
|
| 963 |
},
|
| 964 |
{
|
| 965 |
+
"epoch": 1.5016685205784204,
|
| 966 |
+
"grad_norm": 0.8740651322002517,
|
| 967 |
+
"learning_rate": 3.529203230624747e-06,
|
| 968 |
+
"loss": 0.6334,
|
| 969 |
"step": 675
|
| 970 |
},
|
| 971 |
{
|
| 972 |
+
"epoch": 1.5127919911012235,
|
| 973 |
+
"grad_norm": 0.7540939777667681,
|
| 974 |
+
"learning_rate": 3.3822162040954355e-06,
|
| 975 |
+
"loss": 0.596,
|
| 976 |
"step": 680
|
| 977 |
},
|
| 978 |
{
|
| 979 |
+
"epoch": 1.5239154616240267,
|
| 980 |
+
"grad_norm": 0.7934734563966453,
|
| 981 |
+
"learning_rate": 3.2377301904508163e-06,
|
| 982 |
+
"loss": 0.5951,
|
| 983 |
"step": 685
|
| 984 |
},
|
| 985 |
{
|
| 986 |
+
"epoch": 1.5350389321468298,
|
| 987 |
+
"grad_norm": 0.8131581917254441,
|
| 988 |
+
"learning_rate": 3.0957997942825337e-06,
|
| 989 |
+
"loss": 0.612,
|
| 990 |
"step": 690
|
| 991 |
},
|
| 992 |
{
|
| 993 |
+
"epoch": 1.546162402669633,
|
| 994 |
+
"grad_norm": 0.8454586171154052,
|
| 995 |
+
"learning_rate": 2.956478654355539e-06,
|
| 996 |
+
"loss": 0.6293,
|
| 997 |
"step": 695
|
| 998 |
},
|
| 999 |
{
|
| 1000 |
+
"epoch": 1.557285873192436,
|
| 1001 |
+
"grad_norm": 0.8339559981613245,
|
| 1002 |
+
"learning_rate": 2.8198194233367747e-06,
|
| 1003 |
+
"loss": 0.6088,
|
| 1004 |
"step": 700
|
| 1005 |
},
|
| 1006 |
{
|
| 1007 |
+
"epoch": 1.568409343715239,
|
| 1008 |
+
"grad_norm": 0.821374605567537,
|
| 1009 |
+
"learning_rate": 2.6858737478965036e-06,
|
| 1010 |
+
"loss": 0.6233,
|
| 1011 |
"step": 705
|
| 1012 |
},
|
| 1013 |
{
|
| 1014 |
+
"epoch": 1.5795328142380423,
|
| 1015 |
+
"grad_norm": 0.7711457173871649,
|
| 1016 |
+
"learning_rate": 2.5546922491898497e-06,
|
| 1017 |
+
"loss": 0.6262,
|
| 1018 |
"step": 710
|
| 1019 |
},
|
| 1020 |
{
|
| 1021 |
+
"epoch": 1.5906562847608454,
|
| 1022 |
+
"grad_norm": 0.841438327290974,
|
| 1023 |
+
"learning_rate": 2.4263245037258996e-06,
|
| 1024 |
+
"loss": 0.6359,
|
| 1025 |
"step": 715
|
| 1026 |
},
|
| 1027 |
{
|
| 1028 |
+
"epoch": 1.6017797552836486,
|
| 1029 |
+
"grad_norm": 0.8546385634639357,
|
| 1030 |
+
"learning_rate": 2.3008190246316033e-06,
|
| 1031 |
+
"loss": 0.6312,
|
| 1032 |
"step": 720
|
| 1033 |
},
|
| 1034 |
{
|
| 1035 |
+
"epoch": 1.6129032258064515,
|
| 1036 |
+
"grad_norm": 0.7477788666400325,
|
| 1037 |
+
"learning_rate": 2.178223243317532e-06,
|
| 1038 |
+
"loss": 0.6115,
|
| 1039 |
"step": 725
|
| 1040 |
},
|
| 1041 |
{
|
| 1042 |
+
"epoch": 1.624026696329255,
|
| 1043 |
+
"grad_norm": 0.8448273050299596,
|
| 1044 |
+
"learning_rate": 2.058583491552465e-06,
|
| 1045 |
+
"loss": 0.641,
|
| 1046 |
"step": 730
|
| 1047 |
},
|
| 1048 |
{
|
| 1049 |
+
"epoch": 1.6351501668520578,
|
| 1050 |
+
"grad_norm": 0.8261008969765296,
|
| 1051 |
+
"learning_rate": 1.9419449839535522e-06,
|
| 1052 |
+
"loss": 0.617,
|
| 1053 |
"step": 735
|
| 1054 |
},
|
| 1055 |
{
|
| 1056 |
+
"epoch": 1.646273637374861,
|
| 1057 |
+
"grad_norm": 0.8357122860638048,
|
| 1058 |
+
"learning_rate": 1.8283518008986566e-06,
|
| 1059 |
+
"loss": 0.607,
|
| 1060 |
"step": 740
|
| 1061 |
},
|
| 1062 |
{
|
| 1063 |
+
"epoch": 1.6573971078976641,
|
| 1064 |
+
"grad_norm": 0.8350425917666864,
|
| 1065 |
+
"learning_rate": 1.7178468718673712e-06,
|
| 1066 |
+
"loss": 0.607,
|
| 1067 |
"step": 745
|
| 1068 |
},
|
| 1069 |
{
|
| 1070 |
+
"epoch": 1.668520578420467,
|
| 1071 |
+
"grad_norm": 0.8305517949246249,
|
| 1072 |
+
"learning_rate": 1.6104719592169905e-06,
|
| 1073 |
+
"loss": 0.6151,
|
| 1074 |
"step": 750
|
| 1075 |
},
|
| 1076 |
{
|
| 1077 |
+
"epoch": 1.6796440489432705,
|
| 1078 |
+
"grad_norm": 0.8186870980107259,
|
| 1079 |
+
"learning_rate": 1.506267642399525e-06,
|
| 1080 |
+
"loss": 0.6385,
|
| 1081 |
"step": 755
|
| 1082 |
},
|
| 1083 |
{
|
| 1084 |
+
"epoch": 1.6907675194660734,
|
| 1085 |
+
"grad_norm": 0.7636300249499965,
|
| 1086 |
+
"learning_rate": 1.405273302625828e-06,
|
| 1087 |
+
"loss": 0.6075,
|
| 1088 |
"step": 760
|
| 1089 |
},
|
| 1090 |
{
|
| 1091 |
+
"epoch": 1.7018909899888766,
|
| 1092 |
+
"grad_norm": 0.8363995841599257,
|
| 1093 |
+
"learning_rate": 1.3075271079825035e-06,
|
| 1094 |
+
"loss": 0.638,
|
| 1095 |
"step": 765
|
| 1096 |
},
|
| 1097 |
{
|
| 1098 |
+
"epoch": 1.7130144605116797,
|
| 1099 |
+
"grad_norm": 0.7622345846029582,
|
| 1100 |
+
"learning_rate": 1.2130659990073146e-06,
|
| 1101 |
+
"loss": 0.6125,
|
| 1102 |
"step": 770
|
| 1103 |
},
|
| 1104 |
{
|
| 1105 |
+
"epoch": 1.7241379310344827,
|
| 1106 |
+
"grad_norm": 0.7854431717950116,
|
| 1107 |
+
"learning_rate": 1.1219256747285046e-06,
|
| 1108 |
+
"loss": 0.6203,
|
| 1109 |
"step": 775
|
| 1110 |
},
|
| 1111 |
{
|
| 1112 |
+
"epoch": 1.735261401557286,
|
| 1113 |
+
"grad_norm": 0.80577828518747,
|
| 1114 |
+
"learning_rate": 1.0341405791733183e-06,
|
| 1115 |
+
"loss": 0.6318,
|
| 1116 |
"step": 780
|
| 1117 |
},
|
| 1118 |
{
|
| 1119 |
+
"epoch": 1.746384872080089,
|
| 1120 |
+
"grad_norm": 0.8025241111868774,
|
| 1121 |
+
"learning_rate": 9.497438883507981e-07,
|
| 1122 |
+
"loss": 0.6104,
|
| 1123 |
"step": 785
|
| 1124 |
},
|
| 1125 |
{
|
| 1126 |
+
"epoch": 1.7575083426028921,
|
| 1127 |
+
"grad_norm": 0.7484880734203889,
|
| 1128 |
+
"learning_rate": 8.687674977138116e-07,
|
| 1129 |
+
"loss": 0.6111,
|
| 1130 |
"step": 790
|
| 1131 |
},
|
| 1132 |
{
|
| 1133 |
+
"epoch": 1.7686318131256953,
|
| 1134 |
+
"grad_norm": 0.7844040069263457,
|
| 1135 |
+
"learning_rate": 7.912420101050366e-07,
|
| 1136 |
+
"loss": 0.6058,
|
| 1137 |
"step": 795
|
| 1138 |
},
|
| 1139 |
{
|
| 1140 |
+
"epoch": 1.7797552836484982,
|
| 1141 |
+
"grad_norm": 0.832757853518197,
|
| 1142 |
+
"learning_rate": 7.171967241914224e-07,
|
| 1143 |
+
"loss": 0.6168,
|
| 1144 |
"step": 800
|
| 1145 |
},
|
| 1146 |
{
|
| 1147 |
+
"epoch": 1.7908787541713016,
|
| 1148 |
+
"grad_norm": 0.7982807830933683,
|
| 1149 |
+
"learning_rate": 6.466596233915601e-07,
|
| 1150 |
+
"loss": 0.6111,
|
| 1151 |
"step": 805
|
| 1152 |
},
|
| 1153 |
{
|
| 1154 |
+
"epoch": 1.8020022246941045,
|
| 1155 |
+
"grad_norm": 0.8285138016892091,
|
| 1156 |
+
"learning_rate": 5.796573653001091e-07,
|
| 1157 |
+
"loss": 0.6206,
|
| 1158 |
"step": 810
|
| 1159 |
},
|
| 1160 |
{
|
| 1161 |
+
"epoch": 1.8131256952169077,
|
| 1162 |
+
"grad_norm": 0.8098593367246673,
|
| 1163 |
+
"learning_rate": 5.162152716132662e-07,
|
| 1164 |
+
"loss": 0.6301,
|
| 1165 |
"step": 815
|
| 1166 |
},
|
| 1167 |
{
|
| 1168 |
+
"epoch": 1.8242491657397109,
|
| 1169 |
+
"grad_norm": 0.8317063889098834,
|
| 1170 |
+
"learning_rate": 4.563573185591219e-07,
|
| 1171 |
+
"loss": 0.5913,
|
| 1172 |
"step": 820
|
| 1173 |
},
|
| 1174 |
{
|
| 1175 |
+
"epoch": 1.8353726362625138,
|
| 1176 |
+
"grad_norm": 0.7802676617619877,
|
| 1177 |
+
"learning_rate": 4.0010612783648927e-07,
|
| 1178 |
+
"loss": 0.6009,
|
| 1179 |
"step": 825
|
| 1180 |
},
|
| 1181 |
{
|
| 1182 |
+
"epoch": 1.8464961067853172,
|
| 1183 |
+
"grad_norm": 0.8359715597402506,
|
| 1184 |
+
"learning_rate": 3.474829580656436e-07,
|
| 1185 |
+
"loss": 0.6129,
|
| 1186 |
"step": 830
|
| 1187 |
},
|
| 1188 |
{
|
| 1189 |
+
"epoch": 1.85761957730812,
|
| 1190 |
+
"grad_norm": 0.7764499156414768,
|
| 1191 |
+
"learning_rate": 2.9850769675419776e-07,
|
| 1192 |
+
"loss": 0.6233,
|
| 1193 |
"step": 835
|
| 1194 |
},
|
| 1195 |
{
|
| 1196 |
+
"epoch": 1.8687430478309233,
|
| 1197 |
+
"grad_norm": 0.8028424415714934,
|
| 1198 |
+
"learning_rate": 2.5319885278115907e-07,
|
| 1199 |
+
"loss": 0.6079,
|
| 1200 |
"step": 840
|
| 1201 |
},
|
| 1202 |
{
|
| 1203 |
+
"epoch": 1.8798665183537264,
|
| 1204 |
+
"grad_norm": 0.8831031088714258,
|
| 1205 |
+
"learning_rate": 2.115735494019966e-07,
|
| 1206 |
+
"loss": 0.6258,
|
| 1207 |
"step": 845
|
| 1208 |
},
|
| 1209 |
{
|
| 1210 |
+
"epoch": 1.8909899888765294,
|
| 1211 |
+
"grad_norm": 0.7653050119042255,
|
| 1212 |
+
"learning_rate": 1.7364751777736334e-07,
|
| 1213 |
+
"loss": 0.6212,
|
| 1214 |
"step": 850
|
| 1215 |
},
|
| 1216 |
{
|
| 1217 |
+
"epoch": 1.9021134593993327,
|
| 1218 |
+
"grad_norm": 0.8479149432867373,
|
| 1219 |
+
"learning_rate": 1.394350910279385e-07,
|
| 1220 |
+
"loss": 0.6006,
|
| 1221 |
"step": 855
|
| 1222 |
},
|
| 1223 |
{
|
| 1224 |
+
"epoch": 1.9132369299221357,
|
| 1225 |
+
"grad_norm": 0.8537537714787563,
|
| 1226 |
+
"learning_rate": 1.0894919881760168e-07,
|
| 1227 |
+
"loss": 0.6291,
|
| 1228 |
"step": 860
|
| 1229 |
},
|
| 1230 |
{
|
| 1231 |
+
"epoch": 1.9243604004449388,
|
| 1232 |
+
"grad_norm": 0.8315410821072657,
|
| 1233 |
+
"learning_rate": 8.220136246701926e-08,
|
| 1234 |
+
"loss": 0.6226,
|
| 1235 |
"step": 865
|
| 1236 |
},
|
| 1237 |
{
|
| 1238 |
+
"epoch": 1.935483870967742,
|
| 1239 |
+
"grad_norm": 0.8500259667306754,
|
| 1240 |
+
"learning_rate": 5.920169059947412e-08,
|
| 1241 |
+
"loss": 0.6108,
|
| 1242 |
"step": 870
|
| 1243 |
},
|
| 1244 |
{
|
| 1245 |
+
"epoch": 1.946607341490545,
|
| 1246 |
+
"grad_norm": 0.8522681381543702,
|
| 1247 |
+
"learning_rate": 3.99588753205804e-08,
|
| 1248 |
+
"loss": 0.6239,
|
| 1249 |
"step": 875
|
| 1250 |
},
|
| 1251 |
{
|
| 1252 |
+
"epoch": 1.9577308120133483,
|
| 1253 |
+
"grad_norm": 0.7583572716472294,
|
| 1254 |
+
"learning_rate": 2.4480188933336812e-08,
|
| 1255 |
+
"loss": 0.6128,
|
| 1256 |
"step": 880
|
| 1257 |
},
|
| 1258 |
{
|
| 1259 |
+
"epoch": 1.9688542825361512,
|
| 1260 |
+
"grad_norm": 0.7629282401468074,
|
| 1261 |
+
"learning_rate": 1.277148118975835e-08,
|
| 1262 |
+
"loss": 0.6022,
|
| 1263 |
"step": 885
|
| 1264 |
},
|
| 1265 |
{
|
| 1266 |
+
"epoch": 1.9799777530589544,
|
| 1267 |
+
"grad_norm": 0.817692770966463,
|
| 1268 |
+
"learning_rate": 4.837177080119215e-09,
|
| 1269 |
+
"loss": 0.6154,
|
| 1270 |
"step": 890
|
| 1271 |
},
|
| 1272 |
{
|
| 1273 |
+
"epoch": 1.9911012235817576,
|
| 1274 |
+
"grad_norm": 0.774109128612781,
|
| 1275 |
+
"learning_rate": 6.8027516064606e-10,
|
| 1276 |
+
"loss": 0.616,
|
| 1277 |
"step": 895
|
| 1278 |
},
|
| 1279 |
{
|
| 1280 |
+
"epoch": 1.9977753058954395,
|
| 1281 |
+
"eval_loss": 0.7532592415809631,
|
| 1282 |
+
"eval_runtime": 5.6688,
|
| 1283 |
+
"eval_samples_per_second": 71.797,
|
| 1284 |
+
"eval_steps_per_second": 2.293,
|
| 1285 |
+
"step": 898
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1286 |
},
|
| 1287 |
{
|
| 1288 |
+
"epoch": 1.9977753058954395,
|
| 1289 |
+
"step": 898,
|
| 1290 |
+
"total_flos": 81428314521600.0,
|
| 1291 |
+
"train_loss": 0.7065894536026868,
|
| 1292 |
+
"train_runtime": 5871.8562,
|
| 1293 |
+
"train_samples_per_second": 19.593,
|
| 1294 |
"train_steps_per_second": 0.153
|
| 1295 |
}
|
| 1296 |
],
|
| 1297 |
"logging_steps": 5,
|
| 1298 |
+
"max_steps": 898,
|
| 1299 |
"num_input_tokens_seen": 0,
|
| 1300 |
"num_train_epochs": 2,
|
| 1301 |
"save_steps": 100,
|
| 1302 |
+
"total_flos": 81428314521600.0,
|
| 1303 |
"train_batch_size": 16,
|
| 1304 |
"trial_name": null,
|
| 1305 |
"trial_params": null
|