Model save
Browse files
README.md
CHANGED
|
@@ -2,13 +2,12 @@
|
|
| 2 |
license: gemma
|
| 3 |
library_name: peft
|
| 4 |
tags:
|
| 5 |
-
- alignment-handbook
|
| 6 |
- trl
|
| 7 |
- sft
|
| 8 |
- generated_from_trainer
|
| 9 |
base_model: google/gemma-7b
|
| 10 |
datasets:
|
| 11 |
-
-
|
| 12 |
model-index:
|
| 13 |
- name: gemma7b-summarize-gemini1.5flash-30k
|
| 14 |
results: []
|
|
@@ -17,12 +16,12 @@ model-index:
|
|
| 17 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 18 |
should probably proofread and complete it, then remove this comment. -->
|
| 19 |
|
| 20 |
-
[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="200" height="32"/>](https://wandb.ai/chansung18/huggingface/runs/
|
| 21 |
# gemma7b-summarize-gemini1.5flash-30k
|
| 22 |
|
| 23 |
-
This model is a fine-tuned version of [google/gemma-7b](https://huggingface.co/google/gemma-7b) on the
|
| 24 |
It achieves the following results on the evaluation set:
|
| 25 |
-
- Loss:
|
| 26 |
|
| 27 |
## Model description
|
| 28 |
|
|
@@ -53,17 +52,22 @@ The following hyperparameters were used during training:
|
|
| 53 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 54 |
- lr_scheduler_type: cosine
|
| 55 |
- lr_scheduler_warmup_ratio: 0.1
|
| 56 |
-
- num_epochs:
|
| 57 |
|
| 58 |
### Training results
|
| 59 |
|
| 60 |
-
| Training Loss | Epoch
|
| 61 |
-
|
| 62 |
-
|
|
| 63 |
-
| 0.
|
| 64 |
-
| 0.
|
| 65 |
-
| 0.
|
| 66 |
-
| 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 67 |
|
| 68 |
|
| 69 |
### Framework versions
|
|
|
|
| 2 |
license: gemma
|
| 3 |
library_name: peft
|
| 4 |
tags:
|
|
|
|
| 5 |
- trl
|
| 6 |
- sft
|
| 7 |
- generated_from_trainer
|
| 8 |
base_model: google/gemma-7b
|
| 9 |
datasets:
|
| 10 |
+
- generator
|
| 11 |
model-index:
|
| 12 |
- name: gemma7b-summarize-gemini1.5flash-30k
|
| 13 |
results: []
|
|
|
|
| 16 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 17 |
should probably proofread and complete it, then remove this comment. -->
|
| 18 |
|
| 19 |
+
[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="200" height="32"/>](https://wandb.ai/chansung18/huggingface/runs/gx8z9rab)
|
| 20 |
# gemma7b-summarize-gemini1.5flash-30k
|
| 21 |
|
| 22 |
+
This model is a fine-tuned version of [google/gemma-7b](https://huggingface.co/google/gemma-7b) on the generator dataset.
|
| 23 |
It achieves the following results on the evaluation set:
|
| 24 |
+
- Loss: 3.3804
|
| 25 |
|
| 26 |
## Model description
|
| 27 |
|
|
|
|
| 52 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 53 |
- lr_scheduler_type: cosine
|
| 54 |
- lr_scheduler_warmup_ratio: 0.1
|
| 55 |
+
- num_epochs: 10
|
| 56 |
|
| 57 |
### Training results
|
| 58 |
|
| 59 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
| 60 |
+
|:-------------:|:-----:|:----:|:---------------:|
|
| 61 |
+
| 1.0256 | 1.0 | 110 | 2.3887 |
|
| 62 |
+
| 0.8325 | 2.0 | 220 | 2.2270 |
|
| 63 |
+
| 0.749 | 3.0 | 330 | 2.2333 |
|
| 64 |
+
| 0.6755 | 4.0 | 440 | 2.2993 |
|
| 65 |
+
| 0.6197 | 5.0 | 550 | 2.3820 |
|
| 66 |
+
| 0.5208 | 6.0 | 660 | 2.5869 |
|
| 67 |
+
| 0.4474 | 7.0 | 770 | 2.8389 |
|
| 68 |
+
| 0.4044 | 8.0 | 880 | 3.1029 |
|
| 69 |
+
| 0.3573 | 9.0 | 990 | 3.3573 |
|
| 70 |
+
| 0.354 | 10.0 | 1100 | 3.3804 |
|
| 71 |
|
| 72 |
|
| 73 |
### Framework versions
|
all_results.json
CHANGED
|
@@ -1,14 +1,9 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"
|
| 4 |
-
"
|
| 5 |
-
"
|
| 6 |
-
"
|
| 7 |
-
"
|
| 8 |
-
"
|
| 9 |
-
"train_loss": 2.449154797771521,
|
| 10 |
-
"train_runtime": 4532.0541,
|
| 11 |
-
"train_samples": 30386,
|
| 12 |
-
"train_samples_per_second": 2.02,
|
| 13 |
-
"train_steps_per_second": 0.126
|
| 14 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"total_flos": 1.6817604900715233e+18,
|
| 4 |
+
"train_loss": 1.8264882094209844,
|
| 5 |
+
"train_runtime": 8654.8771,
|
| 6 |
+
"train_samples": 32070,
|
| 7 |
+
"train_samples_per_second": 2.032,
|
| 8 |
+
"train_steps_per_second": 0.127
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 9 |
}
|
runs/May21_04-03-21_deep-diver-main-fluffy-octopus-1-0-0/events.out.tfevents.1716278746.deep-diver-main-fluffy-octopus-1-0-0.384.0
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:d399ed0c7b92a82aef9ca04d710770e51d053c79de1545fbab9634dff190f34b
|
| 3 |
+
size 55216
|
train_results.json
CHANGED
|
@@ -1,9 +1,9 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"total_flos":
|
| 4 |
-
"train_loss":
|
| 5 |
-
"train_runtime":
|
| 6 |
-
"train_samples":
|
| 7 |
-
"train_samples_per_second": 2.
|
| 8 |
-
"train_steps_per_second": 0.
|
| 9 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"total_flos": 1.6817604900715233e+18,
|
| 4 |
+
"train_loss": 1.8264882094209844,
|
| 5 |
+
"train_runtime": 8654.8771,
|
| 6 |
+
"train_samples": 32070,
|
| 7 |
+
"train_samples_per_second": 2.032,
|
| 8 |
+
"train_steps_per_second": 0.127
|
| 9 |
}
|
trainer_state.json
CHANGED
|
@@ -1,872 +1,1654 @@
|
|
| 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":
|
| 16 |
"step": 1
|
| 17 |
},
|
| 18 |
{
|
| 19 |
-
"epoch": 0.
|
| 20 |
-
"grad_norm":
|
| 21 |
-
"learning_rate":
|
| 22 |
-
"loss":
|
| 23 |
"step": 5
|
| 24 |
},
|
| 25 |
{
|
| 26 |
-
"epoch": 0.
|
| 27 |
-
"grad_norm":
|
| 28 |
-
"learning_rate":
|
| 29 |
-
"loss":
|
| 30 |
"step": 10
|
| 31 |
},
|
| 32 |
{
|
| 33 |
-
"epoch": 0.
|
| 34 |
-
"grad_norm":
|
| 35 |
-
"learning_rate":
|
| 36 |
-
"loss":
|
| 37 |
"step": 15
|
| 38 |
},
|
| 39 |
{
|
| 40 |
-
"epoch": 0.
|
| 41 |
-
"grad_norm":
|
| 42 |
-
"learning_rate":
|
| 43 |
-
"loss":
|
| 44 |
"step": 20
|
| 45 |
},
|
| 46 |
{
|
| 47 |
-
"epoch": 0.
|
| 48 |
-
"grad_norm":
|
| 49 |
-
"learning_rate":
|
| 50 |
-
"loss":
|
| 51 |
"step": 25
|
| 52 |
},
|
| 53 |
{
|
| 54 |
-
"epoch": 0.
|
| 55 |
-
"grad_norm":
|
| 56 |
-
"learning_rate":
|
| 57 |
-
"loss":
|
| 58 |
"step": 30
|
| 59 |
},
|
| 60 |
{
|
| 61 |
-
"epoch": 0.
|
| 62 |
-
"grad_norm":
|
| 63 |
-
"learning_rate":
|
| 64 |
-
"loss":
|
| 65 |
"step": 35
|
| 66 |
},
|
| 67 |
{
|
| 68 |
-
"epoch": 0.
|
| 69 |
-
"grad_norm":
|
| 70 |
-
"learning_rate":
|
| 71 |
-
"loss":
|
| 72 |
"step": 40
|
| 73 |
},
|
| 74 |
{
|
| 75 |
-
"epoch": 0.
|
| 76 |
-
"grad_norm":
|
| 77 |
-
"learning_rate":
|
| 78 |
-
"loss":
|
| 79 |
"step": 45
|
| 80 |
},
|
| 81 |
{
|
| 82 |
-
"epoch": 0.
|
| 83 |
-
"grad_norm":
|
| 84 |
-
"learning_rate":
|
| 85 |
-
"loss":
|
| 86 |
"step": 50
|
| 87 |
},
|
| 88 |
{
|
| 89 |
-
"epoch": 0.
|
| 90 |
-
"grad_norm":
|
| 91 |
-
"learning_rate": 0.
|
| 92 |
-
"loss":
|
| 93 |
"step": 55
|
| 94 |
},
|
| 95 |
{
|
| 96 |
-
"epoch": 0.
|
| 97 |
-
"grad_norm":
|
| 98 |
-
"learning_rate": 0.
|
| 99 |
-
"loss": 1.
|
| 100 |
"step": 60
|
| 101 |
},
|
| 102 |
{
|
| 103 |
-
"epoch": 0.
|
| 104 |
-
"grad_norm":
|
| 105 |
-
"learning_rate": 0.
|
| 106 |
-
"loss": 1.
|
| 107 |
"step": 65
|
| 108 |
},
|
| 109 |
{
|
| 110 |
-
"epoch": 0.
|
| 111 |
-
"grad_norm": 3.
|
| 112 |
-
"learning_rate": 0.
|
| 113 |
-
"loss": 1.
|
| 114 |
"step": 70
|
| 115 |
},
|
| 116 |
{
|
| 117 |
-
"epoch": 0.
|
| 118 |
-
"grad_norm":
|
| 119 |
-
"learning_rate": 0.
|
| 120 |
-
"loss": 1.
|
| 121 |
"step": 75
|
| 122 |
},
|
| 123 |
{
|
| 124 |
-
"epoch": 0.
|
| 125 |
-
"grad_norm":
|
| 126 |
-
"learning_rate": 0.
|
| 127 |
-
"loss": 1.
|
| 128 |
"step": 80
|
| 129 |
},
|
| 130 |
{
|
| 131 |
-
"epoch": 0.
|
| 132 |
-
"grad_norm":
|
| 133 |
-
"learning_rate": 0.
|
| 134 |
-
"loss":
|
| 135 |
"step": 85
|
| 136 |
},
|
| 137 |
{
|
| 138 |
-
"epoch": 0.
|
| 139 |
-
"grad_norm":
|
| 140 |
-
"learning_rate": 0.
|
| 141 |
-
"loss":
|
| 142 |
"step": 90
|
| 143 |
},
|
| 144 |
{
|
| 145 |
-
"epoch": 0.
|
| 146 |
-
"grad_norm": 4.
|
| 147 |
-
"learning_rate": 0.
|
| 148 |
-
"loss":
|
| 149 |
"step": 95
|
| 150 |
},
|
| 151 |
{
|
| 152 |
-
"epoch": 0.
|
| 153 |
-
"grad_norm":
|
| 154 |
-
"learning_rate": 0.
|
| 155 |
-
"loss":
|
| 156 |
"step": 100
|
| 157 |
},
|
| 158 |
{
|
| 159 |
-
"epoch": 0.
|
| 160 |
-
"grad_norm":
|
| 161 |
-
"learning_rate": 0.
|
| 162 |
-
"loss":
|
| 163 |
"step": 105
|
| 164 |
},
|
| 165 |
{
|
| 166 |
-
"epoch": 0
|
| 167 |
-
"grad_norm":
|
| 168 |
-
"learning_rate": 0.
|
| 169 |
-
"loss":
|
| 170 |
"step": 110
|
| 171 |
},
|
| 172 |
{
|
| 173 |
-
"epoch": 0
|
| 174 |
-
"eval_loss": 2.
|
| 175 |
-
"eval_runtime": 1.
|
| 176 |
-
"eval_samples_per_second": 4.
|
| 177 |
-
"eval_steps_per_second": 1.
|
| 178 |
-
"step":
|
| 179 |
},
|
| 180 |
{
|
| 181 |
-
"epoch": 1.
|
| 182 |
-
"grad_norm": 1.
|
| 183 |
-
"learning_rate": 0.
|
| 184 |
-
"loss":
|
| 185 |
"step": 115
|
| 186 |
},
|
| 187 |
{
|
| 188 |
-
"epoch": 1.
|
| 189 |
-
"grad_norm":
|
| 190 |
-
"learning_rate": 0.
|
| 191 |
-
"loss": 0.
|
| 192 |
"step": 120
|
| 193 |
},
|
| 194 |
{
|
| 195 |
-
"epoch": 1.
|
| 196 |
-
"grad_norm":
|
| 197 |
-
"learning_rate": 0.
|
| 198 |
-
"loss": 0.
|
| 199 |
"step": 125
|
| 200 |
},
|
| 201 |
{
|
| 202 |
-
"epoch": 1.
|
| 203 |
-
"grad_norm":
|
| 204 |
-
"learning_rate": 0.
|
| 205 |
-
"loss": 0.
|
| 206 |
"step": 130
|
| 207 |
},
|
| 208 |
{
|
| 209 |
-
"epoch": 1.
|
| 210 |
-
"grad_norm": 1.
|
| 211 |
-
"learning_rate": 0.
|
| 212 |
-
"loss": 0.
|
| 213 |
"step": 135
|
| 214 |
},
|
| 215 |
{
|
| 216 |
-
"epoch": 1.
|
| 217 |
-
"grad_norm": 1.
|
| 218 |
-
"learning_rate": 0.
|
| 219 |
-
"loss": 0.
|
| 220 |
"step": 140
|
| 221 |
},
|
| 222 |
{
|
| 223 |
-
"epoch": 1.
|
| 224 |
-
"grad_norm":
|
| 225 |
-
"learning_rate": 0.
|
| 226 |
-
"loss": 0.
|
| 227 |
"step": 145
|
| 228 |
},
|
| 229 |
{
|
| 230 |
-
"epoch": 1.
|
| 231 |
-
"grad_norm":
|
| 232 |
-
"learning_rate": 0.
|
| 233 |
-
"loss": 0.
|
| 234 |
"step": 150
|
| 235 |
},
|
| 236 |
{
|
| 237 |
-
"epoch": 1.
|
| 238 |
-
"grad_norm":
|
| 239 |
-
"learning_rate": 0.
|
| 240 |
-
"loss": 0.
|
| 241 |
"step": 155
|
| 242 |
},
|
| 243 |
{
|
| 244 |
-
"epoch": 1.
|
| 245 |
-
"grad_norm":
|
| 246 |
-
"learning_rate": 0.
|
| 247 |
-
"loss": 0.
|
| 248 |
"step": 160
|
| 249 |
},
|
| 250 |
{
|
| 251 |
-
"epoch": 1.
|
| 252 |
-
"grad_norm": 2.
|
| 253 |
-
"learning_rate": 0.
|
| 254 |
-
"loss": 0.
|
| 255 |
"step": 165
|
| 256 |
},
|
| 257 |
{
|
| 258 |
-
"epoch": 1.
|
| 259 |
-
"grad_norm":
|
| 260 |
-
"learning_rate": 0.
|
| 261 |
-
"loss": 0.
|
| 262 |
"step": 170
|
| 263 |
},
|
| 264 |
{
|
| 265 |
-
"epoch": 1.
|
| 266 |
-
"grad_norm":
|
| 267 |
-
"learning_rate": 0.
|
| 268 |
-
"loss": 0.
|
| 269 |
"step": 175
|
| 270 |
},
|
| 271 |
{
|
| 272 |
-
"epoch": 1.
|
| 273 |
-
"grad_norm": 1.
|
| 274 |
-
"learning_rate": 0.
|
| 275 |
-
"loss": 0.
|
| 276 |
"step": 180
|
| 277 |
},
|
| 278 |
{
|
| 279 |
-
"epoch": 1.
|
| 280 |
-
"grad_norm":
|
| 281 |
-
"learning_rate": 0.
|
| 282 |
-
"loss": 0.
|
| 283 |
"step": 185
|
| 284 |
},
|
| 285 |
{
|
| 286 |
-
"epoch": 1.
|
| 287 |
-
"grad_norm":
|
| 288 |
-
"learning_rate": 0.
|
| 289 |
-
"loss": 0.
|
| 290 |
"step": 190
|
| 291 |
},
|
| 292 |
{
|
| 293 |
-
"epoch": 1.
|
| 294 |
-
"grad_norm": 0.
|
| 295 |
-
"learning_rate": 0.
|
| 296 |
-
"loss": 0.
|
| 297 |
"step": 195
|
| 298 |
},
|
| 299 |
{
|
| 300 |
-
"epoch": 1.
|
| 301 |
-
"grad_norm":
|
| 302 |
-
"learning_rate": 0.
|
| 303 |
-
"loss": 0.
|
| 304 |
"step": 200
|
| 305 |
},
|
| 306 |
{
|
| 307 |
-
"epoch": 1.
|
| 308 |
-
"grad_norm": 1.
|
| 309 |
-
"learning_rate": 0.
|
| 310 |
-
"loss": 0.
|
| 311 |
"step": 205
|
| 312 |
},
|
| 313 |
{
|
| 314 |
-
"epoch": 1.
|
| 315 |
-
"grad_norm":
|
| 316 |
-
"learning_rate": 0.
|
| 317 |
-
"loss": 0.
|
| 318 |
"step": 210
|
| 319 |
},
|
| 320 |
{
|
| 321 |
-
"epoch": 1.
|
| 322 |
-
"grad_norm":
|
| 323 |
-
"learning_rate": 0.
|
| 324 |
-
"loss": 0.
|
| 325 |
"step": 215
|
| 326 |
},
|
| 327 |
{
|
| 328 |
-
"epoch":
|
| 329 |
-
"grad_norm": 0.
|
| 330 |
-
"learning_rate": 0.
|
| 331 |
-
"loss": 0.
|
| 332 |
"step": 220
|
| 333 |
},
|
| 334 |
{
|
| 335 |
-
"epoch":
|
| 336 |
-
"
|
| 337 |
-
"
|
| 338 |
-
"
|
| 339 |
-
"
|
|
|
|
| 340 |
},
|
| 341 |
{
|
| 342 |
-
"epoch": 2.
|
| 343 |
-
"
|
| 344 |
-
"
|
| 345 |
-
"
|
| 346 |
-
"
|
| 347 |
-
"step": 229
|
| 348 |
},
|
| 349 |
{
|
| 350 |
-
"epoch": 2.
|
| 351 |
-
"grad_norm":
|
| 352 |
-
"learning_rate": 0.
|
| 353 |
-
"loss": 0.
|
| 354 |
"step": 230
|
| 355 |
},
|
| 356 |
{
|
| 357 |
-
"epoch": 2.
|
| 358 |
-
"grad_norm": 0.
|
| 359 |
-
"learning_rate": 0.
|
| 360 |
-
"loss": 0.
|
| 361 |
"step": 235
|
| 362 |
},
|
| 363 |
{
|
| 364 |
-
"epoch": 2.
|
| 365 |
-
"grad_norm":
|
| 366 |
-
"learning_rate": 0.
|
| 367 |
-
"loss": 0.
|
| 368 |
"step": 240
|
| 369 |
},
|
| 370 |
{
|
| 371 |
-
"epoch": 2.
|
| 372 |
-
"grad_norm": 0.
|
| 373 |
-
"learning_rate": 0.
|
| 374 |
-
"loss": 0.
|
| 375 |
"step": 245
|
| 376 |
},
|
| 377 |
{
|
| 378 |
-
"epoch": 2.
|
| 379 |
-
"grad_norm":
|
| 380 |
-
"learning_rate": 0.
|
| 381 |
-
"loss": 0.
|
| 382 |
"step": 250
|
| 383 |
},
|
| 384 |
{
|
| 385 |
-
"epoch": 2.
|
| 386 |
-
"grad_norm":
|
| 387 |
-
"learning_rate": 0.
|
| 388 |
-
"loss": 0.
|
| 389 |
"step": 255
|
| 390 |
},
|
| 391 |
{
|
| 392 |
-
"epoch": 2.
|
| 393 |
-
"grad_norm": 1.
|
| 394 |
-
"learning_rate": 0.
|
| 395 |
-
"loss": 0.
|
| 396 |
"step": 260
|
| 397 |
},
|
| 398 |
{
|
| 399 |
-
"epoch": 2.
|
| 400 |
-
"grad_norm":
|
| 401 |
-
"learning_rate": 0.
|
| 402 |
-
"loss": 0.
|
| 403 |
"step": 265
|
| 404 |
},
|
| 405 |
{
|
| 406 |
-
"epoch": 2.
|
| 407 |
-
"grad_norm": 1.
|
| 408 |
-
"learning_rate": 0.
|
| 409 |
-
"loss": 0.
|
| 410 |
"step": 270
|
| 411 |
},
|
| 412 |
{
|
| 413 |
-
"epoch": 2.
|
| 414 |
-
"grad_norm": 1.
|
| 415 |
-
"learning_rate": 0.
|
| 416 |
-
"loss": 0.
|
| 417 |
"step": 275
|
| 418 |
},
|
| 419 |
{
|
| 420 |
-
"epoch": 2.
|
| 421 |
-
"grad_norm": 1.
|
| 422 |
-
"learning_rate": 0.
|
| 423 |
-
"loss": 0.
|
| 424 |
"step": 280
|
| 425 |
},
|
| 426 |
{
|
| 427 |
-
"epoch": 2.
|
| 428 |
-
"grad_norm": 0.
|
| 429 |
-
"learning_rate": 0.
|
| 430 |
-
"loss": 0.
|
| 431 |
"step": 285
|
| 432 |
},
|
| 433 |
{
|
| 434 |
-
"epoch": 2.
|
| 435 |
-
"grad_norm":
|
| 436 |
-
"learning_rate": 0.
|
| 437 |
-
"loss": 0.
|
| 438 |
"step": 290
|
| 439 |
},
|
| 440 |
{
|
| 441 |
-
"epoch": 2.
|
| 442 |
-
"grad_norm":
|
| 443 |
-
"learning_rate": 0.
|
| 444 |
-
"loss": 0.
|
| 445 |
"step": 295
|
| 446 |
},
|
| 447 |
{
|
| 448 |
-
"epoch": 2.
|
| 449 |
-
"grad_norm": 0
|
| 450 |
-
"learning_rate": 0.
|
| 451 |
-
"loss": 0.
|
| 452 |
"step": 300
|
| 453 |
},
|
| 454 |
{
|
| 455 |
-
"epoch": 2.
|
| 456 |
-
"grad_norm":
|
| 457 |
-
"learning_rate": 0.
|
| 458 |
-
"loss": 0.
|
| 459 |
"step": 305
|
| 460 |
},
|
| 461 |
{
|
| 462 |
-
"epoch": 2.
|
| 463 |
-
"grad_norm":
|
| 464 |
-
"learning_rate": 0.
|
| 465 |
-
"loss": 0.
|
| 466 |
"step": 310
|
| 467 |
},
|
| 468 |
{
|
| 469 |
-
"epoch": 2.
|
| 470 |
-
"grad_norm":
|
| 471 |
-
"learning_rate":
|
| 472 |
-
"loss": 0.
|
| 473 |
"step": 315
|
| 474 |
},
|
| 475 |
{
|
| 476 |
-
"epoch": 2.
|
| 477 |
-
"grad_norm":
|
| 478 |
-
"learning_rate":
|
| 479 |
-
"loss": 0.
|
| 480 |
"step": 320
|
| 481 |
},
|
| 482 |
{
|
| 483 |
-
"epoch": 2.
|
| 484 |
-
"grad_norm": 0.
|
| 485 |
-
"learning_rate":
|
| 486 |
-
"loss": 0.
|
| 487 |
"step": 325
|
| 488 |
},
|
| 489 |
{
|
| 490 |
-
"epoch":
|
| 491 |
-
"grad_norm":
|
| 492 |
-
"learning_rate":
|
| 493 |
-
"loss": 0.
|
| 494 |
"step": 330
|
| 495 |
},
|
| 496 |
{
|
| 497 |
-
"epoch":
|
| 498 |
-
"
|
| 499 |
-
"
|
| 500 |
-
"
|
| 501 |
-
"
|
|
|
|
| 502 |
},
|
| 503 |
{
|
| 504 |
-
"epoch":
|
| 505 |
-
"grad_norm":
|
| 506 |
-
"learning_rate":
|
| 507 |
-
"loss": 0.
|
| 508 |
-
"step":
|
| 509 |
},
|
| 510 |
{
|
| 511 |
-
"epoch":
|
| 512 |
-
"
|
| 513 |
-
"
|
| 514 |
-
"
|
| 515 |
-
"
|
| 516 |
-
"step": 343
|
| 517 |
},
|
| 518 |
{
|
| 519 |
-
"epoch": 3.
|
| 520 |
-
"grad_norm": 0.
|
| 521 |
-
"learning_rate":
|
| 522 |
-
"loss": 0.
|
| 523 |
"step": 345
|
| 524 |
},
|
| 525 |
{
|
| 526 |
-
"epoch": 3.
|
| 527 |
-
"grad_norm": 0.
|
| 528 |
-
"learning_rate":
|
| 529 |
-
"loss": 0.
|
| 530 |
"step": 350
|
| 531 |
},
|
| 532 |
{
|
| 533 |
-
"epoch": 3.
|
| 534 |
-
"grad_norm":
|
| 535 |
-
"learning_rate":
|
| 536 |
-
"loss": 0.
|
| 537 |
"step": 355
|
| 538 |
},
|
| 539 |
{
|
| 540 |
-
"epoch": 3.
|
| 541 |
-
"grad_norm":
|
| 542 |
-
"learning_rate":
|
| 543 |
-
"loss": 0.
|
| 544 |
"step": 360
|
| 545 |
},
|
| 546 |
{
|
| 547 |
-
"epoch": 3.
|
| 548 |
-
"grad_norm": 0.
|
| 549 |
-
"learning_rate":
|
| 550 |
-
"loss": 0.
|
| 551 |
"step": 365
|
| 552 |
},
|
| 553 |
{
|
| 554 |
-
"epoch": 3.
|
| 555 |
-
"grad_norm": 0.
|
| 556 |
-
"learning_rate":
|
| 557 |
-
"loss": 0.
|
| 558 |
"step": 370
|
| 559 |
},
|
| 560 |
{
|
| 561 |
-
"epoch": 3.
|
| 562 |
-
"grad_norm": 0.
|
| 563 |
-
"learning_rate":
|
| 564 |
-
"loss": 0.
|
| 565 |
"step": 375
|
| 566 |
},
|
| 567 |
{
|
| 568 |
-
"epoch": 3.
|
| 569 |
-
"grad_norm":
|
| 570 |
-
"learning_rate":
|
| 571 |
-
"loss": 0.
|
| 572 |
"step": 380
|
| 573 |
},
|
| 574 |
{
|
| 575 |
-
"epoch": 3.
|
| 576 |
-
"grad_norm":
|
| 577 |
-
"learning_rate":
|
| 578 |
-
"loss": 0.
|
| 579 |
"step": 385
|
| 580 |
},
|
| 581 |
{
|
| 582 |
-
"epoch": 3.
|
| 583 |
-
"grad_norm":
|
| 584 |
-
"learning_rate":
|
| 585 |
-
"loss": 0.
|
| 586 |
"step": 390
|
| 587 |
},
|
| 588 |
{
|
| 589 |
-
"epoch": 3.
|
| 590 |
-
"grad_norm":
|
| 591 |
-
"learning_rate":
|
| 592 |
-
"loss": 0.
|
| 593 |
"step": 395
|
| 594 |
},
|
| 595 |
{
|
| 596 |
-
"epoch": 3.
|
| 597 |
-
"grad_norm": 0.
|
| 598 |
-
"learning_rate":
|
| 599 |
-
"loss": 0.
|
| 600 |
"step": 400
|
| 601 |
},
|
| 602 |
{
|
| 603 |
-
"epoch": 3.
|
| 604 |
-
"grad_norm":
|
| 605 |
-
"learning_rate":
|
| 606 |
-
"loss": 0.
|
| 607 |
"step": 405
|
| 608 |
},
|
| 609 |
{
|
| 610 |
-
"epoch": 3.
|
| 611 |
-
"grad_norm": 0.
|
| 612 |
-
"learning_rate":
|
| 613 |
-
"loss": 0.
|
| 614 |
"step": 410
|
| 615 |
},
|
| 616 |
{
|
| 617 |
-
"epoch": 3.
|
| 618 |
-
"grad_norm":
|
| 619 |
-
"learning_rate":
|
| 620 |
-
"loss": 0.
|
| 621 |
"step": 415
|
| 622 |
},
|
| 623 |
{
|
| 624 |
-
"epoch": 3.
|
| 625 |
-
"grad_norm": 0.
|
| 626 |
-
"learning_rate":
|
| 627 |
-
"loss": 0.
|
| 628 |
"step": 420
|
| 629 |
},
|
| 630 |
{
|
| 631 |
-
"epoch": 3.
|
| 632 |
-
"grad_norm":
|
| 633 |
-
"learning_rate":
|
| 634 |
-
"loss": 0.
|
| 635 |
"step": 425
|
| 636 |
},
|
| 637 |
{
|
| 638 |
-
"epoch": 3.
|
| 639 |
-
"grad_norm":
|
| 640 |
-
"learning_rate":
|
| 641 |
-
"loss": 0.
|
| 642 |
"step": 430
|
| 643 |
},
|
| 644 |
{
|
| 645 |
-
"epoch": 3.
|
| 646 |
-
"grad_norm": 0.
|
| 647 |
-
"learning_rate":
|
| 648 |
-
"loss": 0.
|
| 649 |
"step": 435
|
| 650 |
},
|
| 651 |
{
|
| 652 |
-
"epoch":
|
| 653 |
-
"grad_norm":
|
| 654 |
-
"learning_rate":
|
| 655 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 656 |
"step": 440
|
| 657 |
},
|
| 658 |
{
|
| 659 |
-
"epoch":
|
| 660 |
-
"grad_norm":
|
| 661 |
-
"learning_rate":
|
| 662 |
-
"loss": 0.
|
| 663 |
"step": 445
|
| 664 |
},
|
| 665 |
{
|
| 666 |
-
"epoch":
|
| 667 |
-
"grad_norm": 0.
|
| 668 |
-
"learning_rate":
|
| 669 |
-
"loss": 0.
|
| 670 |
"step": 450
|
| 671 |
},
|
| 672 |
{
|
| 673 |
-
"epoch":
|
| 674 |
-
"grad_norm": 0.
|
| 675 |
-
"learning_rate":
|
| 676 |
-
"loss": 0.
|
| 677 |
"step": 455
|
| 678 |
},
|
| 679 |
{
|
| 680 |
-
"epoch": 4.
|
| 681 |
-
"
|
| 682 |
-
"
|
| 683 |
-
"
|
| 684 |
-
"eval_steps_per_second": 1.999,
|
| 685 |
-
"step": 458
|
| 686 |
-
},
|
| 687 |
-
{
|
| 688 |
-
"epoch": 4.0174672489082965,
|
| 689 |
-
"grad_norm": 0.458984375,
|
| 690 |
-
"learning_rate": 2.184413228954468e-05,
|
| 691 |
-
"loss": 0.5676,
|
| 692 |
"step": 460
|
| 693 |
},
|
| 694 |
{
|
| 695 |
-
"epoch": 4.
|
| 696 |
-
"grad_norm": 0.
|
| 697 |
-
"learning_rate":
|
| 698 |
-
"loss": 0.
|
| 699 |
"step": 465
|
| 700 |
},
|
| 701 |
{
|
| 702 |
-
"epoch": 4.
|
| 703 |
-
"grad_norm": 0.
|
| 704 |
-
"learning_rate":
|
| 705 |
-
"loss": 0.
|
| 706 |
"step": 470
|
| 707 |
},
|
| 708 |
{
|
| 709 |
-
"epoch": 4.
|
| 710 |
-
"grad_norm": 0.
|
| 711 |
-
"learning_rate":
|
| 712 |
-
"loss": 0.
|
| 713 |
"step": 475
|
| 714 |
},
|
| 715 |
{
|
| 716 |
-
"epoch": 4.
|
| 717 |
-
"grad_norm": 0.
|
| 718 |
-
"learning_rate":
|
| 719 |
-
"loss": 0.
|
| 720 |
"step": 480
|
| 721 |
},
|
| 722 |
{
|
| 723 |
-
"epoch": 4.
|
| 724 |
-
"grad_norm": 0.
|
| 725 |
-
"learning_rate":
|
| 726 |
-
"loss": 0.
|
| 727 |
"step": 485
|
| 728 |
},
|
| 729 |
{
|
| 730 |
-
"epoch": 4.
|
| 731 |
-
"grad_norm": 0.
|
| 732 |
-
"learning_rate":
|
| 733 |
-
"loss": 0.
|
| 734 |
"step": 490
|
| 735 |
},
|
| 736 |
{
|
| 737 |
-
"epoch": 4.
|
| 738 |
-
"grad_norm":
|
| 739 |
-
"learning_rate":
|
| 740 |
-
"loss": 0.
|
| 741 |
"step": 495
|
| 742 |
},
|
| 743 |
{
|
| 744 |
-
"epoch": 4.
|
| 745 |
-
"grad_norm": 0.
|
| 746 |
-
"learning_rate":
|
| 747 |
-
"loss": 0.
|
| 748 |
"step": 500
|
| 749 |
},
|
| 750 |
{
|
| 751 |
-
"epoch": 4.
|
| 752 |
-
"grad_norm":
|
| 753 |
-
"learning_rate":
|
| 754 |
-
"loss": 0.
|
| 755 |
"step": 505
|
| 756 |
},
|
| 757 |
{
|
| 758 |
-
"epoch": 4.
|
| 759 |
-
"grad_norm":
|
| 760 |
-
"learning_rate":
|
| 761 |
-
"loss": 0.
|
| 762 |
"step": 510
|
| 763 |
},
|
| 764 |
{
|
| 765 |
-
"epoch": 4.
|
| 766 |
-
"grad_norm":
|
| 767 |
-
"learning_rate":
|
| 768 |
-
"loss": 0.
|
| 769 |
"step": 515
|
| 770 |
},
|
| 771 |
{
|
| 772 |
-
"epoch": 4.
|
| 773 |
-
"grad_norm": 0.
|
| 774 |
-
"learning_rate":
|
| 775 |
-
"loss": 0.
|
| 776 |
"step": 520
|
| 777 |
},
|
| 778 |
{
|
| 779 |
-
"epoch": 4.
|
| 780 |
-
"grad_norm": 0.
|
| 781 |
-
"learning_rate":
|
| 782 |
-
"loss": 0.
|
| 783 |
"step": 525
|
| 784 |
},
|
| 785 |
{
|
| 786 |
-
"epoch": 4.
|
| 787 |
-
"grad_norm": 0.
|
| 788 |
-
"learning_rate":
|
| 789 |
-
"loss": 0.
|
| 790 |
"step": 530
|
| 791 |
},
|
| 792 |
{
|
| 793 |
-
"epoch": 4.
|
| 794 |
-
"grad_norm": 0.
|
| 795 |
-
"learning_rate":
|
| 796 |
-
"loss": 0.
|
| 797 |
"step": 535
|
| 798 |
},
|
| 799 |
{
|
| 800 |
-
"epoch": 4.
|
| 801 |
-
"grad_norm": 0.
|
| 802 |
-
"learning_rate":
|
| 803 |
-
"loss": 0.
|
| 804 |
"step": 540
|
| 805 |
},
|
| 806 |
{
|
| 807 |
-
"epoch": 4.
|
| 808 |
-
"grad_norm": 0.
|
| 809 |
-
"learning_rate":
|
| 810 |
-
"loss": 0.
|
| 811 |
"step": 545
|
| 812 |
},
|
| 813 |
{
|
| 814 |
-
"epoch":
|
| 815 |
-
"grad_norm": 0.
|
| 816 |
-
"learning_rate":
|
| 817 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 818 |
"step": 550
|
| 819 |
},
|
| 820 |
{
|
| 821 |
-
"epoch":
|
| 822 |
-
"grad_norm": 0.
|
| 823 |
-
"learning_rate":
|
| 824 |
-
"loss": 0.
|
| 825 |
"step": 555
|
| 826 |
},
|
| 827 |
{
|
| 828 |
-
"epoch":
|
| 829 |
-
"grad_norm": 0.
|
| 830 |
-
"learning_rate":
|
| 831 |
-
"loss": 0.
|
| 832 |
"step": 560
|
| 833 |
},
|
| 834 |
{
|
| 835 |
-
"epoch":
|
| 836 |
-
"grad_norm": 0.
|
| 837 |
-
"learning_rate":
|
| 838 |
-
"loss": 0.
|
| 839 |
"step": 565
|
| 840 |
},
|
| 841 |
{
|
| 842 |
-
"epoch":
|
| 843 |
-
"grad_norm": 0.
|
| 844 |
-
"learning_rate": 0.
|
| 845 |
-
"loss": 0.
|
| 846 |
"step": 570
|
| 847 |
},
|
| 848 |
{
|
| 849 |
-
"epoch":
|
| 850 |
-
"
|
| 851 |
-
"
|
| 852 |
-
"
|
| 853 |
-
"
|
| 854 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 855 |
},
|
| 856 |
{
|
| 857 |
-
"epoch":
|
| 858 |
-
"step":
|
| 859 |
-
"total_flos":
|
| 860 |
-
"train_loss":
|
| 861 |
-
"train_runtime":
|
| 862 |
-
"train_samples_per_second": 2.
|
| 863 |
-
"train_steps_per_second": 0.
|
| 864 |
}
|
| 865 |
],
|
| 866 |
"logging_steps": 5,
|
| 867 |
-
"max_steps":
|
| 868 |
"num_input_tokens_seen": 0,
|
| 869 |
-
"num_train_epochs":
|
| 870 |
"save_steps": 100,
|
| 871 |
"stateful_callbacks": {
|
| 872 |
"TrainerControl": {
|
|
@@ -875,12 +1657,12 @@
|
|
| 875 |
"should_evaluate": false,
|
| 876 |
"should_log": false,
|
| 877 |
"should_save": true,
|
| 878 |
-
"should_training_stop":
|
| 879 |
},
|
| 880 |
"attributes": {}
|
| 881 |
}
|
| 882 |
},
|
| 883 |
-
"total_flos":
|
| 884 |
"train_batch_size": 4,
|
| 885 |
"trial_name": null,
|
| 886 |
"trial_params": null
|
|
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 10.0,
|
| 5 |
"eval_steps": 500,
|
| 6 |
+
"global_step": 1100,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
+
"epoch": 0.00909090909090909,
|
| 13 |
+
"grad_norm": 548.0,
|
| 14 |
+
"learning_rate": 1.818181818181818e-06,
|
| 15 |
+
"loss": 53.7965,
|
| 16 |
"step": 1
|
| 17 |
},
|
| 18 |
{
|
| 19 |
+
"epoch": 0.045454545454545456,
|
| 20 |
+
"grad_norm": 668.0,
|
| 21 |
+
"learning_rate": 9.090909090909091e-06,
|
| 22 |
+
"loss": 55.3182,
|
| 23 |
"step": 5
|
| 24 |
},
|
| 25 |
{
|
| 26 |
+
"epoch": 0.09090909090909091,
|
| 27 |
+
"grad_norm": 486.0,
|
| 28 |
+
"learning_rate": 1.8181818181818182e-05,
|
| 29 |
+
"loss": 51.173,
|
| 30 |
"step": 10
|
| 31 |
},
|
| 32 |
{
|
| 33 |
+
"epoch": 0.13636363636363635,
|
| 34 |
+
"grad_norm": 159.0,
|
| 35 |
+
"learning_rate": 2.7272727272727273e-05,
|
| 36 |
+
"loss": 33.7467,
|
| 37 |
"step": 15
|
| 38 |
},
|
| 39 |
{
|
| 40 |
+
"epoch": 0.18181818181818182,
|
| 41 |
+
"grad_norm": 37.5,
|
| 42 |
+
"learning_rate": 3.6363636363636364e-05,
|
| 43 |
+
"loss": 24.0423,
|
| 44 |
"step": 20
|
| 45 |
},
|
| 46 |
{
|
| 47 |
+
"epoch": 0.22727272727272727,
|
| 48 |
+
"grad_norm": 32.25,
|
| 49 |
+
"learning_rate": 4.545454545454546e-05,
|
| 50 |
+
"loss": 22.1424,
|
| 51 |
"step": 25
|
| 52 |
},
|
| 53 |
{
|
| 54 |
+
"epoch": 0.2727272727272727,
|
| 55 |
+
"grad_norm": 21.25,
|
| 56 |
+
"learning_rate": 5.4545454545454546e-05,
|
| 57 |
+
"loss": 20.3818,
|
| 58 |
"step": 30
|
| 59 |
},
|
| 60 |
{
|
| 61 |
+
"epoch": 0.3181818181818182,
|
| 62 |
+
"grad_norm": 7.90625,
|
| 63 |
+
"learning_rate": 6.363636363636364e-05,
|
| 64 |
+
"loss": 18.8064,
|
| 65 |
"step": 35
|
| 66 |
},
|
| 67 |
{
|
| 68 |
+
"epoch": 0.36363636363636365,
|
| 69 |
+
"grad_norm": 11.625,
|
| 70 |
+
"learning_rate": 7.272727272727273e-05,
|
| 71 |
+
"loss": 17.8515,
|
| 72 |
"step": 40
|
| 73 |
},
|
| 74 |
{
|
| 75 |
+
"epoch": 0.4090909090909091,
|
| 76 |
+
"grad_norm": 27.125,
|
| 77 |
+
"learning_rate": 8.181818181818183e-05,
|
| 78 |
+
"loss": 16.8604,
|
| 79 |
"step": 45
|
| 80 |
},
|
| 81 |
{
|
| 82 |
+
"epoch": 0.45454545454545453,
|
| 83 |
+
"grad_norm": 56.25,
|
| 84 |
+
"learning_rate": 9.090909090909092e-05,
|
| 85 |
+
"loss": 12.4796,
|
| 86 |
"step": 50
|
| 87 |
},
|
| 88 |
{
|
| 89 |
+
"epoch": 0.5,
|
| 90 |
+
"grad_norm": 11.4375,
|
| 91 |
+
"learning_rate": 0.0001,
|
| 92 |
+
"loss": 4.4979,
|
| 93 |
"step": 55
|
| 94 |
},
|
| 95 |
{
|
| 96 |
+
"epoch": 0.5454545454545454,
|
| 97 |
+
"grad_norm": 3.25,
|
| 98 |
+
"learning_rate": 0.00010909090909090909,
|
| 99 |
+
"loss": 1.9919,
|
| 100 |
"step": 60
|
| 101 |
},
|
| 102 |
{
|
| 103 |
+
"epoch": 0.5909090909090909,
|
| 104 |
+
"grad_norm": 25.375,
|
| 105 |
+
"learning_rate": 0.0001181818181818182,
|
| 106 |
+
"loss": 1.6619,
|
| 107 |
"step": 65
|
| 108 |
},
|
| 109 |
{
|
| 110 |
+
"epoch": 0.6363636363636364,
|
| 111 |
+
"grad_norm": 3.125,
|
| 112 |
+
"learning_rate": 0.00012727272727272728,
|
| 113 |
+
"loss": 1.527,
|
| 114 |
"step": 70
|
| 115 |
},
|
| 116 |
{
|
| 117 |
+
"epoch": 0.6818181818181818,
|
| 118 |
+
"grad_norm": 9.6875,
|
| 119 |
+
"learning_rate": 0.00013636363636363637,
|
| 120 |
+
"loss": 1.3688,
|
| 121 |
"step": 75
|
| 122 |
},
|
| 123 |
{
|
| 124 |
+
"epoch": 0.7272727272727273,
|
| 125 |
+
"grad_norm": 8.0625,
|
| 126 |
+
"learning_rate": 0.00014545454545454546,
|
| 127 |
+
"loss": 1.2618,
|
| 128 |
"step": 80
|
| 129 |
},
|
| 130 |
{
|
| 131 |
+
"epoch": 0.7727272727272727,
|
| 132 |
+
"grad_norm": 6.9375,
|
| 133 |
+
"learning_rate": 0.00015454545454545454,
|
| 134 |
+
"loss": 1.238,
|
| 135 |
"step": 85
|
| 136 |
},
|
| 137 |
{
|
| 138 |
+
"epoch": 0.8181818181818182,
|
| 139 |
+
"grad_norm": 3.15625,
|
| 140 |
+
"learning_rate": 0.00016363636363636366,
|
| 141 |
+
"loss": 1.1871,
|
| 142 |
"step": 90
|
| 143 |
},
|
| 144 |
{
|
| 145 |
+
"epoch": 0.8636363636363636,
|
| 146 |
+
"grad_norm": 4.34375,
|
| 147 |
+
"learning_rate": 0.00017272727272727275,
|
| 148 |
+
"loss": 1.154,
|
| 149 |
"step": 95
|
| 150 |
},
|
| 151 |
{
|
| 152 |
+
"epoch": 0.9090909090909091,
|
| 153 |
+
"grad_norm": 2.65625,
|
| 154 |
+
"learning_rate": 0.00018181818181818183,
|
| 155 |
+
"loss": 1.1375,
|
| 156 |
"step": 100
|
| 157 |
},
|
| 158 |
{
|
| 159 |
+
"epoch": 0.9545454545454546,
|
| 160 |
+
"grad_norm": 3.296875,
|
| 161 |
+
"learning_rate": 0.00019090909090909092,
|
| 162 |
+
"loss": 1.0352,
|
| 163 |
"step": 105
|
| 164 |
},
|
| 165 |
{
|
| 166 |
+
"epoch": 1.0,
|
| 167 |
+
"grad_norm": 9.25,
|
| 168 |
+
"learning_rate": 0.0002,
|
| 169 |
+
"loss": 1.0256,
|
| 170 |
"step": 110
|
| 171 |
},
|
| 172 |
{
|
| 173 |
+
"epoch": 1.0,
|
| 174 |
+
"eval_loss": 2.388709783554077,
|
| 175 |
+
"eval_runtime": 1.0069,
|
| 176 |
+
"eval_samples_per_second": 4.966,
|
| 177 |
+
"eval_steps_per_second": 1.986,
|
| 178 |
+
"step": 110
|
| 179 |
},
|
| 180 |
{
|
| 181 |
+
"epoch": 1.0454545454545454,
|
| 182 |
+
"grad_norm": 1.53125,
|
| 183 |
+
"learning_rate": 0.00019998741276738754,
|
| 184 |
+
"loss": 1.0186,
|
| 185 |
"step": 115
|
| 186 |
},
|
| 187 |
{
|
| 188 |
+
"epoch": 1.0909090909090908,
|
| 189 |
+
"grad_norm": 2.5625,
|
| 190 |
+
"learning_rate": 0.00019994965423831854,
|
| 191 |
+
"loss": 0.9466,
|
| 192 |
"step": 120
|
| 193 |
},
|
| 194 |
{
|
| 195 |
+
"epoch": 1.1363636363636362,
|
| 196 |
+
"grad_norm": 7.21875,
|
| 197 |
+
"learning_rate": 0.0001998867339183008,
|
| 198 |
+
"loss": 0.9645,
|
| 199 |
"step": 125
|
| 200 |
},
|
| 201 |
{
|
| 202 |
+
"epoch": 1.1818181818181819,
|
| 203 |
+
"grad_norm": 86.5,
|
| 204 |
+
"learning_rate": 0.00019979866764718843,
|
| 205 |
+
"loss": 0.9547,
|
| 206 |
"step": 130
|
| 207 |
},
|
| 208 |
{
|
| 209 |
+
"epoch": 1.2272727272727273,
|
| 210 |
+
"grad_norm": 1.203125,
|
| 211 |
+
"learning_rate": 0.00019968547759519425,
|
| 212 |
+
"loss": 0.9527,
|
| 213 |
"step": 135
|
| 214 |
},
|
| 215 |
{
|
| 216 |
+
"epoch": 1.2727272727272727,
|
| 217 |
+
"grad_norm": 1.265625,
|
| 218 |
+
"learning_rate": 0.00019954719225730847,
|
| 219 |
+
"loss": 0.9091,
|
| 220 |
"step": 140
|
| 221 |
},
|
| 222 |
{
|
| 223 |
+
"epoch": 1.3181818181818181,
|
| 224 |
+
"grad_norm": 1.703125,
|
| 225 |
+
"learning_rate": 0.00019938384644612543,
|
| 226 |
+
"loss": 0.8789,
|
| 227 |
"step": 145
|
| 228 |
},
|
| 229 |
{
|
| 230 |
+
"epoch": 1.3636363636363638,
|
| 231 |
+
"grad_norm": 2.40625,
|
| 232 |
+
"learning_rate": 0.00019919548128307954,
|
| 233 |
+
"loss": 0.9036,
|
| 234 |
"step": 150
|
| 235 |
},
|
| 236 |
{
|
| 237 |
+
"epoch": 1.4090909090909092,
|
| 238 |
+
"grad_norm": 5.09375,
|
| 239 |
+
"learning_rate": 0.0001989821441880933,
|
| 240 |
+
"loss": 0.8787,
|
| 241 |
"step": 155
|
| 242 |
},
|
| 243 |
{
|
| 244 |
+
"epoch": 1.4545454545454546,
|
| 245 |
+
"grad_norm": 1.0859375,
|
| 246 |
+
"learning_rate": 0.00019874388886763944,
|
| 247 |
+
"loss": 0.8671,
|
| 248 |
"step": 160
|
| 249 |
},
|
| 250 |
{
|
| 251 |
+
"epoch": 1.5,
|
| 252 |
+
"grad_norm": 2.046875,
|
| 253 |
+
"learning_rate": 0.00019848077530122083,
|
| 254 |
+
"loss": 0.8872,
|
| 255 |
"step": 165
|
| 256 |
},
|
| 257 |
{
|
| 258 |
+
"epoch": 1.5454545454545454,
|
| 259 |
+
"grad_norm": 4.0625,
|
| 260 |
+
"learning_rate": 0.00019819286972627066,
|
| 261 |
+
"loss": 0.9179,
|
| 262 |
"step": 170
|
| 263 |
},
|
| 264 |
{
|
| 265 |
+
"epoch": 1.5909090909090908,
|
| 266 |
+
"grad_norm": 3.25,
|
| 267 |
+
"learning_rate": 0.00019788024462147788,
|
| 268 |
+
"loss": 0.8857,
|
| 269 |
"step": 175
|
| 270 |
},
|
| 271 |
{
|
| 272 |
+
"epoch": 1.6363636363636362,
|
| 273 |
+
"grad_norm": 1.9140625,
|
| 274 |
+
"learning_rate": 0.00019754297868854073,
|
| 275 |
+
"loss": 0.8474,
|
| 276 |
"step": 180
|
| 277 |
},
|
| 278 |
{
|
| 279 |
+
"epoch": 1.6818181818181817,
|
| 280 |
+
"grad_norm": 3.03125,
|
| 281 |
+
"learning_rate": 0.00019718115683235417,
|
| 282 |
+
"loss": 0.861,
|
| 283 |
"step": 185
|
| 284 |
},
|
| 285 |
{
|
| 286 |
+
"epoch": 1.7272727272727273,
|
| 287 |
+
"grad_norm": 3.515625,
|
| 288 |
+
"learning_rate": 0.00019679487013963564,
|
| 289 |
+
"loss": 0.8266,
|
| 290 |
"step": 190
|
| 291 |
},
|
| 292 |
{
|
| 293 |
+
"epoch": 1.7727272727272727,
|
| 294 |
+
"grad_norm": 0.9140625,
|
| 295 |
+
"learning_rate": 0.00019638421585599423,
|
| 296 |
+
"loss": 0.8515,
|
| 297 |
"step": 195
|
| 298 |
},
|
| 299 |
{
|
| 300 |
+
"epoch": 1.8181818181818183,
|
| 301 |
+
"grad_norm": 2.015625,
|
| 302 |
+
"learning_rate": 0.00019594929736144976,
|
| 303 |
+
"loss": 0.8328,
|
| 304 |
"step": 200
|
| 305 |
},
|
| 306 |
{
|
| 307 |
+
"epoch": 1.8636363636363638,
|
| 308 |
+
"grad_norm": 1.7578125,
|
| 309 |
+
"learning_rate": 0.0001954902241444074,
|
| 310 |
+
"loss": 0.8601,
|
| 311 |
"step": 205
|
| 312 |
},
|
| 313 |
{
|
| 314 |
+
"epoch": 1.9090909090909092,
|
| 315 |
+
"grad_norm": 1.65625,
|
| 316 |
+
"learning_rate": 0.00019500711177409454,
|
| 317 |
+
"loss": 0.8435,
|
| 318 |
"step": 210
|
| 319 |
},
|
| 320 |
{
|
| 321 |
+
"epoch": 1.9545454545454546,
|
| 322 |
+
"grad_norm": 2.421875,
|
| 323 |
+
"learning_rate": 0.00019450008187146684,
|
| 324 |
+
"loss": 0.8082,
|
| 325 |
"step": 215
|
| 326 |
},
|
| 327 |
{
|
| 328 |
+
"epoch": 2.0,
|
| 329 |
+
"grad_norm": 0.75390625,
|
| 330 |
+
"learning_rate": 0.00019396926207859084,
|
| 331 |
+
"loss": 0.8325,
|
| 332 |
"step": 220
|
| 333 |
},
|
| 334 |
{
|
| 335 |
+
"epoch": 2.0,
|
| 336 |
+
"eval_loss": 2.227036952972412,
|
| 337 |
+
"eval_runtime": 1.0053,
|
| 338 |
+
"eval_samples_per_second": 4.974,
|
| 339 |
+
"eval_steps_per_second": 1.989,
|
| 340 |
+
"step": 220
|
| 341 |
},
|
| 342 |
{
|
| 343 |
+
"epoch": 2.0454545454545454,
|
| 344 |
+
"grad_norm": 0.70703125,
|
| 345 |
+
"learning_rate": 0.00019341478602651069,
|
| 346 |
+
"loss": 0.7413,
|
| 347 |
+
"step": 225
|
|
|
|
| 348 |
},
|
| 349 |
{
|
| 350 |
+
"epoch": 2.090909090909091,
|
| 351 |
+
"grad_norm": 3.796875,
|
| 352 |
+
"learning_rate": 0.00019283679330160726,
|
| 353 |
+
"loss": 0.7526,
|
| 354 |
"step": 230
|
| 355 |
},
|
| 356 |
{
|
| 357 |
+
"epoch": 2.1363636363636362,
|
| 358 |
+
"grad_norm": 0.69921875,
|
| 359 |
+
"learning_rate": 0.00019223542941045817,
|
| 360 |
+
"loss": 0.739,
|
| 361 |
"step": 235
|
| 362 |
},
|
| 363 |
{
|
| 364 |
+
"epoch": 2.1818181818181817,
|
| 365 |
+
"grad_norm": 1.09375,
|
| 366 |
+
"learning_rate": 0.00019161084574320696,
|
| 367 |
+
"loss": 0.7665,
|
| 368 |
"step": 240
|
| 369 |
},
|
| 370 |
{
|
| 371 |
+
"epoch": 2.227272727272727,
|
| 372 |
+
"grad_norm": 0.859375,
|
| 373 |
+
"learning_rate": 0.00019096319953545185,
|
| 374 |
+
"loss": 0.7655,
|
| 375 |
"step": 245
|
| 376 |
},
|
| 377 |
{
|
| 378 |
+
"epoch": 2.2727272727272725,
|
| 379 |
+
"grad_norm": 2.53125,
|
| 380 |
+
"learning_rate": 0.00019029265382866214,
|
| 381 |
+
"loss": 0.753,
|
| 382 |
"step": 250
|
| 383 |
},
|
| 384 |
{
|
| 385 |
+
"epoch": 2.3181818181818183,
|
| 386 |
+
"grad_norm": 2.09375,
|
| 387 |
+
"learning_rate": 0.00018959937742913359,
|
| 388 |
+
"loss": 0.7557,
|
| 389 |
"step": 255
|
| 390 |
},
|
| 391 |
{
|
| 392 |
+
"epoch": 2.3636363636363638,
|
| 393 |
+
"grad_norm": 1.2109375,
|
| 394 |
+
"learning_rate": 0.00018888354486549237,
|
| 395 |
+
"loss": 0.7805,
|
| 396 |
"step": 260
|
| 397 |
},
|
| 398 |
{
|
| 399 |
+
"epoch": 2.409090909090909,
|
| 400 |
+
"grad_norm": 4.75,
|
| 401 |
+
"learning_rate": 0.00018814533634475822,
|
| 402 |
+
"loss": 0.7451,
|
| 403 |
"step": 265
|
| 404 |
},
|
| 405 |
{
|
| 406 |
+
"epoch": 2.4545454545454546,
|
| 407 |
+
"grad_norm": 1.109375,
|
| 408 |
+
"learning_rate": 0.00018738493770697852,
|
| 409 |
+
"loss": 0.7308,
|
| 410 |
"step": 270
|
| 411 |
},
|
| 412 |
{
|
| 413 |
+
"epoch": 2.5,
|
| 414 |
+
"grad_norm": 1.0703125,
|
| 415 |
+
"learning_rate": 0.00018660254037844388,
|
| 416 |
+
"loss": 0.8117,
|
| 417 |
"step": 275
|
| 418 |
},
|
| 419 |
{
|
| 420 |
+
"epoch": 2.5454545454545454,
|
| 421 |
+
"grad_norm": 1.515625,
|
| 422 |
+
"learning_rate": 0.00018579834132349772,
|
| 423 |
+
"loss": 0.7756,
|
| 424 |
"step": 280
|
| 425 |
},
|
| 426 |
{
|
| 427 |
+
"epoch": 2.590909090909091,
|
| 428 |
+
"grad_norm": 0.64453125,
|
| 429 |
+
"learning_rate": 0.00018497254299495146,
|
| 430 |
+
"loss": 0.7503,
|
| 431 |
"step": 285
|
| 432 |
},
|
| 433 |
{
|
| 434 |
+
"epoch": 2.6363636363636362,
|
| 435 |
+
"grad_norm": 1.25,
|
| 436 |
+
"learning_rate": 0.00018412535328311814,
|
| 437 |
+
"loss": 0.7886,
|
| 438 |
"step": 290
|
| 439 |
},
|
| 440 |
{
|
| 441 |
+
"epoch": 2.6818181818181817,
|
| 442 |
+
"grad_norm": 1.1328125,
|
| 443 |
+
"learning_rate": 0.00018325698546347715,
|
| 444 |
+
"loss": 0.7452,
|
| 445 |
"step": 295
|
| 446 |
},
|
| 447 |
{
|
| 448 |
+
"epoch": 2.7272727272727275,
|
| 449 |
+
"grad_norm": 4.0,
|
| 450 |
+
"learning_rate": 0.0001823676581429833,
|
| 451 |
+
"loss": 0.7579,
|
| 452 |
"step": 300
|
| 453 |
},
|
| 454 |
{
|
| 455 |
+
"epoch": 2.7727272727272725,
|
| 456 |
+
"grad_norm": 1.59375,
|
| 457 |
+
"learning_rate": 0.00018145759520503358,
|
| 458 |
+
"loss": 0.7494,
|
| 459 |
"step": 305
|
| 460 |
},
|
| 461 |
{
|
| 462 |
+
"epoch": 2.8181818181818183,
|
| 463 |
+
"grad_norm": 1.6953125,
|
| 464 |
+
"learning_rate": 0.00018052702575310588,
|
| 465 |
+
"loss": 0.7343,
|
| 466 |
"step": 310
|
| 467 |
},
|
| 468 |
{
|
| 469 |
+
"epoch": 2.8636363636363638,
|
| 470 |
+
"grad_norm": 3.28125,
|
| 471 |
+
"learning_rate": 0.00017957618405308324,
|
| 472 |
+
"loss": 0.7402,
|
| 473 |
"step": 315
|
| 474 |
},
|
| 475 |
{
|
| 476 |
+
"epoch": 2.909090909090909,
|
| 477 |
+
"grad_norm": 1.1328125,
|
| 478 |
+
"learning_rate": 0.00017860530947427875,
|
| 479 |
+
"loss": 0.7428,
|
| 480 |
"step": 320
|
| 481 |
},
|
| 482 |
{
|
| 483 |
+
"epoch": 2.9545454545454546,
|
| 484 |
+
"grad_norm": 0.828125,
|
| 485 |
+
"learning_rate": 0.0001776146464291757,
|
| 486 |
+
"loss": 0.7699,
|
| 487 |
"step": 325
|
| 488 |
},
|
| 489 |
{
|
| 490 |
+
"epoch": 3.0,
|
| 491 |
+
"grad_norm": 1.21875,
|
| 492 |
+
"learning_rate": 0.0001766044443118978,
|
| 493 |
+
"loss": 0.749,
|
| 494 |
"step": 330
|
| 495 |
},
|
| 496 |
{
|
| 497 |
+
"epoch": 3.0,
|
| 498 |
+
"eval_loss": 2.2333221435546875,
|
| 499 |
+
"eval_runtime": 1.0046,
|
| 500 |
+
"eval_samples_per_second": 4.977,
|
| 501 |
+
"eval_steps_per_second": 1.991,
|
| 502 |
+
"step": 330
|
| 503 |
},
|
| 504 |
{
|
| 505 |
+
"epoch": 3.0454545454545454,
|
| 506 |
+
"grad_norm": 1.1328125,
|
| 507 |
+
"learning_rate": 0.00017557495743542585,
|
| 508 |
+
"loss": 0.6475,
|
| 509 |
+
"step": 335
|
| 510 |
},
|
| 511 |
{
|
| 512 |
+
"epoch": 3.090909090909091,
|
| 513 |
+
"grad_norm": 0.63671875,
|
| 514 |
+
"learning_rate": 0.0001745264449675755,
|
| 515 |
+
"loss": 0.6833,
|
| 516 |
+
"step": 340
|
|
|
|
| 517 |
},
|
| 518 |
{
|
| 519 |
+
"epoch": 3.1363636363636362,
|
| 520 |
+
"grad_norm": 0.671875,
|
| 521 |
+
"learning_rate": 0.00017345917086575332,
|
| 522 |
+
"loss": 0.6776,
|
| 523 |
"step": 345
|
| 524 |
},
|
| 525 |
{
|
| 526 |
+
"epoch": 3.1818181818181817,
|
| 527 |
+
"grad_norm": 0.921875,
|
| 528 |
+
"learning_rate": 0.00017237340381050703,
|
| 529 |
+
"loss": 0.6798,
|
| 530 |
"step": 350
|
| 531 |
},
|
| 532 |
{
|
| 533 |
+
"epoch": 3.227272727272727,
|
| 534 |
+
"grad_norm": 1.5078125,
|
| 535 |
+
"learning_rate": 0.00017126941713788632,
|
| 536 |
+
"loss": 0.6606,
|
| 537 |
"step": 355
|
| 538 |
},
|
| 539 |
{
|
| 540 |
+
"epoch": 3.2727272727272725,
|
| 541 |
+
"grad_norm": 1.203125,
|
| 542 |
+
"learning_rate": 0.00017014748877063214,
|
| 543 |
+
"loss": 0.6629,
|
| 544 |
"step": 360
|
| 545 |
},
|
| 546 |
{
|
| 547 |
+
"epoch": 3.3181818181818183,
|
| 548 |
+
"grad_norm": 0.9375,
|
| 549 |
+
"learning_rate": 0.00016900790114821122,
|
| 550 |
+
"loss": 0.6871,
|
| 551 |
"step": 365
|
| 552 |
},
|
| 553 |
{
|
| 554 |
+
"epoch": 3.3636363636363638,
|
| 555 |
+
"grad_norm": 0.578125,
|
| 556 |
+
"learning_rate": 0.00016785094115571322,
|
| 557 |
+
"loss": 0.6575,
|
| 558 |
"step": 370
|
| 559 |
},
|
| 560 |
{
|
| 561 |
+
"epoch": 3.409090909090909,
|
| 562 |
+
"grad_norm": 0.9140625,
|
| 563 |
+
"learning_rate": 0.00016667690005162916,
|
| 564 |
+
"loss": 0.6674,
|
| 565 |
"step": 375
|
| 566 |
},
|
| 567 |
{
|
| 568 |
+
"epoch": 3.4545454545454546,
|
| 569 |
+
"grad_norm": 0.6171875,
|
| 570 |
+
"learning_rate": 0.00016548607339452853,
|
| 571 |
+
"loss": 0.7054,
|
| 572 |
"step": 380
|
| 573 |
},
|
| 574 |
{
|
| 575 |
+
"epoch": 3.5,
|
| 576 |
+
"grad_norm": 1.0546875,
|
| 577 |
+
"learning_rate": 0.00016427876096865394,
|
| 578 |
+
"loss": 0.6646,
|
| 579 |
"step": 385
|
| 580 |
},
|
| 581 |
{
|
| 582 |
+
"epoch": 3.5454545454545454,
|
| 583 |
+
"grad_norm": 1.453125,
|
| 584 |
+
"learning_rate": 0.00016305526670845226,
|
| 585 |
+
"loss": 0.6766,
|
| 586 |
"step": 390
|
| 587 |
},
|
| 588 |
{
|
| 589 |
+
"epoch": 3.590909090909091,
|
| 590 |
+
"grad_norm": 1.5703125,
|
| 591 |
+
"learning_rate": 0.00016181589862206052,
|
| 592 |
+
"loss": 0.6716,
|
| 593 |
"step": 395
|
| 594 |
},
|
| 595 |
{
|
| 596 |
+
"epoch": 3.6363636363636362,
|
| 597 |
+
"grad_norm": 0.9140625,
|
| 598 |
+
"learning_rate": 0.00016056096871376667,
|
| 599 |
+
"loss": 0.6976,
|
| 600 |
"step": 400
|
| 601 |
},
|
| 602 |
{
|
| 603 |
+
"epoch": 3.6818181818181817,
|
| 604 |
+
"grad_norm": 1.734375,
|
| 605 |
+
"learning_rate": 0.00015929079290546408,
|
| 606 |
+
"loss": 0.6624,
|
| 607 |
"step": 405
|
| 608 |
},
|
| 609 |
{
|
| 610 |
+
"epoch": 3.7272727272727275,
|
| 611 |
+
"grad_norm": 0.83984375,
|
| 612 |
+
"learning_rate": 0.00015800569095711982,
|
| 613 |
+
"loss": 0.6907,
|
| 614 |
"step": 410
|
| 615 |
},
|
| 616 |
{
|
| 617 |
+
"epoch": 3.7727272727272725,
|
| 618 |
+
"grad_norm": 1.078125,
|
| 619 |
+
"learning_rate": 0.00015670598638627706,
|
| 620 |
+
"loss": 0.6664,
|
| 621 |
"step": 415
|
| 622 |
},
|
| 623 |
{
|
| 624 |
+
"epoch": 3.8181818181818183,
|
| 625 |
+
"grad_norm": 0.88671875,
|
| 626 |
+
"learning_rate": 0.00015539200638661104,
|
| 627 |
+
"loss": 0.6642,
|
| 628 |
"step": 420
|
| 629 |
},
|
| 630 |
{
|
| 631 |
+
"epoch": 3.8636363636363638,
|
| 632 |
+
"grad_norm": 3.203125,
|
| 633 |
+
"learning_rate": 0.00015406408174555976,
|
| 634 |
+
"loss": 0.7091,
|
| 635 |
"step": 425
|
| 636 |
},
|
| 637 |
{
|
| 638 |
+
"epoch": 3.909090909090909,
|
| 639 |
+
"grad_norm": 1.9296875,
|
| 640 |
+
"learning_rate": 0.00015272254676105025,
|
| 641 |
+
"loss": 0.7034,
|
| 642 |
"step": 430
|
| 643 |
},
|
| 644 |
{
|
| 645 |
+
"epoch": 3.9545454545454546,
|
| 646 |
+
"grad_norm": 0.98828125,
|
| 647 |
+
"learning_rate": 0.00015136773915734066,
|
| 648 |
+
"loss": 0.6679,
|
| 649 |
"step": 435
|
| 650 |
},
|
| 651 |
{
|
| 652 |
+
"epoch": 4.0,
|
| 653 |
+
"grad_norm": 2.140625,
|
| 654 |
+
"learning_rate": 0.00015000000000000001,
|
| 655 |
+
"loss": 0.6755,
|
| 656 |
+
"step": 440
|
| 657 |
+
},
|
| 658 |
+
{
|
| 659 |
+
"epoch": 4.0,
|
| 660 |
+
"eval_loss": 2.29927659034729,
|
| 661 |
+
"eval_runtime": 1.0067,
|
| 662 |
+
"eval_samples_per_second": 4.967,
|
| 663 |
+
"eval_steps_per_second": 1.987,
|
| 664 |
"step": 440
|
| 665 |
},
|
| 666 |
{
|
| 667 |
+
"epoch": 4.045454545454546,
|
| 668 |
+
"grad_norm": 2.75,
|
| 669 |
+
"learning_rate": 0.00014861967361004687,
|
| 670 |
+
"loss": 0.6121,
|
| 671 |
"step": 445
|
| 672 |
},
|
| 673 |
{
|
| 674 |
+
"epoch": 4.090909090909091,
|
| 675 |
+
"grad_norm": 0.625,
|
| 676 |
+
"learning_rate": 0.0001472271074772683,
|
| 677 |
+
"loss": 0.598,
|
| 678 |
"step": 450
|
| 679 |
},
|
| 680 |
{
|
| 681 |
+
"epoch": 4.136363636363637,
|
| 682 |
+
"grad_norm": 0.94140625,
|
| 683 |
+
"learning_rate": 0.00014582265217274104,
|
| 684 |
+
"loss": 0.5832,
|
| 685 |
"step": 455
|
| 686 |
},
|
| 687 |
{
|
| 688 |
+
"epoch": 4.181818181818182,
|
| 689 |
+
"grad_norm": 1.1640625,
|
| 690 |
+
"learning_rate": 0.00014440666126057744,
|
| 691 |
+
"loss": 0.591,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 692 |
"step": 460
|
| 693 |
},
|
| 694 |
{
|
| 695 |
+
"epoch": 4.2272727272727275,
|
| 696 |
+
"grad_norm": 0.61328125,
|
| 697 |
+
"learning_rate": 0.00014297949120891718,
|
| 698 |
+
"loss": 0.5767,
|
| 699 |
"step": 465
|
| 700 |
},
|
| 701 |
{
|
| 702 |
+
"epoch": 4.2727272727272725,
|
| 703 |
+
"grad_norm": 0.69140625,
|
| 704 |
+
"learning_rate": 0.00014154150130018866,
|
| 705 |
+
"loss": 0.5742,
|
| 706 |
"step": 470
|
| 707 |
},
|
| 708 |
{
|
| 709 |
+
"epoch": 4.318181818181818,
|
| 710 |
+
"grad_norm": 0.6328125,
|
| 711 |
+
"learning_rate": 0.00014009305354066137,
|
| 712 |
+
"loss": 0.5806,
|
| 713 |
"step": 475
|
| 714 |
},
|
| 715 |
{
|
| 716 |
+
"epoch": 4.363636363636363,
|
| 717 |
+
"grad_norm": 0.7109375,
|
| 718 |
+
"learning_rate": 0.00013863451256931287,
|
| 719 |
+
"loss": 0.605,
|
| 720 |
"step": 480
|
| 721 |
},
|
| 722 |
{
|
| 723 |
+
"epoch": 4.409090909090909,
|
| 724 |
+
"grad_norm": 0.6796875,
|
| 725 |
+
"learning_rate": 0.00013716624556603274,
|
| 726 |
+
"loss": 0.6142,
|
| 727 |
"step": 485
|
| 728 |
},
|
| 729 |
{
|
| 730 |
+
"epoch": 4.454545454545454,
|
| 731 |
+
"grad_norm": 0.80859375,
|
| 732 |
+
"learning_rate": 0.00013568862215918717,
|
| 733 |
+
"loss": 0.586,
|
| 734 |
"step": 490
|
| 735 |
},
|
| 736 |
{
|
| 737 |
+
"epoch": 4.5,
|
| 738 |
+
"grad_norm": 1.34375,
|
| 739 |
+
"learning_rate": 0.00013420201433256689,
|
| 740 |
+
"loss": 0.6088,
|
| 741 |
"step": 495
|
| 742 |
},
|
| 743 |
{
|
| 744 |
+
"epoch": 4.545454545454545,
|
| 745 |
+
"grad_norm": 0.9765625,
|
| 746 |
+
"learning_rate": 0.00013270679633174218,
|
| 747 |
+
"loss": 0.5974,
|
| 748 |
"step": 500
|
| 749 |
},
|
| 750 |
{
|
| 751 |
+
"epoch": 4.590909090909091,
|
| 752 |
+
"grad_norm": 1.171875,
|
| 753 |
+
"learning_rate": 0.0001312033445698487,
|
| 754 |
+
"loss": 0.5808,
|
| 755 |
"step": 505
|
| 756 |
},
|
| 757 |
{
|
| 758 |
+
"epoch": 4.636363636363637,
|
| 759 |
+
"grad_norm": 1.515625,
|
| 760 |
+
"learning_rate": 0.0001296920375328275,
|
| 761 |
+
"loss": 0.6008,
|
| 762 |
"step": 510
|
| 763 |
},
|
| 764 |
{
|
| 765 |
+
"epoch": 4.681818181818182,
|
| 766 |
+
"grad_norm": 1.6640625,
|
| 767 |
+
"learning_rate": 0.00012817325568414297,
|
| 768 |
+
"loss": 0.6069,
|
| 769 |
"step": 515
|
| 770 |
},
|
| 771 |
{
|
| 772 |
+
"epoch": 4.7272727272727275,
|
| 773 |
+
"grad_norm": 0.6484375,
|
| 774 |
+
"learning_rate": 0.00012664738136900348,
|
| 775 |
+
"loss": 0.6111,
|
| 776 |
"step": 520
|
| 777 |
},
|
| 778 |
{
|
| 779 |
+
"epoch": 4.7727272727272725,
|
| 780 |
+
"grad_norm": 0.640625,
|
| 781 |
+
"learning_rate": 0.0001251147987181079,
|
| 782 |
+
"loss": 0.6114,
|
| 783 |
"step": 525
|
| 784 |
},
|
| 785 |
{
|
| 786 |
+
"epoch": 4.818181818181818,
|
| 787 |
+
"grad_norm": 0.6015625,
|
| 788 |
+
"learning_rate": 0.00012357589355094275,
|
| 789 |
+
"loss": 0.5905,
|
| 790 |
"step": 530
|
| 791 |
},
|
| 792 |
{
|
| 793 |
+
"epoch": 4.863636363636363,
|
| 794 |
+
"grad_norm": 0.6171875,
|
| 795 |
+
"learning_rate": 0.00012203105327865407,
|
| 796 |
+
"loss": 0.6153,
|
| 797 |
"step": 535
|
| 798 |
},
|
| 799 |
{
|
| 800 |
+
"epoch": 4.909090909090909,
|
| 801 |
+
"grad_norm": 0.73828125,
|
| 802 |
+
"learning_rate": 0.00012048066680651908,
|
| 803 |
+
"loss": 0.6179,
|
| 804 |
"step": 540
|
| 805 |
},
|
| 806 |
{
|
| 807 |
+
"epoch": 4.954545454545455,
|
| 808 |
+
"grad_norm": 0.62890625,
|
| 809 |
+
"learning_rate": 0.00011892512443604102,
|
| 810 |
+
"loss": 0.5833,
|
| 811 |
"step": 545
|
| 812 |
},
|
| 813 |
{
|
| 814 |
+
"epoch": 5.0,
|
| 815 |
+
"grad_norm": 0.54296875,
|
| 816 |
+
"learning_rate": 0.00011736481776669306,
|
| 817 |
+
"loss": 0.6197,
|
| 818 |
+
"step": 550
|
| 819 |
+
},
|
| 820 |
+
{
|
| 821 |
+
"epoch": 5.0,
|
| 822 |
+
"eval_loss": 2.3820290565490723,
|
| 823 |
+
"eval_runtime": 1.0056,
|
| 824 |
+
"eval_samples_per_second": 4.972,
|
| 825 |
+
"eval_steps_per_second": 1.989,
|
| 826 |
"step": 550
|
| 827 |
},
|
| 828 |
{
|
| 829 |
+
"epoch": 5.045454545454546,
|
| 830 |
+
"grad_norm": 0.71484375,
|
| 831 |
+
"learning_rate": 0.000115800139597335,
|
| 832 |
+
"loss": 0.5034,
|
| 833 |
"step": 555
|
| 834 |
},
|
| 835 |
{
|
| 836 |
+
"epoch": 5.090909090909091,
|
| 837 |
+
"grad_norm": 0.66796875,
|
| 838 |
+
"learning_rate": 0.00011423148382732853,
|
| 839 |
+
"loss": 0.5171,
|
| 840 |
"step": 560
|
| 841 |
},
|
| 842 |
{
|
| 843 |
+
"epoch": 5.136363636363637,
|
| 844 |
+
"grad_norm": 0.6875,
|
| 845 |
+
"learning_rate": 0.00011265924535737493,
|
| 846 |
+
"loss": 0.5233,
|
| 847 |
"step": 565
|
| 848 |
},
|
| 849 |
{
|
| 850 |
+
"epoch": 5.181818181818182,
|
| 851 |
+
"grad_norm": 0.61328125,
|
| 852 |
+
"learning_rate": 0.00011108381999010111,
|
| 853 |
+
"loss": 0.5103,
|
| 854 |
"step": 570
|
| 855 |
},
|
| 856 |
{
|
| 857 |
+
"epoch": 5.2272727272727275,
|
| 858 |
+
"grad_norm": 0.63671875,
|
| 859 |
+
"learning_rate": 0.00010950560433041826,
|
| 860 |
+
"loss": 0.5371,
|
| 861 |
+
"step": 575
|
| 862 |
+
},
|
| 863 |
+
{
|
| 864 |
+
"epoch": 5.2727272727272725,
|
| 865 |
+
"grad_norm": 0.6171875,
|
| 866 |
+
"learning_rate": 0.00010792499568567884,
|
| 867 |
+
"loss": 0.5228,
|
| 868 |
+
"step": 580
|
| 869 |
+
},
|
| 870 |
+
{
|
| 871 |
+
"epoch": 5.318181818181818,
|
| 872 |
+
"grad_norm": 0.7734375,
|
| 873 |
+
"learning_rate": 0.00010634239196565646,
|
| 874 |
+
"loss": 0.5308,
|
| 875 |
+
"step": 585
|
| 876 |
+
},
|
| 877 |
+
{
|
| 878 |
+
"epoch": 5.363636363636363,
|
| 879 |
+
"grad_norm": 0.6015625,
|
| 880 |
+
"learning_rate": 0.00010475819158237425,
|
| 881 |
+
"loss": 0.5349,
|
| 882 |
+
"step": 590
|
| 883 |
+
},
|
| 884 |
+
{
|
| 885 |
+
"epoch": 5.409090909090909,
|
| 886 |
+
"grad_norm": 0.59765625,
|
| 887 |
+
"learning_rate": 0.00010317279334980678,
|
| 888 |
+
"loss": 0.5204,
|
| 889 |
+
"step": 595
|
| 890 |
+
},
|
| 891 |
+
{
|
| 892 |
+
"epoch": 5.454545454545454,
|
| 893 |
+
"grad_norm": 0.70703125,
|
| 894 |
+
"learning_rate": 0.00010158659638348081,
|
| 895 |
+
"loss": 0.5182,
|
| 896 |
+
"step": 600
|
| 897 |
+
},
|
| 898 |
+
{
|
| 899 |
+
"epoch": 5.5,
|
| 900 |
+
"grad_norm": 0.58984375,
|
| 901 |
+
"learning_rate": 0.0001,
|
| 902 |
+
"loss": 0.5215,
|
| 903 |
+
"step": 605
|
| 904 |
+
},
|
| 905 |
+
{
|
| 906 |
+
"epoch": 5.545454545454545,
|
| 907 |
+
"grad_norm": 0.6484375,
|
| 908 |
+
"learning_rate": 9.84134036165192e-05,
|
| 909 |
+
"loss": 0.5282,
|
| 910 |
+
"step": 610
|
| 911 |
+
},
|
| 912 |
+
{
|
| 913 |
+
"epoch": 5.590909090909091,
|
| 914 |
+
"grad_norm": 0.6015625,
|
| 915 |
+
"learning_rate": 9.682720665019325e-05,
|
| 916 |
+
"loss": 0.5103,
|
| 917 |
+
"step": 615
|
| 918 |
+
},
|
| 919 |
+
{
|
| 920 |
+
"epoch": 5.636363636363637,
|
| 921 |
+
"grad_norm": 0.63671875,
|
| 922 |
+
"learning_rate": 9.524180841762577e-05,
|
| 923 |
+
"loss": 0.5257,
|
| 924 |
+
"step": 620
|
| 925 |
+
},
|
| 926 |
+
{
|
| 927 |
+
"epoch": 5.681818181818182,
|
| 928 |
+
"grad_norm": 0.6640625,
|
| 929 |
+
"learning_rate": 9.365760803434355e-05,
|
| 930 |
+
"loss": 0.5228,
|
| 931 |
+
"step": 625
|
| 932 |
+
},
|
| 933 |
+
{
|
| 934 |
+
"epoch": 5.7272727272727275,
|
| 935 |
+
"grad_norm": 0.60546875,
|
| 936 |
+
"learning_rate": 9.207500431432115e-05,
|
| 937 |
+
"loss": 0.5487,
|
| 938 |
+
"step": 630
|
| 939 |
+
},
|
| 940 |
+
{
|
| 941 |
+
"epoch": 5.7727272727272725,
|
| 942 |
+
"grad_norm": 0.7734375,
|
| 943 |
+
"learning_rate": 9.049439566958175e-05,
|
| 944 |
+
"loss": 0.528,
|
| 945 |
+
"step": 635
|
| 946 |
+
},
|
| 947 |
+
{
|
| 948 |
+
"epoch": 5.818181818181818,
|
| 949 |
+
"grad_norm": 0.79296875,
|
| 950 |
+
"learning_rate": 8.891618000989891e-05,
|
| 951 |
+
"loss": 0.5213,
|
| 952 |
+
"step": 640
|
| 953 |
+
},
|
| 954 |
+
{
|
| 955 |
+
"epoch": 5.863636363636363,
|
| 956 |
+
"grad_norm": 0.640625,
|
| 957 |
+
"learning_rate": 8.734075464262507e-05,
|
| 958 |
+
"loss": 0.5312,
|
| 959 |
+
"step": 645
|
| 960 |
+
},
|
| 961 |
+
{
|
| 962 |
+
"epoch": 5.909090909090909,
|
| 963 |
+
"grad_norm": 0.62890625,
|
| 964 |
+
"learning_rate": 8.57685161726715e-05,
|
| 965 |
+
"loss": 0.5208,
|
| 966 |
+
"step": 650
|
| 967 |
+
},
|
| 968 |
+
{
|
| 969 |
+
"epoch": 5.954545454545455,
|
| 970 |
+
"grad_norm": 0.6953125,
|
| 971 |
+
"learning_rate": 8.4199860402665e-05,
|
| 972 |
+
"loss": 0.5372,
|
| 973 |
+
"step": 655
|
| 974 |
+
},
|
| 975 |
+
{
|
| 976 |
+
"epoch": 6.0,
|
| 977 |
+
"grad_norm": 0.640625,
|
| 978 |
+
"learning_rate": 8.263518223330697e-05,
|
| 979 |
+
"loss": 0.5208,
|
| 980 |
+
"step": 660
|
| 981 |
+
},
|
| 982 |
+
{
|
| 983 |
+
"epoch": 6.0,
|
| 984 |
+
"eval_loss": 2.586942672729492,
|
| 985 |
+
"eval_runtime": 1.0065,
|
| 986 |
+
"eval_samples_per_second": 4.968,
|
| 987 |
+
"eval_steps_per_second": 1.987,
|
| 988 |
+
"step": 660
|
| 989 |
+
},
|
| 990 |
+
{
|
| 991 |
+
"epoch": 6.045454545454546,
|
| 992 |
+
"grad_norm": 0.7734375,
|
| 993 |
+
"learning_rate": 8.107487556395901e-05,
|
| 994 |
+
"loss": 0.46,
|
| 995 |
+
"step": 665
|
| 996 |
+
},
|
| 997 |
+
{
|
| 998 |
+
"epoch": 6.090909090909091,
|
| 999 |
+
"grad_norm": 0.703125,
|
| 1000 |
+
"learning_rate": 7.951933319348095e-05,
|
| 1001 |
+
"loss": 0.4614,
|
| 1002 |
+
"step": 670
|
| 1003 |
+
},
|
| 1004 |
+
{
|
| 1005 |
+
"epoch": 6.136363636363637,
|
| 1006 |
+
"grad_norm": 0.6640625,
|
| 1007 |
+
"learning_rate": 7.796894672134594e-05,
|
| 1008 |
+
"loss": 0.4648,
|
| 1009 |
+
"step": 675
|
| 1010 |
+
},
|
| 1011 |
+
{
|
| 1012 |
+
"epoch": 6.181818181818182,
|
| 1013 |
+
"grad_norm": 0.76171875,
|
| 1014 |
+
"learning_rate": 7.642410644905726e-05,
|
| 1015 |
+
"loss": 0.4445,
|
| 1016 |
+
"step": 680
|
| 1017 |
+
},
|
| 1018 |
+
{
|
| 1019 |
+
"epoch": 6.2272727272727275,
|
| 1020 |
+
"grad_norm": 0.67578125,
|
| 1021 |
+
"learning_rate": 7.488520128189209e-05,
|
| 1022 |
+
"loss": 0.4483,
|
| 1023 |
+
"step": 685
|
| 1024 |
+
},
|
| 1025 |
+
{
|
| 1026 |
+
"epoch": 6.2727272727272725,
|
| 1027 |
+
"grad_norm": 0.6875,
|
| 1028 |
+
"learning_rate": 7.335261863099651e-05,
|
| 1029 |
+
"loss": 0.4484,
|
| 1030 |
+
"step": 690
|
| 1031 |
+
},
|
| 1032 |
+
{
|
| 1033 |
+
"epoch": 6.318181818181818,
|
| 1034 |
+
"grad_norm": 0.65234375,
|
| 1035 |
+
"learning_rate": 7.182674431585704e-05,
|
| 1036 |
+
"loss": 0.4452,
|
| 1037 |
+
"step": 695
|
| 1038 |
+
},
|
| 1039 |
+
{
|
| 1040 |
+
"epoch": 6.363636363636363,
|
| 1041 |
+
"grad_norm": 0.69140625,
|
| 1042 |
+
"learning_rate": 7.030796246717255e-05,
|
| 1043 |
+
"loss": 0.4615,
|
| 1044 |
+
"step": 700
|
| 1045 |
+
},
|
| 1046 |
+
{
|
| 1047 |
+
"epoch": 6.409090909090909,
|
| 1048 |
+
"grad_norm": 0.66796875,
|
| 1049 |
+
"learning_rate": 6.87966554301513e-05,
|
| 1050 |
+
"loss": 0.4443,
|
| 1051 |
+
"step": 705
|
| 1052 |
+
},
|
| 1053 |
+
{
|
| 1054 |
+
"epoch": 6.454545454545454,
|
| 1055 |
+
"grad_norm": 0.65625,
|
| 1056 |
+
"learning_rate": 6.729320366825784e-05,
|
| 1057 |
+
"loss": 0.4509,
|
| 1058 |
+
"step": 710
|
| 1059 |
+
},
|
| 1060 |
+
{
|
| 1061 |
+
"epoch": 6.5,
|
| 1062 |
+
"grad_norm": 0.734375,
|
| 1063 |
+
"learning_rate": 6.579798566743314e-05,
|
| 1064 |
+
"loss": 0.461,
|
| 1065 |
+
"step": 715
|
| 1066 |
+
},
|
| 1067 |
+
{
|
| 1068 |
+
"epoch": 6.545454545454545,
|
| 1069 |
+
"grad_norm": 0.66015625,
|
| 1070 |
+
"learning_rate": 6.431137784081282e-05,
|
| 1071 |
+
"loss": 0.4571,
|
| 1072 |
+
"step": 720
|
| 1073 |
+
},
|
| 1074 |
+
{
|
| 1075 |
+
"epoch": 6.590909090909091,
|
| 1076 |
+
"grad_norm": 0.63671875,
|
| 1077 |
+
"learning_rate": 6.283375443396726e-05,
|
| 1078 |
+
"loss": 0.4698,
|
| 1079 |
+
"step": 725
|
| 1080 |
+
},
|
| 1081 |
+
{
|
| 1082 |
+
"epoch": 6.636363636363637,
|
| 1083 |
+
"grad_norm": 0.62890625,
|
| 1084 |
+
"learning_rate": 6.136548743068713e-05,
|
| 1085 |
+
"loss": 0.441,
|
| 1086 |
+
"step": 730
|
| 1087 |
+
},
|
| 1088 |
+
{
|
| 1089 |
+
"epoch": 6.681818181818182,
|
| 1090 |
+
"grad_norm": 0.65625,
|
| 1091 |
+
"learning_rate": 5.9906946459338656e-05,
|
| 1092 |
+
"loss": 0.4464,
|
| 1093 |
+
"step": 735
|
| 1094 |
+
},
|
| 1095 |
+
{
|
| 1096 |
+
"epoch": 6.7272727272727275,
|
| 1097 |
+
"grad_norm": 0.71875,
|
| 1098 |
+
"learning_rate": 5.845849869981137e-05,
|
| 1099 |
+
"loss": 0.4455,
|
| 1100 |
+
"step": 740
|
| 1101 |
+
},
|
| 1102 |
+
{
|
| 1103 |
+
"epoch": 6.7727272727272725,
|
| 1104 |
+
"grad_norm": 0.8359375,
|
| 1105 |
+
"learning_rate": 5.702050879108284e-05,
|
| 1106 |
+
"loss": 0.4541,
|
| 1107 |
+
"step": 745
|
| 1108 |
+
},
|
| 1109 |
+
{
|
| 1110 |
+
"epoch": 6.818181818181818,
|
| 1111 |
+
"grad_norm": 0.69921875,
|
| 1112 |
+
"learning_rate": 5.559333873942259e-05,
|
| 1113 |
+
"loss": 0.4546,
|
| 1114 |
+
"step": 750
|
| 1115 |
+
},
|
| 1116 |
+
{
|
| 1117 |
+
"epoch": 6.863636363636363,
|
| 1118 |
+
"grad_norm": 0.66015625,
|
| 1119 |
+
"learning_rate": 5.417734782725896e-05,
|
| 1120 |
+
"loss": 0.4427,
|
| 1121 |
+
"step": 755
|
| 1122 |
+
},
|
| 1123 |
+
{
|
| 1124 |
+
"epoch": 6.909090909090909,
|
| 1125 |
+
"grad_norm": 0.703125,
|
| 1126 |
+
"learning_rate": 5.277289252273174e-05,
|
| 1127 |
+
"loss": 0.4497,
|
| 1128 |
+
"step": 760
|
| 1129 |
+
},
|
| 1130 |
+
{
|
| 1131 |
+
"epoch": 6.954545454545455,
|
| 1132 |
+
"grad_norm": 0.66796875,
|
| 1133 |
+
"learning_rate": 5.138032638995315e-05,
|
| 1134 |
+
"loss": 0.4519,
|
| 1135 |
+
"step": 765
|
| 1136 |
+
},
|
| 1137 |
+
{
|
| 1138 |
+
"epoch": 7.0,
|
| 1139 |
+
"grad_norm": 0.6953125,
|
| 1140 |
+
"learning_rate": 5.000000000000002e-05,
|
| 1141 |
+
"loss": 0.4474,
|
| 1142 |
+
"step": 770
|
| 1143 |
+
},
|
| 1144 |
+
{
|
| 1145 |
+
"epoch": 7.0,
|
| 1146 |
+
"eval_loss": 2.838920831680298,
|
| 1147 |
+
"eval_runtime": 1.0059,
|
| 1148 |
+
"eval_samples_per_second": 4.971,
|
| 1149 |
+
"eval_steps_per_second": 1.988,
|
| 1150 |
+
"step": 770
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 7.045454545454546,
|
| 1154 |
+
"grad_norm": 0.66015625,
|
| 1155 |
+
"learning_rate": 4.8632260842659393e-05,
|
| 1156 |
+
"loss": 0.3915,
|
| 1157 |
+
"step": 775
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 7.090909090909091,
|
| 1161 |
+
"grad_norm": 0.69140625,
|
| 1162 |
+
"learning_rate": 4.727745323894976e-05,
|
| 1163 |
+
"loss": 0.3938,
|
| 1164 |
+
"step": 780
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 7.136363636363637,
|
| 1168 |
+
"grad_norm": 0.81640625,
|
| 1169 |
+
"learning_rate": 4.593591825444028e-05,
|
| 1170 |
+
"loss": 0.3943,
|
| 1171 |
+
"step": 785
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 7.181818181818182,
|
| 1175 |
+
"grad_norm": 0.8515625,
|
| 1176 |
+
"learning_rate": 4.4607993613388976e-05,
|
| 1177 |
+
"loss": 0.3993,
|
| 1178 |
+
"step": 790
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 7.2272727272727275,
|
| 1182 |
+
"grad_norm": 0.7109375,
|
| 1183 |
+
"learning_rate": 4.329401361372294e-05,
|
| 1184 |
+
"loss": 0.397,
|
| 1185 |
+
"step": 795
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 7.2727272727272725,
|
| 1189 |
+
"grad_norm": 0.609375,
|
| 1190 |
+
"learning_rate": 4.19943090428802e-05,
|
| 1191 |
+
"loss": 0.3952,
|
| 1192 |
+
"step": 800
|
| 1193 |
+
},
|
| 1194 |
+
{
|
| 1195 |
+
"epoch": 7.318181818181818,
|
| 1196 |
+
"grad_norm": 0.703125,
|
| 1197 |
+
"learning_rate": 4.070920709453597e-05,
|
| 1198 |
+
"loss": 0.3992,
|
| 1199 |
+
"step": 805
|
| 1200 |
+
},
|
| 1201 |
+
{
|
| 1202 |
+
"epoch": 7.363636363636363,
|
| 1203 |
+
"grad_norm": 0.6953125,
|
| 1204 |
+
"learning_rate": 3.943903128623335e-05,
|
| 1205 |
+
"loss": 0.3869,
|
| 1206 |
+
"step": 810
|
| 1207 |
+
},
|
| 1208 |
+
{
|
| 1209 |
+
"epoch": 7.409090909090909,
|
| 1210 |
+
"grad_norm": 0.65234375,
|
| 1211 |
+
"learning_rate": 3.8184101377939476e-05,
|
| 1212 |
+
"loss": 0.3931,
|
| 1213 |
+
"step": 815
|
| 1214 |
+
},
|
| 1215 |
+
{
|
| 1216 |
+
"epoch": 7.454545454545454,
|
| 1217 |
+
"grad_norm": 0.65625,
|
| 1218 |
+
"learning_rate": 3.694473329154778e-05,
|
| 1219 |
+
"loss": 0.398,
|
| 1220 |
+
"step": 820
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 7.5,
|
| 1224 |
+
"grad_norm": 0.66015625,
|
| 1225 |
+
"learning_rate": 3.5721239031346066e-05,
|
| 1226 |
+
"loss": 0.4006,
|
| 1227 |
+
"step": 825
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"epoch": 7.545454545454545,
|
| 1231 |
+
"grad_norm": 0.6640625,
|
| 1232 |
+
"learning_rate": 3.45139266054715e-05,
|
| 1233 |
+
"loss": 0.3935,
|
| 1234 |
+
"step": 830
|
| 1235 |
+
},
|
| 1236 |
+
{
|
| 1237 |
+
"epoch": 7.590909090909091,
|
| 1238 |
+
"grad_norm": 0.74609375,
|
| 1239 |
+
"learning_rate": 3.332309994837085e-05,
|
| 1240 |
+
"loss": 0.3984,
|
| 1241 |
+
"step": 835
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 7.636363636363637,
|
| 1245 |
+
"grad_norm": 0.6953125,
|
| 1246 |
+
"learning_rate": 3.21490588442868e-05,
|
| 1247 |
+
"loss": 0.3981,
|
| 1248 |
+
"step": 840
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 7.681818181818182,
|
| 1252 |
+
"grad_norm": 0.640625,
|
| 1253 |
+
"learning_rate": 3.099209885178882e-05,
|
| 1254 |
+
"loss": 0.3981,
|
| 1255 |
+
"step": 845
|
| 1256 |
+
},
|
| 1257 |
+
{
|
| 1258 |
+
"epoch": 7.7272727272727275,
|
| 1259 |
+
"grad_norm": 0.65234375,
|
| 1260 |
+
"learning_rate": 2.9852511229367865e-05,
|
| 1261 |
+
"loss": 0.3978,
|
| 1262 |
+
"step": 850
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 7.7727272727272725,
|
| 1266 |
+
"grad_norm": 0.6796875,
|
| 1267 |
+
"learning_rate": 2.8730582862113742e-05,
|
| 1268 |
+
"loss": 0.3943,
|
| 1269 |
+
"step": 855
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 7.818181818181818,
|
| 1273 |
+
"grad_norm": 0.6484375,
|
| 1274 |
+
"learning_rate": 2.7626596189492983e-05,
|
| 1275 |
+
"loss": 0.3913,
|
| 1276 |
+
"step": 860
|
| 1277 |
+
},
|
| 1278 |
+
{
|
| 1279 |
+
"epoch": 7.863636363636363,
|
| 1280 |
+
"grad_norm": 0.73046875,
|
| 1281 |
+
"learning_rate": 2.654082913424668e-05,
|
| 1282 |
+
"loss": 0.4045,
|
| 1283 |
+
"step": 865
|
| 1284 |
+
},
|
| 1285 |
+
{
|
| 1286 |
+
"epoch": 7.909090909090909,
|
| 1287 |
+
"grad_norm": 0.6796875,
|
| 1288 |
+
"learning_rate": 2.5473555032424533e-05,
|
| 1289 |
+
"loss": 0.3828,
|
| 1290 |
+
"step": 870
|
| 1291 |
+
},
|
| 1292 |
+
{
|
| 1293 |
+
"epoch": 7.954545454545455,
|
| 1294 |
+
"grad_norm": 0.6953125,
|
| 1295 |
+
"learning_rate": 2.4425042564574184e-05,
|
| 1296 |
+
"loss": 0.388,
|
| 1297 |
+
"step": 875
|
| 1298 |
+
},
|
| 1299 |
+
{
|
| 1300 |
+
"epoch": 8.0,
|
| 1301 |
+
"grad_norm": 0.7734375,
|
| 1302 |
+
"learning_rate": 2.339555568810221e-05,
|
| 1303 |
+
"loss": 0.4044,
|
| 1304 |
+
"step": 880
|
| 1305 |
+
},
|
| 1306 |
+
{
|
| 1307 |
+
"epoch": 8.0,
|
| 1308 |
+
"eval_loss": 3.102931499481201,
|
| 1309 |
+
"eval_runtime": 1.0057,
|
| 1310 |
+
"eval_samples_per_second": 4.972,
|
| 1311 |
+
"eval_steps_per_second": 1.989,
|
| 1312 |
+
"step": 880
|
| 1313 |
+
},
|
| 1314 |
+
{
|
| 1315 |
+
"epoch": 8.045454545454545,
|
| 1316 |
+
"grad_norm": 0.59375,
|
| 1317 |
+
"learning_rate": 2.2385353570824308e-05,
|
| 1318 |
+
"loss": 0.3693,
|
| 1319 |
+
"step": 885
|
| 1320 |
+
},
|
| 1321 |
+
{
|
| 1322 |
+
"epoch": 8.090909090909092,
|
| 1323 |
+
"grad_norm": 0.65625,
|
| 1324 |
+
"learning_rate": 2.139469052572127e-05,
|
| 1325 |
+
"loss": 0.3641,
|
| 1326 |
+
"step": 890
|
| 1327 |
+
},
|
| 1328 |
+
{
|
| 1329 |
+
"epoch": 8.136363636363637,
|
| 1330 |
+
"grad_norm": 0.65234375,
|
| 1331 |
+
"learning_rate": 2.042381594691678e-05,
|
| 1332 |
+
"loss": 0.3647,
|
| 1333 |
+
"step": 895
|
| 1334 |
+
},
|
| 1335 |
+
{
|
| 1336 |
+
"epoch": 8.181818181818182,
|
| 1337 |
+
"grad_norm": 0.671875,
|
| 1338 |
+
"learning_rate": 1.947297424689414e-05,
|
| 1339 |
+
"loss": 0.3619,
|
| 1340 |
+
"step": 900
|
| 1341 |
+
},
|
| 1342 |
+
{
|
| 1343 |
+
"epoch": 8.227272727272727,
|
| 1344 |
+
"grad_norm": 0.64453125,
|
| 1345 |
+
"learning_rate": 1.854240479496643e-05,
|
| 1346 |
+
"loss": 0.3566,
|
| 1347 |
+
"step": 905
|
| 1348 |
+
},
|
| 1349 |
+
{
|
| 1350 |
+
"epoch": 8.272727272727273,
|
| 1351 |
+
"grad_norm": 0.6640625,
|
| 1352 |
+
"learning_rate": 1.763234185701673e-05,
|
| 1353 |
+
"loss": 0.354,
|
| 1354 |
+
"step": 910
|
| 1355 |
+
},
|
| 1356 |
+
{
|
| 1357 |
+
"epoch": 8.318181818181818,
|
| 1358 |
+
"grad_norm": 0.6953125,
|
| 1359 |
+
"learning_rate": 1.6743014536522873e-05,
|
| 1360 |
+
"loss": 0.3703,
|
| 1361 |
+
"step": 915
|
| 1362 |
+
},
|
| 1363 |
+
{
|
| 1364 |
+
"epoch": 8.363636363636363,
|
| 1365 |
+
"grad_norm": 0.75,
|
| 1366 |
+
"learning_rate": 1.587464671688187e-05,
|
| 1367 |
+
"loss": 0.3597,
|
| 1368 |
+
"step": 920
|
| 1369 |
+
},
|
| 1370 |
+
{
|
| 1371 |
+
"epoch": 8.409090909090908,
|
| 1372 |
+
"grad_norm": 0.6875,
|
| 1373 |
+
"learning_rate": 1.5027457005048573e-05,
|
| 1374 |
+
"loss": 0.3705,
|
| 1375 |
+
"step": 925
|
| 1376 |
+
},
|
| 1377 |
+
{
|
| 1378 |
+
"epoch": 8.454545454545455,
|
| 1379 |
+
"grad_norm": 0.70703125,
|
| 1380 |
+
"learning_rate": 1.4201658676502294e-05,
|
| 1381 |
+
"loss": 0.367,
|
| 1382 |
+
"step": 930
|
| 1383 |
+
},
|
| 1384 |
+
{
|
| 1385 |
+
"epoch": 8.5,
|
| 1386 |
+
"grad_norm": 0.640625,
|
| 1387 |
+
"learning_rate": 1.339745962155613e-05,
|
| 1388 |
+
"loss": 0.3607,
|
| 1389 |
+
"step": 935
|
| 1390 |
+
},
|
| 1391 |
+
{
|
| 1392 |
+
"epoch": 8.545454545454545,
|
| 1393 |
+
"grad_norm": 0.6484375,
|
| 1394 |
+
"learning_rate": 1.2615062293021507e-05,
|
| 1395 |
+
"loss": 0.3633,
|
| 1396 |
+
"step": 940
|
| 1397 |
+
},
|
| 1398 |
+
{
|
| 1399 |
+
"epoch": 8.590909090909092,
|
| 1400 |
+
"grad_norm": 0.66015625,
|
| 1401 |
+
"learning_rate": 1.1854663655241805e-05,
|
| 1402 |
+
"loss": 0.3622,
|
| 1403 |
+
"step": 945
|
| 1404 |
+
},
|
| 1405 |
+
{
|
| 1406 |
+
"epoch": 8.636363636363637,
|
| 1407 |
+
"grad_norm": 0.65625,
|
| 1408 |
+
"learning_rate": 1.1116455134507664e-05,
|
| 1409 |
+
"loss": 0.3734,
|
| 1410 |
+
"step": 950
|
| 1411 |
+
},
|
| 1412 |
+
{
|
| 1413 |
+
"epoch": 8.681818181818182,
|
| 1414 |
+
"grad_norm": 0.66015625,
|
| 1415 |
+
"learning_rate": 1.0400622570866425e-05,
|
| 1416 |
+
"loss": 0.367,
|
| 1417 |
+
"step": 955
|
| 1418 |
+
},
|
| 1419 |
+
{
|
| 1420 |
+
"epoch": 8.727272727272727,
|
| 1421 |
+
"grad_norm": 0.63671875,
|
| 1422 |
+
"learning_rate": 9.707346171337894e-06,
|
| 1423 |
+
"loss": 0.3578,
|
| 1424 |
+
"step": 960
|
| 1425 |
+
},
|
| 1426 |
+
{
|
| 1427 |
+
"epoch": 8.772727272727273,
|
| 1428 |
+
"grad_norm": 0.6640625,
|
| 1429 |
+
"learning_rate": 9.036800464548157e-06,
|
| 1430 |
+
"loss": 0.364,
|
| 1431 |
+
"step": 965
|
| 1432 |
+
},
|
| 1433 |
+
{
|
| 1434 |
+
"epoch": 8.818181818181818,
|
| 1435 |
+
"grad_norm": 0.61328125,
|
| 1436 |
+
"learning_rate": 8.38915425679304e-06,
|
| 1437 |
+
"loss": 0.3558,
|
| 1438 |
+
"step": 970
|
| 1439 |
+
},
|
| 1440 |
+
{
|
| 1441 |
+
"epoch": 8.863636363636363,
|
| 1442 |
+
"grad_norm": 0.65625,
|
| 1443 |
+
"learning_rate": 7.764570589541875e-06,
|
| 1444 |
+
"loss": 0.3664,
|
| 1445 |
+
"step": 975
|
| 1446 |
+
},
|
| 1447 |
+
{
|
| 1448 |
+
"epoch": 8.909090909090908,
|
| 1449 |
+
"grad_norm": 0.61328125,
|
| 1450 |
+
"learning_rate": 7.163206698392744e-06,
|
| 1451 |
+
"loss": 0.3573,
|
| 1452 |
+
"step": 980
|
| 1453 |
+
},
|
| 1454 |
+
{
|
| 1455 |
+
"epoch": 8.954545454545455,
|
| 1456 |
+
"grad_norm": 0.6484375,
|
| 1457 |
+
"learning_rate": 6.585213973489335e-06,
|
| 1458 |
+
"loss": 0.3717,
|
| 1459 |
+
"step": 985
|
| 1460 |
+
},
|
| 1461 |
+
{
|
| 1462 |
+
"epoch": 9.0,
|
| 1463 |
+
"grad_norm": 0.640625,
|
| 1464 |
+
"learning_rate": 6.030737921409169e-06,
|
| 1465 |
+
"loss": 0.3573,
|
| 1466 |
+
"step": 990
|
| 1467 |
+
},
|
| 1468 |
+
{
|
| 1469 |
+
"epoch": 9.0,
|
| 1470 |
+
"eval_loss": 3.357294797897339,
|
| 1471 |
+
"eval_runtime": 1.005,
|
| 1472 |
+
"eval_samples_per_second": 4.975,
|
| 1473 |
+
"eval_steps_per_second": 1.99,
|
| 1474 |
+
"step": 990
|
| 1475 |
+
},
|
| 1476 |
+
{
|
| 1477 |
+
"epoch": 9.045454545454545,
|
| 1478 |
+
"grad_norm": 0.58984375,
|
| 1479 |
+
"learning_rate": 5.499918128533155e-06,
|
| 1480 |
+
"loss": 0.3595,
|
| 1481 |
+
"step": 995
|
| 1482 |
+
},
|
| 1483 |
+
{
|
| 1484 |
+
"epoch": 9.090909090909092,
|
| 1485 |
+
"grad_norm": 0.5859375,
|
| 1486 |
+
"learning_rate": 4.992888225905468e-06,
|
| 1487 |
+
"loss": 0.359,
|
| 1488 |
+
"step": 1000
|
| 1489 |
+
},
|
| 1490 |
+
{
|
| 1491 |
+
"epoch": 9.136363636363637,
|
| 1492 |
+
"grad_norm": 0.6328125,
|
| 1493 |
+
"learning_rate": 4.509775855592613e-06,
|
| 1494 |
+
"loss": 0.3584,
|
| 1495 |
+
"step": 1005
|
| 1496 |
+
},
|
| 1497 |
+
{
|
| 1498 |
+
"epoch": 9.181818181818182,
|
| 1499 |
+
"grad_norm": 0.6328125,
|
| 1500 |
+
"learning_rate": 4.050702638550275e-06,
|
| 1501 |
+
"loss": 0.3575,
|
| 1502 |
+
"step": 1010
|
| 1503 |
+
},
|
| 1504 |
+
{
|
| 1505 |
+
"epoch": 9.227272727272727,
|
| 1506 |
+
"grad_norm": 0.640625,
|
| 1507 |
+
"learning_rate": 3.615784144005796e-06,
|
| 1508 |
+
"loss": 0.3449,
|
| 1509 |
+
"step": 1015
|
| 1510 |
+
},
|
| 1511 |
+
{
|
| 1512 |
+
"epoch": 9.272727272727273,
|
| 1513 |
+
"grad_norm": 0.61328125,
|
| 1514 |
+
"learning_rate": 3.2051298603643753e-06,
|
| 1515 |
+
"loss": 0.354,
|
| 1516 |
+
"step": 1020
|
| 1517 |
+
},
|
| 1518 |
+
{
|
| 1519 |
+
"epoch": 9.318181818181818,
|
| 1520 |
+
"grad_norm": 0.64453125,
|
| 1521 |
+
"learning_rate": 2.818843167645835e-06,
|
| 1522 |
+
"loss": 0.3463,
|
| 1523 |
+
"step": 1025
|
| 1524 |
+
},
|
| 1525 |
+
{
|
| 1526 |
+
"epoch": 9.363636363636363,
|
| 1527 |
+
"grad_norm": 0.65625,
|
| 1528 |
+
"learning_rate": 2.4570213114592954e-06,
|
| 1529 |
+
"loss": 0.351,
|
| 1530 |
+
"step": 1030
|
| 1531 |
+
},
|
| 1532 |
+
{
|
| 1533 |
+
"epoch": 9.409090909090908,
|
| 1534 |
+
"grad_norm": 0.6640625,
|
| 1535 |
+
"learning_rate": 2.119755378522137e-06,
|
| 1536 |
+
"loss": 0.3594,
|
| 1537 |
+
"step": 1035
|
| 1538 |
+
},
|
| 1539 |
+
{
|
| 1540 |
+
"epoch": 9.454545454545455,
|
| 1541 |
+
"grad_norm": 0.60546875,
|
| 1542 |
+
"learning_rate": 1.8071302737293295e-06,
|
| 1543 |
+
"loss": 0.3541,
|
| 1544 |
+
"step": 1040
|
| 1545 |
+
},
|
| 1546 |
+
{
|
| 1547 |
+
"epoch": 9.5,
|
| 1548 |
+
"grad_norm": 0.66796875,
|
| 1549 |
+
"learning_rate": 1.5192246987791981e-06,
|
| 1550 |
+
"loss": 0.36,
|
| 1551 |
+
"step": 1045
|
| 1552 |
+
},
|
| 1553 |
+
{
|
| 1554 |
+
"epoch": 9.545454545454545,
|
| 1555 |
+
"grad_norm": 0.609375,
|
| 1556 |
+
"learning_rate": 1.2561111323605712e-06,
|
| 1557 |
+
"loss": 0.3573,
|
| 1558 |
+
"step": 1050
|
| 1559 |
+
},
|
| 1560 |
+
{
|
| 1561 |
+
"epoch": 9.590909090909092,
|
| 1562 |
+
"grad_norm": 0.62109375,
|
| 1563 |
+
"learning_rate": 1.0178558119067315e-06,
|
| 1564 |
+
"loss": 0.3601,
|
| 1565 |
+
"step": 1055
|
| 1566 |
+
},
|
| 1567 |
+
{
|
| 1568 |
+
"epoch": 9.636363636363637,
|
| 1569 |
+
"grad_norm": 0.6171875,
|
| 1570 |
+
"learning_rate": 8.04518716920466e-07,
|
| 1571 |
+
"loss": 0.3574,
|
| 1572 |
+
"step": 1060
|
| 1573 |
+
},
|
| 1574 |
+
{
|
| 1575 |
+
"epoch": 9.681818181818182,
|
| 1576 |
+
"grad_norm": 0.64453125,
|
| 1577 |
+
"learning_rate": 6.161535538745878e-07,
|
| 1578 |
+
"loss": 0.3664,
|
| 1579 |
+
"step": 1065
|
| 1580 |
+
},
|
| 1581 |
+
{
|
| 1582 |
+
"epoch": 9.727272727272727,
|
| 1583 |
+
"grad_norm": 0.578125,
|
| 1584 |
+
"learning_rate": 4.5280774269154115e-07,
|
| 1585 |
+
"loss": 0.3542,
|
| 1586 |
+
"step": 1070
|
| 1587 |
+
},
|
| 1588 |
+
{
|
| 1589 |
+
"epoch": 9.772727272727273,
|
| 1590 |
+
"grad_norm": 0.62109375,
|
| 1591 |
+
"learning_rate": 3.145224048057727e-07,
|
| 1592 |
+
"loss": 0.36,
|
| 1593 |
+
"step": 1075
|
| 1594 |
+
},
|
| 1595 |
+
{
|
| 1596 |
+
"epoch": 9.818181818181818,
|
| 1597 |
+
"grad_norm": 0.63671875,
|
| 1598 |
+
"learning_rate": 2.0133235281156736e-07,
|
| 1599 |
+
"loss": 0.3565,
|
| 1600 |
+
"step": 1080
|
| 1601 |
+
},
|
| 1602 |
+
{
|
| 1603 |
+
"epoch": 9.863636363636363,
|
| 1604 |
+
"grad_norm": 0.62109375,
|
| 1605 |
+
"learning_rate": 1.1326608169920372e-07,
|
| 1606 |
+
"loss": 0.3694,
|
| 1607 |
+
"step": 1085
|
| 1608 |
+
},
|
| 1609 |
+
{
|
| 1610 |
+
"epoch": 9.909090909090908,
|
| 1611 |
+
"grad_norm": 0.62109375,
|
| 1612 |
+
"learning_rate": 5.0345761681491746e-08,
|
| 1613 |
+
"loss": 0.354,
|
| 1614 |
+
"step": 1090
|
| 1615 |
+
},
|
| 1616 |
+
{
|
| 1617 |
+
"epoch": 9.954545454545455,
|
| 1618 |
+
"grad_norm": 0.59765625,
|
| 1619 |
+
"learning_rate": 1.2587232612493172e-08,
|
| 1620 |
+
"loss": 0.3516,
|
| 1621 |
+
"step": 1095
|
| 1622 |
+
},
|
| 1623 |
+
{
|
| 1624 |
+
"epoch": 10.0,
|
| 1625 |
+
"grad_norm": 0.5859375,
|
| 1626 |
+
"learning_rate": 0.0,
|
| 1627 |
+
"loss": 0.354,
|
| 1628 |
+
"step": 1100
|
| 1629 |
+
},
|
| 1630 |
+
{
|
| 1631 |
+
"epoch": 10.0,
|
| 1632 |
+
"eval_loss": 3.380370616912842,
|
| 1633 |
+
"eval_runtime": 1.0228,
|
| 1634 |
+
"eval_samples_per_second": 4.888,
|
| 1635 |
+
"eval_steps_per_second": 1.955,
|
| 1636 |
+
"step": 1100
|
| 1637 |
},
|
| 1638 |
{
|
| 1639 |
+
"epoch": 10.0,
|
| 1640 |
+
"step": 1100,
|
| 1641 |
+
"total_flos": 1.6817604900715233e+18,
|
| 1642 |
+
"train_loss": 1.8264882094209844,
|
| 1643 |
+
"train_runtime": 8654.8771,
|
| 1644 |
+
"train_samples_per_second": 2.032,
|
| 1645 |
+
"train_steps_per_second": 0.127
|
| 1646 |
}
|
| 1647 |
],
|
| 1648 |
"logging_steps": 5,
|
| 1649 |
+
"max_steps": 1100,
|
| 1650 |
"num_input_tokens_seen": 0,
|
| 1651 |
+
"num_train_epochs": 10,
|
| 1652 |
"save_steps": 100,
|
| 1653 |
"stateful_callbacks": {
|
| 1654 |
"TrainerControl": {
|
|
|
|
| 1657 |
"should_evaluate": false,
|
| 1658 |
"should_log": false,
|
| 1659 |
"should_save": true,
|
| 1660 |
+
"should_training_stop": true
|
| 1661 |
},
|
| 1662 |
"attributes": {}
|
| 1663 |
}
|
| 1664 |
},
|
| 1665 |
+
"total_flos": 1.6817604900715233e+18,
|
| 1666 |
"train_batch_size": 4,
|
| 1667 |
"trial_name": null,
|
| 1668 |
"trial_params": null
|