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-gpt4o-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-gpt4o-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 | Step | Validation Loss |
|
| 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-gpt4o-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/gtgsbwvu)
|
| 20 |
# gemma7b-summarize-gpt4o-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: 3.2430
|
| 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 |
+
| 1.1572 | 1.0 | 111 | 2.3072 |
|
| 62 |
+
| 0.9296 | 2.0 | 222 | 2.1789 |
|
| 63 |
+
| 0.8273 | 3.0 | 333 | 2.1709 |
|
| 64 |
+
| 0.7586 | 4.0 | 444 | 2.2164 |
|
| 65 |
+
| 0.6613 | 5.0 | 555 | 2.3182 |
|
| 66 |
+
| 0.577 | 6.0 | 666 | 2.4774 |
|
| 67 |
+
| 0.4958 | 7.0 | 777 | 2.7036 |
|
| 68 |
+
| 0.4205 | 8.0 | 888 | 2.9689 |
|
| 69 |
+
| 0.382 | 9.0 | 999 | 3.2252 |
|
| 70 |
+
| 0.372 | 10.0 | 1110 | 3.2430 |
|
| 71 |
|
| 72 |
|
| 73 |
### Framework versions
|
adapter_model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 50056096
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:32b4355c727acdb0c6029f34cd21b7f1e40baf4881b93221c8019898e95b873f
|
| 3 |
size 50056096
|
all_results.json
CHANGED
|
@@ -1,14 +1,9 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"
|
| 4 |
-
"
|
| 5 |
-
"
|
| 6 |
-
"
|
| 7 |
-
"
|
| 8 |
-
"
|
| 9 |
-
"train_loss": 2.151051264783762,
|
| 10 |
-
"train_runtime": 5341.9856,
|
| 11 |
-
"train_samples": 29787,
|
| 12 |
-
"train_samples_per_second": 2.052,
|
| 13 |
-
"train_steps_per_second": 0.128
|
| 14 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"total_flos": 1.697049221804327e+18,
|
| 4 |
+
"train_loss": 1.8630313719715084,
|
| 5 |
+
"train_runtime": 9058.6901,
|
| 6 |
+
"train_samples": 32782,
|
| 7 |
+
"train_samples_per_second": 1.957,
|
| 8 |
+
"train_steps_per_second": 0.123
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 9 |
}
|
runs/May21_04-02-00_deep-diver-main-tough-snake-1-0-0/events.out.tfevents.1716278664.deep-diver-main-tough-snake-1-0-0.385.0
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 sha256:7fb3394383ff9d99394f680955b2c1d92f3e2570009d97a580da145d58da55e7
|
| 3 |
+
size 55599
|
train_results.json
CHANGED
|
@@ -1,9 +1,9 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"total_flos": 1.
|
| 4 |
-
"train_loss":
|
| 5 |
-
"train_runtime":
|
| 6 |
-
"train_samples":
|
| 7 |
-
"train_samples_per_second":
|
| 8 |
-
"train_steps_per_second": 0.
|
| 9 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"total_flos": 1.697049221804327e+18,
|
| 4 |
+
"train_loss": 1.8630313719715084,
|
| 5 |
+
"train_runtime": 9058.6901,
|
| 6 |
+
"train_samples": 32782,
|
| 7 |
+
"train_samples_per_second": 1.957,
|
| 8 |
+
"train_steps_per_second": 0.123
|
| 9 |
}
|
trainer_state.json
CHANGED
|
@@ -1,1033 +1,1668 @@
|
|
| 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":
|
| 92 |
-
"loss":
|
| 93 |
"step": 55
|
| 94 |
},
|
| 95 |
{
|
| 96 |
-
"epoch": 0.
|
| 97 |
-
"grad_norm": 3.
|
| 98 |
-
"learning_rate": 0.
|
| 99 |
-
"loss":
|
| 100 |
"step": 60
|
| 101 |
},
|
| 102 |
{
|
| 103 |
-
"epoch": 0.
|
| 104 |
-
"grad_norm": 2.
|
| 105 |
-
"learning_rate": 0.
|
| 106 |
-
"loss": 1.
|
| 107 |
"step": 65
|
| 108 |
},
|
| 109 |
{
|
| 110 |
-
"epoch": 0.
|
| 111 |
-
"grad_norm":
|
| 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": 1.
|
| 135 |
"step": 85
|
| 136 |
},
|
| 137 |
{
|
| 138 |
-
"epoch": 0.
|
| 139 |
-
"grad_norm":
|
| 140 |
-
"learning_rate": 0.
|
| 141 |
-
"loss": 1.
|
| 142 |
"step": 90
|
| 143 |
},
|
| 144 |
{
|
| 145 |
-
"epoch": 0.
|
| 146 |
-
"grad_norm":
|
| 147 |
-
"learning_rate": 0.
|
| 148 |
-
"loss": 1.
|
| 149 |
"step": 95
|
| 150 |
},
|
| 151 |
{
|
| 152 |
-
"epoch": 0.
|
| 153 |
-
"grad_norm":
|
| 154 |
-
"learning_rate": 0.
|
| 155 |
-
"loss": 1.
|
| 156 |
"step": 100
|
| 157 |
},
|
| 158 |
{
|
| 159 |
-
"epoch": 0.
|
| 160 |
-
"grad_norm":
|
| 161 |
-
"learning_rate": 0.
|
| 162 |
-
"loss": 1.
|
| 163 |
"step": 105
|
| 164 |
},
|
| 165 |
{
|
| 166 |
-
"epoch": 0.
|
| 167 |
-
"grad_norm": 2.
|
| 168 |
-
"learning_rate": 0.
|
| 169 |
-
"loss": 1.
|
| 170 |
"step": 110
|
| 171 |
},
|
| 172 |
{
|
| 173 |
-
"epoch": 0
|
| 174 |
-
"
|
| 175 |
-
"
|
| 176 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 177 |
"step": 115
|
| 178 |
},
|
| 179 |
{
|
| 180 |
-
"epoch":
|
| 181 |
-
"grad_norm":
|
| 182 |
-
"learning_rate": 0.
|
| 183 |
-
"loss":
|
| 184 |
"step": 120
|
| 185 |
},
|
| 186 |
{
|
| 187 |
-
"epoch":
|
| 188 |
-
"grad_norm":
|
| 189 |
-
"learning_rate": 0.
|
| 190 |
-
"loss":
|
| 191 |
"step": 125
|
| 192 |
},
|
| 193 |
{
|
| 194 |
-
"epoch":
|
| 195 |
-
"grad_norm":
|
| 196 |
-
"learning_rate": 0.
|
| 197 |
-
"loss":
|
| 198 |
"step": 130
|
| 199 |
},
|
| 200 |
{
|
| 201 |
-
"epoch":
|
| 202 |
-
"grad_norm": 0
|
| 203 |
-
"learning_rate": 0.
|
| 204 |
-
"loss":
|
| 205 |
"step": 135
|
| 206 |
},
|
| 207 |
{
|
| 208 |
-
"epoch": 1.
|
| 209 |
-
"
|
| 210 |
-
"
|
| 211 |
-
"
|
| 212 |
-
"eval_steps_per_second": 2.008,
|
| 213 |
-
"step": 137
|
| 214 |
-
},
|
| 215 |
-
{
|
| 216 |
-
"epoch": 1.0218978102189782,
|
| 217 |
-
"grad_norm": 2.125,
|
| 218 |
-
"learning_rate": 0.00019351552846298025,
|
| 219 |
-
"loss": 0.9374,
|
| 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": 1.
|
| 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":
|
| 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":
|
| 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":
|
| 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":
|
| 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": 1.
|
| 329 |
-
"grad_norm":
|
| 330 |
-
"learning_rate": 0.
|
| 331 |
-
"loss": 0.
|
| 332 |
"step": 220
|
| 333 |
},
|
| 334 |
{
|
| 335 |
-
"epoch":
|
| 336 |
-
"
|
| 337 |
-
"
|
| 338 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 339 |
"step": 225
|
| 340 |
},
|
| 341 |
{
|
| 342 |
-
"epoch":
|
| 343 |
-
"grad_norm":
|
| 344 |
-
"learning_rate": 0.
|
| 345 |
-
"loss": 0.
|
| 346 |
"step": 230
|
| 347 |
},
|
| 348 |
{
|
| 349 |
-
"epoch":
|
| 350 |
-
"grad_norm":
|
| 351 |
-
"learning_rate": 0.
|
| 352 |
-
"loss": 0.
|
| 353 |
"step": 235
|
| 354 |
},
|
| 355 |
{
|
| 356 |
-
"epoch":
|
| 357 |
-
"grad_norm":
|
| 358 |
-
"learning_rate": 0.
|
| 359 |
-
"loss": 0.
|
| 360 |
"step": 240
|
| 361 |
},
|
| 362 |
{
|
| 363 |
-
"epoch":
|
| 364 |
-
"grad_norm": 0.
|
| 365 |
-
"learning_rate": 0.
|
| 366 |
-
"loss": 0.
|
| 367 |
"step": 245
|
| 368 |
},
|
| 369 |
{
|
| 370 |
-
"epoch":
|
| 371 |
-
"grad_norm": 0.
|
| 372 |
-
"learning_rate": 0.
|
| 373 |
-
"loss": 0.
|
| 374 |
"step": 250
|
| 375 |
},
|
| 376 |
{
|
| 377 |
-
"epoch":
|
| 378 |
-
"grad_norm": 1.
|
| 379 |
-
"learning_rate": 0.
|
| 380 |
-
"loss": 0.
|
| 381 |
"step": 255
|
| 382 |
},
|
| 383 |
{
|
| 384 |
-
"epoch":
|
| 385 |
-
"grad_norm":
|
| 386 |
-
"learning_rate": 0.
|
| 387 |
-
"loss": 0.
|
| 388 |
"step": 260
|
| 389 |
},
|
| 390 |
{
|
| 391 |
-
"epoch":
|
| 392 |
-
"grad_norm":
|
| 393 |
-
"learning_rate": 0.
|
| 394 |
-
"loss": 0.
|
| 395 |
"step": 265
|
| 396 |
},
|
| 397 |
{
|
| 398 |
-
"epoch":
|
| 399 |
-
"grad_norm":
|
| 400 |
-
"learning_rate": 0.
|
| 401 |
-
"loss": 0.
|
| 402 |
"step": 270
|
| 403 |
},
|
| 404 |
{
|
| 405 |
-
"epoch": 2.
|
| 406 |
-
"
|
| 407 |
-
"
|
| 408 |
-
"
|
| 409 |
-
"eval_steps_per_second": 2.004,
|
| 410 |
-
"step": 274
|
| 411 |
-
},
|
| 412 |
-
{
|
| 413 |
-
"epoch": 2.0072992700729926,
|
| 414 |
-
"grad_norm": 0.80859375,
|
| 415 |
-
"learning_rate": 0.00014970526362019079,
|
| 416 |
-
"loss": 0.8435,
|
| 417 |
"step": 275
|
| 418 |
},
|
| 419 |
{
|
| 420 |
-
"epoch": 2.
|
| 421 |
-
"grad_norm":
|
| 422 |
-
"learning_rate": 0.
|
| 423 |
-
"loss": 0.
|
| 424 |
"step": 280
|
| 425 |
},
|
| 426 |
{
|
| 427 |
-
"epoch": 2.
|
| 428 |
-
"grad_norm":
|
| 429 |
-
"learning_rate": 0.
|
| 430 |
-
"loss": 0.
|
| 431 |
"step": 285
|
| 432 |
},
|
| 433 |
{
|
| 434 |
-
"epoch": 2.
|
| 435 |
-
"grad_norm": 0.
|
| 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": 0.
|
| 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": 0.
|
| 471 |
-
"learning_rate": 0.
|
| 472 |
-
"loss": 0.
|
| 473 |
"step": 315
|
| 474 |
},
|
| 475 |
{
|
| 476 |
-
"epoch": 2.
|
| 477 |
-
"grad_norm": 0.
|
| 478 |
-
"learning_rate": 0.
|
| 479 |
-
"loss": 0.
|
| 480 |
"step": 320
|
| 481 |
},
|
| 482 |
{
|
| 483 |
-
"epoch": 2.
|
| 484 |
-
"grad_norm": 1.
|
| 485 |
-
"learning_rate": 0.
|
| 486 |
-
"loss": 0.
|
| 487 |
"step": 325
|
| 488 |
},
|
| 489 |
{
|
| 490 |
-
"epoch": 2.
|
| 491 |
-
"grad_norm": 1.
|
| 492 |
-
"learning_rate": 0.
|
| 493 |
-
"loss": 0.
|
| 494 |
"step": 330
|
| 495 |
},
|
| 496 |
{
|
| 497 |
-
"epoch":
|
| 498 |
-
"
|
| 499 |
-
"
|
| 500 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 501 |
"step": 335
|
| 502 |
},
|
| 503 |
{
|
| 504 |
-
"epoch":
|
| 505 |
-
"grad_norm":
|
| 506 |
-
"learning_rate": 0.
|
| 507 |
-
"loss": 0.
|
| 508 |
"step": 340
|
| 509 |
},
|
| 510 |
{
|
| 511 |
-
"epoch":
|
| 512 |
-
"grad_norm": 0.
|
| 513 |
-
"learning_rate": 0.
|
| 514 |
-
"loss": 0.
|
| 515 |
"step": 345
|
| 516 |
},
|
| 517 |
{
|
| 518 |
-
"epoch":
|
| 519 |
-
"grad_norm":
|
| 520 |
-
"learning_rate": 0.
|
| 521 |
-
"loss": 0.
|
| 522 |
"step": 350
|
| 523 |
},
|
| 524 |
{
|
| 525 |
-
"epoch":
|
| 526 |
-
"grad_norm":
|
| 527 |
-
"learning_rate": 0.
|
| 528 |
-
"loss": 0.
|
| 529 |
"step": 355
|
| 530 |
},
|
| 531 |
{
|
| 532 |
-
"epoch":
|
| 533 |
-
"grad_norm":
|
| 534 |
-
"learning_rate": 0.
|
| 535 |
-
"loss": 0.
|
| 536 |
"step": 360
|
| 537 |
},
|
| 538 |
{
|
| 539 |
-
"epoch":
|
| 540 |
-
"grad_norm":
|
| 541 |
-
"learning_rate": 0.
|
| 542 |
-
"loss": 0.
|
| 543 |
"step": 365
|
| 544 |
},
|
| 545 |
{
|
| 546 |
-
"epoch":
|
| 547 |
-
"grad_norm":
|
| 548 |
-
"learning_rate": 0.
|
| 549 |
-
"loss": 0.
|
| 550 |
"step": 370
|
| 551 |
},
|
| 552 |
{
|
| 553 |
-
"epoch":
|
| 554 |
-
"grad_norm":
|
| 555 |
-
"learning_rate": 0.
|
| 556 |
-
"loss": 0.
|
| 557 |
"step": 375
|
| 558 |
},
|
| 559 |
{
|
| 560 |
-
"epoch":
|
| 561 |
-
"grad_norm":
|
| 562 |
-
"learning_rate":
|
| 563 |
-
"loss": 0.
|
| 564 |
"step": 380
|
| 565 |
},
|
| 566 |
{
|
| 567 |
-
"epoch":
|
| 568 |
-
"grad_norm":
|
| 569 |
-
"learning_rate":
|
| 570 |
-
"loss": 0.
|
| 571 |
"step": 385
|
| 572 |
},
|
| 573 |
{
|
| 574 |
-
"epoch":
|
| 575 |
-
"grad_norm":
|
| 576 |
-
"learning_rate":
|
| 577 |
-
"loss": 0.
|
| 578 |
"step": 390
|
| 579 |
},
|
| 580 |
{
|
| 581 |
-
"epoch":
|
| 582 |
-
"grad_norm":
|
| 583 |
-
"learning_rate":
|
| 584 |
-
"loss": 0.
|
| 585 |
"step": 395
|
| 586 |
},
|
| 587 |
{
|
| 588 |
-
"epoch":
|
| 589 |
-
"grad_norm": 0.
|
| 590 |
-
"learning_rate":
|
| 591 |
-
"loss": 0.
|
| 592 |
"step": 400
|
| 593 |
},
|
| 594 |
{
|
| 595 |
-
"epoch":
|
| 596 |
-
"grad_norm": 0.
|
| 597 |
-
"learning_rate":
|
| 598 |
-
"loss": 0.
|
| 599 |
"step": 405
|
| 600 |
},
|
| 601 |
{
|
| 602 |
-
"epoch":
|
| 603 |
-
"grad_norm":
|
| 604 |
-
"learning_rate":
|
| 605 |
-
"loss": 0.
|
| 606 |
"step": 410
|
| 607 |
},
|
| 608 |
{
|
| 609 |
-
"epoch": 3.
|
| 610 |
-
"
|
| 611 |
-
"
|
| 612 |
-
"
|
| 613 |
-
"eval_steps_per_second": 2.011,
|
| 614 |
-
"step": 411
|
| 615 |
-
},
|
| 616 |
-
{
|
| 617 |
-
"epoch": 3.0291970802919708,
|
| 618 |
-
"grad_norm": 0.8359375,
|
| 619 |
-
"learning_rate": 8.074113061741397e-05,
|
| 620 |
-
"loss": 0.7329,
|
| 621 |
"step": 415
|
| 622 |
},
|
| 623 |
{
|
| 624 |
-
"epoch": 3.
|
| 625 |
-
"grad_norm":
|
| 626 |
-
"learning_rate":
|
| 627 |
-
"loss": 0.
|
| 628 |
"step": 420
|
| 629 |
},
|
| 630 |
{
|
| 631 |
-
"epoch": 3.
|
| 632 |
-
"grad_norm": 0.
|
| 633 |
-
"learning_rate":
|
| 634 |
-
"loss": 0.
|
| 635 |
"step": 425
|
| 636 |
},
|
| 637 |
{
|
| 638 |
-
"epoch": 3.
|
| 639 |
-
"grad_norm": 0.
|
| 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": 3.
|
| 653 |
-
"grad_norm": 0.
|
| 654 |
-
"learning_rate":
|
| 655 |
-
"loss": 0.
|
| 656 |
"step": 440
|
| 657 |
},
|
| 658 |
{
|
| 659 |
-
"epoch":
|
| 660 |
-
"
|
| 661 |
-
"
|
| 662 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 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":
|
| 681 |
-
"grad_norm": 0.
|
| 682 |
-
"learning_rate":
|
| 683 |
-
"loss": 0.
|
| 684 |
"step": 460
|
| 685 |
},
|
| 686 |
{
|
| 687 |
-
"epoch":
|
| 688 |
-
"grad_norm":
|
| 689 |
-
"learning_rate":
|
| 690 |
-
"loss": 0.
|
| 691 |
"step": 465
|
| 692 |
},
|
| 693 |
{
|
| 694 |
-
"epoch":
|
| 695 |
-
"grad_norm": 0.
|
| 696 |
-
"learning_rate":
|
| 697 |
-
"loss": 0.
|
| 698 |
"step": 470
|
| 699 |
},
|
| 700 |
{
|
| 701 |
-
"epoch":
|
| 702 |
-
"grad_norm": 0.
|
| 703 |
-
"learning_rate":
|
| 704 |
-
"loss": 0.
|
| 705 |
"step": 475
|
| 706 |
},
|
| 707 |
{
|
| 708 |
-
"epoch":
|
| 709 |
-
"grad_norm": 0.
|
| 710 |
-
"learning_rate":
|
| 711 |
-
"loss": 0.
|
| 712 |
"step": 480
|
| 713 |
},
|
| 714 |
{
|
| 715 |
-
"epoch":
|
| 716 |
-
"grad_norm":
|
| 717 |
-
"learning_rate":
|
| 718 |
-
"loss": 0.
|
| 719 |
"step": 485
|
| 720 |
},
|
| 721 |
{
|
| 722 |
-
"epoch":
|
| 723 |
-
"grad_norm":
|
| 724 |
-
"learning_rate":
|
| 725 |
-
"loss": 0.
|
| 726 |
"step": 490
|
| 727 |
},
|
| 728 |
{
|
| 729 |
-
"epoch":
|
| 730 |
-
"grad_norm": 0.
|
| 731 |
-
"learning_rate":
|
| 732 |
-
"loss": 0.
|
| 733 |
"step": 495
|
| 734 |
},
|
| 735 |
{
|
| 736 |
-
"epoch":
|
| 737 |
-
"grad_norm": 0.
|
| 738 |
-
"learning_rate":
|
| 739 |
-
"loss": 0.
|
| 740 |
"step": 500
|
| 741 |
},
|
| 742 |
{
|
| 743 |
-
"epoch":
|
| 744 |
-
"grad_norm": 0.
|
| 745 |
-
"learning_rate":
|
| 746 |
-
"loss": 0.
|
| 747 |
"step": 505
|
| 748 |
},
|
| 749 |
{
|
| 750 |
-
"epoch":
|
| 751 |
-
"grad_norm": 0.
|
| 752 |
-
"learning_rate":
|
| 753 |
-
"loss": 0.
|
| 754 |
"step": 510
|
| 755 |
},
|
| 756 |
{
|
| 757 |
-
"epoch":
|
| 758 |
-
"grad_norm": 0.
|
| 759 |
-
"learning_rate":
|
| 760 |
-
"loss": 0.
|
| 761 |
"step": 515
|
| 762 |
},
|
| 763 |
{
|
| 764 |
-
"epoch":
|
| 765 |
-
"grad_norm": 0.
|
| 766 |
-
"learning_rate":
|
| 767 |
-
"loss": 0.
|
| 768 |
"step": 520
|
| 769 |
},
|
| 770 |
{
|
| 771 |
-
"epoch":
|
| 772 |
-
"grad_norm": 0.
|
| 773 |
-
"learning_rate":
|
| 774 |
-
"loss": 0.
|
| 775 |
"step": 525
|
| 776 |
},
|
| 777 |
{
|
| 778 |
-
"epoch":
|
| 779 |
-
"grad_norm": 0.
|
| 780 |
-
"learning_rate":
|
| 781 |
-
"loss": 0.
|
| 782 |
"step": 530
|
| 783 |
},
|
| 784 |
{
|
| 785 |
-
"epoch":
|
| 786 |
-
"grad_norm": 0.
|
| 787 |
-
"learning_rate":
|
| 788 |
-
"loss": 0.
|
| 789 |
"step": 535
|
| 790 |
},
|
| 791 |
{
|
| 792 |
-
"epoch":
|
| 793 |
-
"grad_norm": 0.
|
| 794 |
-
"learning_rate":
|
| 795 |
-
"loss": 0.
|
| 796 |
"step": 540
|
| 797 |
},
|
| 798 |
{
|
| 799 |
-
"epoch":
|
| 800 |
-
"grad_norm": 0.
|
| 801 |
-
"learning_rate":
|
| 802 |
-
"loss": 0.
|
| 803 |
"step": 545
|
| 804 |
},
|
| 805 |
{
|
| 806 |
-
"epoch": 4.
|
| 807 |
-
"
|
| 808 |
-
"
|
| 809 |
-
"
|
| 810 |
-
"
|
| 811 |
-
"step": 548
|
| 812 |
},
|
| 813 |
{
|
| 814 |
-
"epoch":
|
| 815 |
-
"grad_norm": 0.
|
| 816 |
-
"learning_rate":
|
| 817 |
-
"loss": 0.
|
| 818 |
-
"step":
|
| 819 |
},
|
| 820 |
{
|
| 821 |
-
"epoch":
|
| 822 |
-
"
|
| 823 |
-
"
|
| 824 |
-
"
|
|
|
|
| 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":
|
| 845 |
-
"loss": 0.
|
| 846 |
"step": 570
|
| 847 |
},
|
| 848 |
{
|
| 849 |
-
"epoch":
|
| 850 |
-
"grad_norm": 0.
|
| 851 |
-
"learning_rate":
|
| 852 |
-
"loss": 0.
|
| 853 |
"step": 575
|
| 854 |
},
|
| 855 |
{
|
| 856 |
-
"epoch":
|
| 857 |
-
"grad_norm": 0.
|
| 858 |
-
"learning_rate":
|
| 859 |
-
"loss": 0.
|
| 860 |
"step": 580
|
| 861 |
},
|
| 862 |
{
|
| 863 |
-
"epoch":
|
| 864 |
-
"grad_norm": 0.
|
| 865 |
-
"learning_rate":
|
| 866 |
-
"loss": 0.
|
| 867 |
"step": 585
|
| 868 |
},
|
| 869 |
{
|
| 870 |
-
"epoch":
|
| 871 |
-
"grad_norm": 0.
|
| 872 |
-
"learning_rate":
|
| 873 |
-
"loss": 0.
|
| 874 |
"step": 590
|
| 875 |
},
|
| 876 |
{
|
| 877 |
-
"epoch":
|
| 878 |
-
"grad_norm": 0.
|
| 879 |
-
"learning_rate":
|
| 880 |
-
"loss": 0.
|
| 881 |
"step": 595
|
| 882 |
},
|
| 883 |
{
|
| 884 |
-
"epoch":
|
| 885 |
-
"grad_norm": 0.
|
| 886 |
-
"learning_rate":
|
| 887 |
-
"loss": 0.
|
| 888 |
"step": 600
|
| 889 |
},
|
| 890 |
{
|
| 891 |
-
"epoch":
|
| 892 |
-
"grad_norm": 0.
|
| 893 |
-
"learning_rate":
|
| 894 |
-
"loss": 0.
|
| 895 |
"step": 605
|
| 896 |
},
|
| 897 |
{
|
| 898 |
-
"epoch":
|
| 899 |
-
"grad_norm": 0.
|
| 900 |
-
"learning_rate":
|
| 901 |
-
"loss": 0.
|
| 902 |
"step": 610
|
| 903 |
},
|
| 904 |
{
|
| 905 |
-
"epoch":
|
| 906 |
-
"grad_norm": 0.
|
| 907 |
-
"learning_rate":
|
| 908 |
-
"loss": 0.
|
| 909 |
"step": 615
|
| 910 |
},
|
| 911 |
{
|
| 912 |
-
"epoch":
|
| 913 |
-
"grad_norm": 0.
|
| 914 |
-
"learning_rate":
|
| 915 |
-
"loss": 0.
|
| 916 |
"step": 620
|
| 917 |
},
|
| 918 |
{
|
| 919 |
-
"epoch":
|
| 920 |
-
"grad_norm": 0.
|
| 921 |
-
"learning_rate":
|
| 922 |
-
"loss": 0.
|
| 923 |
"step": 625
|
| 924 |
},
|
| 925 |
{
|
| 926 |
-
"epoch":
|
| 927 |
-
"grad_norm": 0.
|
| 928 |
-
"learning_rate":
|
| 929 |
-
"loss": 0.
|
| 930 |
"step": 630
|
| 931 |
},
|
| 932 |
{
|
| 933 |
-
"epoch":
|
| 934 |
-
"grad_norm": 0.
|
| 935 |
-
"learning_rate":
|
| 936 |
-
"loss": 0.
|
| 937 |
"step": 635
|
| 938 |
},
|
| 939 |
{
|
| 940 |
-
"epoch":
|
| 941 |
-
"grad_norm": 0.
|
| 942 |
-
"learning_rate":
|
| 943 |
-
"loss": 0.
|
| 944 |
"step": 640
|
| 945 |
},
|
| 946 |
{
|
| 947 |
-
"epoch":
|
| 948 |
-
"grad_norm": 0.
|
| 949 |
-
"learning_rate":
|
| 950 |
-
"loss": 0.
|
| 951 |
"step": 645
|
| 952 |
},
|
| 953 |
{
|
| 954 |
-
"epoch":
|
| 955 |
-
"grad_norm": 0.
|
| 956 |
-
"learning_rate":
|
| 957 |
-
"loss": 0.
|
| 958 |
"step": 650
|
| 959 |
},
|
| 960 |
{
|
| 961 |
-
"epoch":
|
| 962 |
-
"grad_norm": 0.
|
| 963 |
-
"learning_rate":
|
| 964 |
-
"loss": 0.
|
| 965 |
"step": 655
|
| 966 |
},
|
| 967 |
{
|
| 968 |
-
"epoch":
|
| 969 |
-
"grad_norm": 0.
|
| 970 |
-
"learning_rate": 8.
|
| 971 |
-
"loss": 0.
|
| 972 |
"step": 660
|
| 973 |
},
|
| 974 |
{
|
| 975 |
-
"epoch":
|
| 976 |
-
"grad_norm": 0.
|
| 977 |
-
"learning_rate":
|
| 978 |
-
"loss": 0.
|
| 979 |
"step": 665
|
| 980 |
},
|
| 981 |
{
|
| 982 |
-
"epoch":
|
| 983 |
-
"
|
| 984 |
-
"
|
| 985 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 986 |
"step": 670
|
| 987 |
},
|
| 988 |
{
|
| 989 |
-
"epoch":
|
| 990 |
-
"grad_norm": 0.
|
| 991 |
-
"learning_rate":
|
| 992 |
-
"loss": 0.
|
| 993 |
"step": 675
|
| 994 |
},
|
| 995 |
{
|
| 996 |
-
"epoch":
|
| 997 |
-
"grad_norm": 0.
|
| 998 |
-
"learning_rate":
|
| 999 |
-
"loss": 0.
|
| 1000 |
"step": 680
|
| 1001 |
},
|
| 1002 |
{
|
| 1003 |
-
"epoch":
|
| 1004 |
-
"grad_norm": 0.
|
| 1005 |
-
"learning_rate":
|
| 1006 |
-
"loss": 0.
|
| 1007 |
"step": 685
|
| 1008 |
},
|
| 1009 |
{
|
| 1010 |
-
"epoch":
|
| 1011 |
-
"
|
| 1012 |
-
"
|
| 1013 |
-
"
|
| 1014 |
-
"
|
| 1015 |
-
"step": 685
|
| 1016 |
},
|
| 1017 |
{
|
| 1018 |
-
"epoch":
|
| 1019 |
-
"
|
| 1020 |
-
"
|
| 1021 |
-
"
|
| 1022 |
-
"
|
| 1023 |
-
|
| 1024 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1025 |
}
|
| 1026 |
],
|
| 1027 |
"logging_steps": 5,
|
| 1028 |
-
"max_steps":
|
| 1029 |
"num_input_tokens_seen": 0,
|
| 1030 |
-
"num_train_epochs":
|
| 1031 |
"save_steps": 100,
|
| 1032 |
"stateful_callbacks": {
|
| 1033 |
"TrainerControl": {
|
|
@@ -1041,7 +1676,7 @@
|
|
| 1041 |
"attributes": {}
|
| 1042 |
}
|
| 1043 |
},
|
| 1044 |
-
"total_flos": 1.
|
| 1045 |
"train_batch_size": 4,
|
| 1046 |
"trial_name": null,
|
| 1047 |
"trial_params": null
|
|
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 10.0,
|
| 5 |
"eval_steps": 500,
|
| 6 |
+
"global_step": 1110,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
+
"epoch": 0.009009009009009009,
|
| 13 |
+
"grad_norm": 608.0,
|
| 14 |
+
"learning_rate": 1.801801801801802e-06,
|
| 15 |
+
"loss": 58.5641,
|
| 16 |
"step": 1
|
| 17 |
},
|
| 18 |
{
|
| 19 |
+
"epoch": 0.04504504504504504,
|
| 20 |
+
"grad_norm": 532.0,
|
| 21 |
+
"learning_rate": 9.00900900900901e-06,
|
| 22 |
+
"loss": 54.6181,
|
| 23 |
"step": 5
|
| 24 |
},
|
| 25 |
{
|
| 26 |
+
"epoch": 0.09009009009009009,
|
| 27 |
+
"grad_norm": 446.0,
|
| 28 |
+
"learning_rate": 1.801801801801802e-05,
|
| 29 |
+
"loss": 50.0236,
|
| 30 |
"step": 10
|
| 31 |
},
|
| 32 |
{
|
| 33 |
+
"epoch": 0.13513513513513514,
|
| 34 |
+
"grad_norm": 193.0,
|
| 35 |
+
"learning_rate": 2.702702702702703e-05,
|
| 36 |
+
"loss": 33.1549,
|
| 37 |
"step": 15
|
| 38 |
},
|
| 39 |
{
|
| 40 |
+
"epoch": 0.18018018018018017,
|
| 41 |
+
"grad_norm": 44.5,
|
| 42 |
+
"learning_rate": 3.603603603603604e-05,
|
| 43 |
+
"loss": 25.2428,
|
| 44 |
"step": 20
|
| 45 |
},
|
| 46 |
{
|
| 47 |
+
"epoch": 0.22522522522522523,
|
| 48 |
+
"grad_norm": 26.625,
|
| 49 |
+
"learning_rate": 4.5045045045045046e-05,
|
| 50 |
+
"loss": 22.4735,
|
| 51 |
"step": 25
|
| 52 |
},
|
| 53 |
{
|
| 54 |
+
"epoch": 0.2702702702702703,
|
| 55 |
+
"grad_norm": 17.25,
|
| 56 |
+
"learning_rate": 5.405405405405406e-05,
|
| 57 |
+
"loss": 20.4661,
|
| 58 |
"step": 30
|
| 59 |
},
|
| 60 |
{
|
| 61 |
+
"epoch": 0.3153153153153153,
|
| 62 |
+
"grad_norm": 7.6875,
|
| 63 |
+
"learning_rate": 6.306306306306306e-05,
|
| 64 |
+
"loss": 19.1401,
|
| 65 |
"step": 35
|
| 66 |
},
|
| 67 |
{
|
| 68 |
+
"epoch": 0.36036036036036034,
|
| 69 |
+
"grad_norm": 11.6875,
|
| 70 |
+
"learning_rate": 7.207207207207208e-05,
|
| 71 |
+
"loss": 18.3188,
|
| 72 |
"step": 40
|
| 73 |
},
|
| 74 |
{
|
| 75 |
+
"epoch": 0.40540540540540543,
|
| 76 |
+
"grad_norm": 23.5,
|
| 77 |
+
"learning_rate": 8.108108108108109e-05,
|
| 78 |
+
"loss": 16.7622,
|
| 79 |
"step": 45
|
| 80 |
},
|
| 81 |
{
|
| 82 |
+
"epoch": 0.45045045045045046,
|
| 83 |
+
"grad_norm": 56.25,
|
| 84 |
+
"learning_rate": 9.009009009009009e-05,
|
| 85 |
+
"loss": 12.6183,
|
| 86 |
"step": 50
|
| 87 |
},
|
| 88 |
{
|
| 89 |
+
"epoch": 0.4954954954954955,
|
| 90 |
+
"grad_norm": 13.0,
|
| 91 |
+
"learning_rate": 9.90990990990991e-05,
|
| 92 |
+
"loss": 4.3593,
|
| 93 |
"step": 55
|
| 94 |
},
|
| 95 |
{
|
| 96 |
+
"epoch": 0.5405405405405406,
|
| 97 |
+
"grad_norm": 3.875,
|
| 98 |
+
"learning_rate": 0.00010810810810810812,
|
| 99 |
+
"loss": 2.18,
|
| 100 |
"step": 60
|
| 101 |
},
|
| 102 |
{
|
| 103 |
+
"epoch": 0.5855855855855856,
|
| 104 |
+
"grad_norm": 2.421875,
|
| 105 |
+
"learning_rate": 0.00011711711711711712,
|
| 106 |
+
"loss": 1.8179,
|
| 107 |
"step": 65
|
| 108 |
},
|
| 109 |
{
|
| 110 |
+
"epoch": 0.6306306306306306,
|
| 111 |
+
"grad_norm": 3.265625,
|
| 112 |
+
"learning_rate": 0.00012612612612612612,
|
| 113 |
+
"loss": 1.5974,
|
| 114 |
"step": 70
|
| 115 |
},
|
| 116 |
{
|
| 117 |
+
"epoch": 0.6756756756756757,
|
| 118 |
+
"grad_norm": 2.375,
|
| 119 |
+
"learning_rate": 0.00013513513513513514,
|
| 120 |
+
"loss": 1.486,
|
| 121 |
"step": 75
|
| 122 |
},
|
| 123 |
{
|
| 124 |
+
"epoch": 0.7207207207207207,
|
| 125 |
+
"grad_norm": 1.5078125,
|
| 126 |
+
"learning_rate": 0.00014414414414414415,
|
| 127 |
+
"loss": 1.361,
|
| 128 |
"step": 80
|
| 129 |
},
|
| 130 |
{
|
| 131 |
+
"epoch": 0.7657657657657657,
|
| 132 |
+
"grad_norm": 3.890625,
|
| 133 |
+
"learning_rate": 0.00015315315315315314,
|
| 134 |
+
"loss": 1.3001,
|
| 135 |
"step": 85
|
| 136 |
},
|
| 137 |
{
|
| 138 |
+
"epoch": 0.8108108108108109,
|
| 139 |
+
"grad_norm": 4.46875,
|
| 140 |
+
"learning_rate": 0.00016216216216216218,
|
| 141 |
+
"loss": 1.261,
|
| 142 |
"step": 90
|
| 143 |
},
|
| 144 |
{
|
| 145 |
+
"epoch": 0.8558558558558559,
|
| 146 |
+
"grad_norm": 5.03125,
|
| 147 |
+
"learning_rate": 0.0001711711711711712,
|
| 148 |
+
"loss": 1.2015,
|
| 149 |
"step": 95
|
| 150 |
},
|
| 151 |
{
|
| 152 |
+
"epoch": 0.9009009009009009,
|
| 153 |
+
"grad_norm": 32.25,
|
| 154 |
+
"learning_rate": 0.00018018018018018018,
|
| 155 |
+
"loss": 1.1886,
|
| 156 |
"step": 100
|
| 157 |
},
|
| 158 |
{
|
| 159 |
+
"epoch": 0.9459459459459459,
|
| 160 |
+
"grad_norm": 1.703125,
|
| 161 |
+
"learning_rate": 0.0001891891891891892,
|
| 162 |
+
"loss": 1.1679,
|
| 163 |
"step": 105
|
| 164 |
},
|
| 165 |
{
|
| 166 |
+
"epoch": 0.990990990990991,
|
| 167 |
+
"grad_norm": 2.984375,
|
| 168 |
+
"learning_rate": 0.0001981981981981982,
|
| 169 |
+
"loss": 1.1572,
|
| 170 |
"step": 110
|
| 171 |
},
|
| 172 |
{
|
| 173 |
+
"epoch": 1.0,
|
| 174 |
+
"eval_loss": 2.307225465774536,
|
| 175 |
+
"eval_runtime": 1.0056,
|
| 176 |
+
"eval_samples_per_second": 4.972,
|
| 177 |
+
"eval_steps_per_second": 1.989,
|
| 178 |
+
"step": 111
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 1.0360360360360361,
|
| 182 |
+
"grad_norm": 1.546875,
|
| 183 |
+
"learning_rate": 0.00019999208860571255,
|
| 184 |
+
"loss": 1.0473,
|
| 185 |
"step": 115
|
| 186 |
},
|
| 187 |
{
|
| 188 |
+
"epoch": 1.0810810810810811,
|
| 189 |
+
"grad_norm": 1.546875,
|
| 190 |
+
"learning_rate": 0.0001999599507118322,
|
| 191 |
+
"loss": 1.0618,
|
| 192 |
"step": 120
|
| 193 |
},
|
| 194 |
{
|
| 195 |
+
"epoch": 1.1261261261261262,
|
| 196 |
+
"grad_norm": 10.0,
|
| 197 |
+
"learning_rate": 0.00019990309979553045,
|
| 198 |
+
"loss": 1.0458,
|
| 199 |
"step": 125
|
| 200 |
},
|
| 201 |
{
|
| 202 |
+
"epoch": 1.1711711711711712,
|
| 203 |
+
"grad_norm": 8.4375,
|
| 204 |
+
"learning_rate": 0.00019982154991201608,
|
| 205 |
+
"loss": 1.0364,
|
| 206 |
"step": 130
|
| 207 |
},
|
| 208 |
{
|
| 209 |
+
"epoch": 1.2162162162162162,
|
| 210 |
+
"grad_norm": 2.0,
|
| 211 |
+
"learning_rate": 0.00019971532122280464,
|
| 212 |
+
"loss": 1.0457,
|
| 213 |
"step": 135
|
| 214 |
},
|
| 215 |
{
|
| 216 |
+
"epoch": 1.2612612612612613,
|
| 217 |
+
"grad_norm": 1.4453125,
|
| 218 |
+
"learning_rate": 0.00019958443999073397,
|
| 219 |
+
"loss": 0.9906,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 220 |
"step": 140
|
| 221 |
},
|
| 222 |
{
|
| 223 |
+
"epoch": 1.3063063063063063,
|
| 224 |
+
"grad_norm": 18.25,
|
| 225 |
+
"learning_rate": 0.00019942893857347128,
|
| 226 |
+
"loss": 0.9911,
|
| 227 |
"step": 145
|
| 228 |
},
|
| 229 |
{
|
| 230 |
+
"epoch": 1.3513513513513513,
|
| 231 |
+
"grad_norm": 2.578125,
|
| 232 |
+
"learning_rate": 0.0001992488554155135,
|
| 233 |
+
"loss": 0.9996,
|
| 234 |
"step": 150
|
| 235 |
},
|
| 236 |
{
|
| 237 |
+
"epoch": 1.3963963963963963,
|
| 238 |
+
"grad_norm": 1.7734375,
|
| 239 |
+
"learning_rate": 0.00019904423503868247,
|
| 240 |
+
"loss": 0.9656,
|
| 241 |
"step": 155
|
| 242 |
},
|
| 243 |
{
|
| 244 |
+
"epoch": 1.4414414414414414,
|
| 245 |
+
"grad_norm": 5.65625,
|
| 246 |
+
"learning_rate": 0.00019881512803111796,
|
| 247 |
+
"loss": 0.9753,
|
| 248 |
"step": 160
|
| 249 |
},
|
| 250 |
{
|
| 251 |
+
"epoch": 1.4864864864864864,
|
| 252 |
+
"grad_norm": 3.78125,
|
| 253 |
+
"learning_rate": 0.00019856159103477086,
|
| 254 |
+
"loss": 0.9239,
|
| 255 |
"step": 165
|
| 256 |
},
|
| 257 |
{
|
| 258 |
+
"epoch": 1.5315315315315314,
|
| 259 |
+
"grad_norm": 0.86328125,
|
| 260 |
+
"learning_rate": 0.00019828368673139947,
|
| 261 |
+
"loss": 0.9428,
|
| 262 |
"step": 170
|
| 263 |
},
|
| 264 |
{
|
| 265 |
+
"epoch": 1.5765765765765765,
|
| 266 |
+
"grad_norm": 0.7265625,
|
| 267 |
+
"learning_rate": 0.00019798148382707296,
|
| 268 |
+
"loss": 0.9455,
|
| 269 |
"step": 175
|
| 270 |
},
|
| 271 |
{
|
| 272 |
+
"epoch": 1.6216216216216215,
|
| 273 |
+
"grad_norm": 1.8125,
|
| 274 |
+
"learning_rate": 0.00019765505703518496,
|
| 275 |
+
"loss": 0.9373,
|
| 276 |
"step": 180
|
| 277 |
},
|
| 278 |
{
|
| 279 |
+
"epoch": 1.6666666666666665,
|
| 280 |
+
"grad_norm": 1.0078125,
|
| 281 |
+
"learning_rate": 0.00019730448705798239,
|
| 282 |
+
"loss": 0.9659,
|
| 283 |
"step": 185
|
| 284 |
},
|
| 285 |
{
|
| 286 |
+
"epoch": 1.7117117117117115,
|
| 287 |
+
"grad_norm": 3.859375,
|
| 288 |
+
"learning_rate": 0.00019692986056661356,
|
| 289 |
+
"loss": 0.9271,
|
| 290 |
"step": 190
|
| 291 |
},
|
| 292 |
{
|
| 293 |
+
"epoch": 1.7567567567567568,
|
| 294 |
+
"grad_norm": 3.34375,
|
| 295 |
+
"learning_rate": 0.00019653127017970034,
|
| 296 |
+
"loss": 0.9303,
|
| 297 |
"step": 195
|
| 298 |
},
|
| 299 |
{
|
| 300 |
+
"epoch": 1.8018018018018018,
|
| 301 |
+
"grad_norm": 0.91796875,
|
| 302 |
+
"learning_rate": 0.0001961088144404403,
|
| 303 |
+
"loss": 0.9333,
|
| 304 |
"step": 200
|
| 305 |
},
|
| 306 |
{
|
| 307 |
+
"epoch": 1.8468468468468469,
|
| 308 |
+
"grad_norm": 1.4453125,
|
| 309 |
+
"learning_rate": 0.00019566259779224378,
|
| 310 |
+
"loss": 0.8923,
|
| 311 |
"step": 205
|
| 312 |
},
|
| 313 |
{
|
| 314 |
+
"epoch": 1.8918918918918919,
|
| 315 |
+
"grad_norm": 3.171875,
|
| 316 |
+
"learning_rate": 0.00019519273055291266,
|
| 317 |
+
"loss": 0.9,
|
| 318 |
"step": 210
|
| 319 |
},
|
| 320 |
{
|
| 321 |
+
"epoch": 1.936936936936937,
|
| 322 |
+
"grad_norm": 14.3125,
|
| 323 |
+
"learning_rate": 0.00019469932888736632,
|
| 324 |
+
"loss": 0.8988,
|
| 325 |
"step": 215
|
| 326 |
},
|
| 327 |
{
|
| 328 |
+
"epoch": 1.981981981981982,
|
| 329 |
+
"grad_norm": 3.46875,
|
| 330 |
+
"learning_rate": 0.0001941825147789225,
|
| 331 |
+
"loss": 0.9296,
|
| 332 |
"step": 220
|
| 333 |
},
|
| 334 |
{
|
| 335 |
+
"epoch": 2.0,
|
| 336 |
+
"eval_loss": 2.178852081298828,
|
| 337 |
+
"eval_runtime": 1.0053,
|
| 338 |
+
"eval_samples_per_second": 4.973,
|
| 339 |
+
"eval_steps_per_second": 1.989,
|
| 340 |
+
"step": 222
|
| 341 |
+
},
|
| 342 |
+
{
|
| 343 |
+
"epoch": 2.027027027027027,
|
| 344 |
+
"grad_norm": 1.4921875,
|
| 345 |
+
"learning_rate": 0.00019364241599913924,
|
| 346 |
+
"loss": 0.8696,
|
| 347 |
"step": 225
|
| 348 |
},
|
| 349 |
{
|
| 350 |
+
"epoch": 2.0720720720720722,
|
| 351 |
+
"grad_norm": 2.65625,
|
| 352 |
+
"learning_rate": 0.0001930791660762262,
|
| 353 |
+
"loss": 0.8363,
|
| 354 |
"step": 230
|
| 355 |
},
|
| 356 |
{
|
| 357 |
+
"epoch": 2.1171171171171173,
|
| 358 |
+
"grad_norm": 1.265625,
|
| 359 |
+
"learning_rate": 0.00019249290426203252,
|
| 360 |
+
"loss": 0.821,
|
| 361 |
"step": 235
|
| 362 |
},
|
| 363 |
{
|
| 364 |
+
"epoch": 2.1621621621621623,
|
| 365 |
+
"grad_norm": 2.546875,
|
| 366 |
+
"learning_rate": 0.00019188377549761963,
|
| 367 |
+
"loss": 0.8511,
|
| 368 |
"step": 240
|
| 369 |
},
|
| 370 |
{
|
| 371 |
+
"epoch": 2.2072072072072073,
|
| 372 |
+
"grad_norm": 0.828125,
|
| 373 |
+
"learning_rate": 0.0001912519303774276,
|
| 374 |
+
"loss": 0.8231,
|
| 375 |
"step": 245
|
| 376 |
},
|
| 377 |
{
|
| 378 |
+
"epoch": 2.2522522522522523,
|
| 379 |
+
"grad_norm": 0.73046875,
|
| 380 |
+
"learning_rate": 0.000190597525112044,
|
| 381 |
+
"loss": 0.8496,
|
| 382 |
"step": 250
|
| 383 |
},
|
| 384 |
{
|
| 385 |
+
"epoch": 2.2972972972972974,
|
| 386 |
+
"grad_norm": 1.2421875,
|
| 387 |
+
"learning_rate": 0.00018992072148958368,
|
| 388 |
+
"loss": 0.852,
|
| 389 |
"step": 255
|
| 390 |
},
|
| 391 |
{
|
| 392 |
+
"epoch": 2.3423423423423424,
|
| 393 |
+
"grad_norm": 1.578125,
|
| 394 |
+
"learning_rate": 0.0001892216868356904,
|
| 395 |
+
"loss": 0.8131,
|
| 396 |
"step": 260
|
| 397 |
},
|
| 398 |
{
|
| 399 |
+
"epoch": 2.3873873873873874,
|
| 400 |
+
"grad_norm": 1.5078125,
|
| 401 |
+
"learning_rate": 0.00018850059397216876,
|
| 402 |
+
"loss": 0.8483,
|
| 403 |
"step": 265
|
| 404 |
},
|
| 405 |
{
|
| 406 |
+
"epoch": 2.4324324324324325,
|
| 407 |
+
"grad_norm": 1.125,
|
| 408 |
+
"learning_rate": 0.00018775762117425777,
|
| 409 |
+
"loss": 0.8432,
|
| 410 |
"step": 270
|
| 411 |
},
|
| 412 |
{
|
| 413 |
+
"epoch": 2.4774774774774775,
|
| 414 |
+
"grad_norm": 0.6015625,
|
| 415 |
+
"learning_rate": 0.00018699295212655596,
|
| 416 |
+
"loss": 0.8493,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 417 |
"step": 275
|
| 418 |
},
|
| 419 |
{
|
| 420 |
+
"epoch": 2.5225225225225225,
|
| 421 |
+
"grad_norm": 0.8828125,
|
| 422 |
+
"learning_rate": 0.00018620677587760916,
|
| 423 |
+
"loss": 0.7998,
|
| 424 |
"step": 280
|
| 425 |
},
|
| 426 |
{
|
| 427 |
+
"epoch": 2.5675675675675675,
|
| 428 |
+
"grad_norm": 0.73046875,
|
| 429 |
+
"learning_rate": 0.0001853992867931721,
|
| 430 |
+
"loss": 0.8256,
|
| 431 |
"step": 285
|
| 432 |
},
|
| 433 |
{
|
| 434 |
+
"epoch": 2.6126126126126126,
|
| 435 |
+
"grad_norm": 0.6796875,
|
| 436 |
+
"learning_rate": 0.00018457068450815562,
|
| 437 |
+
"loss": 0.8162,
|
| 438 |
"step": 290
|
| 439 |
},
|
| 440 |
{
|
| 441 |
+
"epoch": 2.6576576576576576,
|
| 442 |
+
"grad_norm": 0.671875,
|
| 443 |
+
"learning_rate": 0.0001837211738772711,
|
| 444 |
+
"loss": 0.8338,
|
| 445 |
"step": 295
|
| 446 |
},
|
| 447 |
{
|
| 448 |
+
"epoch": 2.7027027027027026,
|
| 449 |
+
"grad_norm": 0.9140625,
|
| 450 |
+
"learning_rate": 0.00018285096492438424,
|
| 451 |
+
"loss": 0.8279,
|
| 452 |
"step": 300
|
| 453 |
},
|
| 454 |
{
|
| 455 |
+
"epoch": 2.7477477477477477,
|
| 456 |
+
"grad_norm": 0.60546875,
|
| 457 |
+
"learning_rate": 0.00018196027279059117,
|
| 458 |
+
"loss": 0.7962,
|
| 459 |
"step": 305
|
| 460 |
},
|
| 461 |
{
|
| 462 |
+
"epoch": 2.7927927927927927,
|
| 463 |
+
"grad_norm": 2.78125,
|
| 464 |
+
"learning_rate": 0.0001810493176810292,
|
| 465 |
+
"loss": 0.8192,
|
| 466 |
"step": 310
|
| 467 |
},
|
| 468 |
{
|
| 469 |
+
"epoch": 2.8378378378378377,
|
| 470 |
+
"grad_norm": 0.63671875,
|
| 471 |
+
"learning_rate": 0.00018011832481043576,
|
| 472 |
+
"loss": 0.8147,
|
| 473 |
"step": 315
|
| 474 |
},
|
| 475 |
{
|
| 476 |
+
"epoch": 2.8828828828828827,
|
| 477 |
+
"grad_norm": 0.56640625,
|
| 478 |
+
"learning_rate": 0.00017916752434746856,
|
| 479 |
+
"loss": 0.8255,
|
| 480 |
"step": 320
|
| 481 |
},
|
| 482 |
{
|
| 483 |
+
"epoch": 2.9279279279279278,
|
| 484 |
+
"grad_norm": 1.8046875,
|
| 485 |
+
"learning_rate": 0.0001781971513578013,
|
| 486 |
+
"loss": 0.8059,
|
| 487 |
"step": 325
|
| 488 |
},
|
| 489 |
{
|
| 490 |
+
"epoch": 2.972972972972973,
|
| 491 |
+
"grad_norm": 1.1640625,
|
| 492 |
+
"learning_rate": 0.00017720744574600863,
|
| 493 |
+
"loss": 0.8273,
|
| 494 |
"step": 330
|
| 495 |
},
|
| 496 |
{
|
| 497 |
+
"epoch": 3.0,
|
| 498 |
+
"eval_loss": 2.1709225177764893,
|
| 499 |
+
"eval_runtime": 1.0054,
|
| 500 |
+
"eval_samples_per_second": 4.973,
|
| 501 |
+
"eval_steps_per_second": 1.989,
|
| 502 |
+
"step": 333
|
| 503 |
+
},
|
| 504 |
+
{
|
| 505 |
+
"epoch": 3.018018018018018,
|
| 506 |
+
"grad_norm": 0.83984375,
|
| 507 |
+
"learning_rate": 0.00017619865219625452,
|
| 508 |
+
"loss": 0.7934,
|
| 509 |
"step": 335
|
| 510 |
},
|
| 511 |
{
|
| 512 |
+
"epoch": 3.063063063063063,
|
| 513 |
+
"grad_norm": 1.5859375,
|
| 514 |
+
"learning_rate": 0.00017517102011179933,
|
| 515 |
+
"loss": 0.7096,
|
| 516 |
"step": 340
|
| 517 |
},
|
| 518 |
{
|
| 519 |
+
"epoch": 3.108108108108108,
|
| 520 |
+
"grad_norm": 0.91796875,
|
| 521 |
+
"learning_rate": 0.00017412480355334005,
|
| 522 |
+
"loss": 0.7203,
|
| 523 |
"step": 345
|
| 524 |
},
|
| 525 |
{
|
| 526 |
+
"epoch": 3.153153153153153,
|
| 527 |
+
"grad_norm": 1.9296875,
|
| 528 |
+
"learning_rate": 0.00017306026117619889,
|
| 529 |
+
"loss": 0.7237,
|
| 530 |
"step": 350
|
| 531 |
},
|
| 532 |
{
|
| 533 |
+
"epoch": 3.1981981981981984,
|
| 534 |
+
"grad_norm": 1.6328125,
|
| 535 |
+
"learning_rate": 0.00017197765616637636,
|
| 536 |
+
"loss": 0.738,
|
| 537 |
"step": 355
|
| 538 |
},
|
| 539 |
{
|
| 540 |
+
"epoch": 3.2432432432432434,
|
| 541 |
+
"grad_norm": 2.0625,
|
| 542 |
+
"learning_rate": 0.00017087725617548385,
|
| 543 |
+
"loss": 0.7214,
|
| 544 |
"step": 360
|
| 545 |
},
|
| 546 |
{
|
| 547 |
+
"epoch": 3.2882882882882885,
|
| 548 |
+
"grad_norm": 3.53125,
|
| 549 |
+
"learning_rate": 0.0001697593332545723,
|
| 550 |
+
"loss": 0.7549,
|
| 551 |
"step": 365
|
| 552 |
},
|
| 553 |
{
|
| 554 |
+
"epoch": 3.3333333333333335,
|
| 555 |
+
"grad_norm": 4.25,
|
| 556 |
+
"learning_rate": 0.0001686241637868734,
|
| 557 |
+
"loss": 0.7575,
|
| 558 |
"step": 370
|
| 559 |
},
|
| 560 |
{
|
| 561 |
+
"epoch": 3.3783783783783785,
|
| 562 |
+
"grad_norm": 1.6484375,
|
| 563 |
+
"learning_rate": 0.00016747202841946928,
|
| 564 |
+
"loss": 0.7392,
|
| 565 |
"step": 375
|
| 566 |
},
|
| 567 |
{
|
| 568 |
+
"epoch": 3.4234234234234235,
|
| 569 |
+
"grad_norm": 1.6171875,
|
| 570 |
+
"learning_rate": 0.00016630321199390867,
|
| 571 |
+
"loss": 0.7251,
|
| 572 |
"step": 380
|
| 573 |
},
|
| 574 |
{
|
| 575 |
+
"epoch": 3.4684684684684686,
|
| 576 |
+
"grad_norm": 1.5546875,
|
| 577 |
+
"learning_rate": 0.0001651180034757856,
|
| 578 |
+
"loss": 0.7285,
|
| 579 |
"step": 385
|
| 580 |
},
|
| 581 |
{
|
| 582 |
+
"epoch": 3.5135135135135136,
|
| 583 |
+
"grad_norm": 1.0078125,
|
| 584 |
+
"learning_rate": 0.0001639166958832985,
|
| 585 |
+
"loss": 0.7114,
|
| 586 |
"step": 390
|
| 587 |
},
|
| 588 |
{
|
| 589 |
+
"epoch": 3.5585585585585586,
|
| 590 |
+
"grad_norm": 1.2421875,
|
| 591 |
+
"learning_rate": 0.00016269958621480788,
|
| 592 |
+
"loss": 0.7223,
|
| 593 |
"step": 395
|
| 594 |
},
|
| 595 |
{
|
| 596 |
+
"epoch": 3.6036036036036037,
|
| 597 |
+
"grad_norm": 0.61328125,
|
| 598 |
+
"learning_rate": 0.00016146697537540924,
|
| 599 |
+
"loss": 0.7273,
|
| 600 |
"step": 400
|
| 601 |
},
|
| 602 |
{
|
| 603 |
+
"epoch": 3.6486486486486487,
|
| 604 |
+
"grad_norm": 0.67578125,
|
| 605 |
+
"learning_rate": 0.00016021916810254097,
|
| 606 |
+
"loss": 0.7328,
|
| 607 |
"step": 405
|
| 608 |
},
|
| 609 |
{
|
| 610 |
+
"epoch": 3.6936936936936937,
|
| 611 |
+
"grad_norm": 1.1015625,
|
| 612 |
+
"learning_rate": 0.00015895647289064396,
|
| 613 |
+
"loss": 0.7409,
|
| 614 |
"step": 410
|
| 615 |
},
|
| 616 |
{
|
| 617 |
+
"epoch": 3.7387387387387387,
|
| 618 |
+
"grad_norm": 0.7578125,
|
| 619 |
+
"learning_rate": 0.000157679201914893,
|
| 620 |
+
"loss": 0.7247,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 621 |
"step": 415
|
| 622 |
},
|
| 623 |
{
|
| 624 |
+
"epoch": 3.7837837837837838,
|
| 625 |
+
"grad_norm": 1.890625,
|
| 626 |
+
"learning_rate": 0.0001563876709540178,
|
| 627 |
+
"loss": 0.7446,
|
| 628 |
"step": 420
|
| 629 |
},
|
| 630 |
{
|
| 631 |
+
"epoch": 3.828828828828829,
|
| 632 |
+
"grad_norm": 0.7109375,
|
| 633 |
+
"learning_rate": 0.0001550821993122334,
|
| 634 |
+
"loss": 0.7421,
|
| 635 |
"step": 425
|
| 636 |
},
|
| 637 |
{
|
| 638 |
+
"epoch": 3.873873873873874,
|
| 639 |
+
"grad_norm": 0.73046875,
|
| 640 |
+
"learning_rate": 0.00015376310974029873,
|
| 641 |
+
"loss": 0.7362,
|
| 642 |
"step": 430
|
| 643 |
},
|
| 644 |
{
|
| 645 |
+
"epoch": 3.918918918918919,
|
| 646 |
+
"grad_norm": 0.66015625,
|
| 647 |
+
"learning_rate": 0.00015243072835572318,
|
| 648 |
+
"loss": 0.7398,
|
| 649 |
"step": 435
|
| 650 |
},
|
| 651 |
{
|
| 652 |
+
"epoch": 3.963963963963964,
|
| 653 |
+
"grad_norm": 0.69921875,
|
| 654 |
+
"learning_rate": 0.0001510853845621409,
|
| 655 |
+
"loss": 0.7586,
|
| 656 |
"step": 440
|
| 657 |
},
|
| 658 |
{
|
| 659 |
+
"epoch": 4.0,
|
| 660 |
+
"eval_loss": 2.2163968086242676,
|
| 661 |
+
"eval_runtime": 1.0061,
|
| 662 |
+
"eval_samples_per_second": 4.97,
|
| 663 |
+
"eval_steps_per_second": 1.988,
|
| 664 |
+
"step": 444
|
| 665 |
+
},
|
| 666 |
+
{
|
| 667 |
+
"epoch": 4.009009009009009,
|
| 668 |
+
"grad_norm": 0.58203125,
|
| 669 |
+
"learning_rate": 0.00014972741096787242,
|
| 670 |
+
"loss": 0.7128,
|
| 671 |
"step": 445
|
| 672 |
},
|
| 673 |
{
|
| 674 |
+
"epoch": 4.054054054054054,
|
| 675 |
+
"grad_norm": 0.75,
|
| 676 |
+
"learning_rate": 0.00014835714330369446,
|
| 677 |
+
"loss": 0.6463,
|
| 678 |
"step": 450
|
| 679 |
},
|
| 680 |
{
|
| 681 |
+
"epoch": 4.099099099099099,
|
| 682 |
+
"grad_norm": 0.83203125,
|
| 683 |
+
"learning_rate": 0.00014697492033983707,
|
| 684 |
+
"loss": 0.6453,
|
| 685 |
"step": 455
|
| 686 |
},
|
| 687 |
{
|
| 688 |
+
"epoch": 4.1441441441441444,
|
| 689 |
+
"grad_norm": 0.55859375,
|
| 690 |
+
"learning_rate": 0.00014558108380223012,
|
| 691 |
+
"loss": 0.647,
|
| 692 |
"step": 460
|
| 693 |
},
|
| 694 |
{
|
| 695 |
+
"epoch": 4.1891891891891895,
|
| 696 |
+
"grad_norm": 1.3125,
|
| 697 |
+
"learning_rate": 0.00014417597828801832,
|
| 698 |
+
"loss": 0.626,
|
| 699 |
"step": 465
|
| 700 |
},
|
| 701 |
{
|
| 702 |
+
"epoch": 4.2342342342342345,
|
| 703 |
+
"grad_norm": 0.85546875,
|
| 704 |
+
"learning_rate": 0.00014275995118036693,
|
| 705 |
+
"loss": 0.6334,
|
| 706 |
"step": 470
|
| 707 |
},
|
| 708 |
{
|
| 709 |
+
"epoch": 4.2792792792792795,
|
| 710 |
+
"grad_norm": 0.69921875,
|
| 711 |
+
"learning_rate": 0.0001413333525625784,
|
| 712 |
+
"loss": 0.6435,
|
| 713 |
"step": 475
|
| 714 |
},
|
| 715 |
{
|
| 716 |
+
"epoch": 4.324324324324325,
|
| 717 |
+
"grad_norm": 0.8046875,
|
| 718 |
+
"learning_rate": 0.00013989653513154165,
|
| 719 |
+
"loss": 0.6439,
|
| 720 |
"step": 480
|
| 721 |
},
|
| 722 |
{
|
| 723 |
+
"epoch": 4.36936936936937,
|
| 724 |
+
"grad_norm": 1.0859375,
|
| 725 |
+
"learning_rate": 0.00013844985411053492,
|
| 726 |
+
"loss": 0.6559,
|
| 727 |
"step": 485
|
| 728 |
},
|
| 729 |
{
|
| 730 |
+
"epoch": 4.414414414414415,
|
| 731 |
+
"grad_norm": 1.3359375,
|
| 732 |
+
"learning_rate": 0.00013699366716140435,
|
| 733 |
+
"loss": 0.6654,
|
| 734 |
"step": 490
|
| 735 |
},
|
| 736 |
{
|
| 737 |
+
"epoch": 4.45945945945946,
|
| 738 |
+
"grad_norm": 0.80859375,
|
| 739 |
+
"learning_rate": 0.00013552833429613938,
|
| 740 |
+
"loss": 0.6783,
|
| 741 |
"step": 495
|
| 742 |
},
|
| 743 |
{
|
| 744 |
+
"epoch": 4.504504504504505,
|
| 745 |
+
"grad_norm": 0.6875,
|
| 746 |
+
"learning_rate": 0.00013405421778786737,
|
| 747 |
+
"loss": 0.6543,
|
| 748 |
"step": 500
|
| 749 |
},
|
| 750 |
{
|
| 751 |
+
"epoch": 4.54954954954955,
|
| 752 |
+
"grad_norm": 0.62890625,
|
| 753 |
+
"learning_rate": 0.00013257168208128908,
|
| 754 |
+
"loss": 0.6608,
|
| 755 |
"step": 505
|
| 756 |
},
|
| 757 |
{
|
| 758 |
+
"epoch": 4.594594594594595,
|
| 759 |
+
"grad_norm": 0.60546875,
|
| 760 |
+
"learning_rate": 0.00013108109370257712,
|
| 761 |
+
"loss": 0.6621,
|
| 762 |
"step": 510
|
| 763 |
},
|
| 764 |
{
|
| 765 |
+
"epoch": 4.63963963963964,
|
| 766 |
+
"grad_norm": 0.67578125,
|
| 767 |
+
"learning_rate": 0.00012958282116876026,
|
| 768 |
+
"loss": 0.656,
|
| 769 |
"step": 515
|
| 770 |
},
|
| 771 |
{
|
| 772 |
+
"epoch": 4.684684684684685,
|
| 773 |
+
"grad_norm": 0.65234375,
|
| 774 |
+
"learning_rate": 0.00012807723489661495,
|
| 775 |
+
"loss": 0.6505,
|
| 776 |
"step": 520
|
| 777 |
},
|
| 778 |
{
|
| 779 |
+
"epoch": 4.72972972972973,
|
| 780 |
+
"grad_norm": 0.921875,
|
| 781 |
+
"learning_rate": 0.00012656470711108764,
|
| 782 |
+
"loss": 0.6789,
|
| 783 |
"step": 525
|
| 784 |
},
|
| 785 |
{
|
| 786 |
+
"epoch": 4.774774774774775,
|
| 787 |
+
"grad_norm": 0.61328125,
|
| 788 |
+
"learning_rate": 0.00012504561175326985,
|
| 789 |
+
"loss": 0.6588,
|
| 790 |
"step": 530
|
| 791 |
},
|
| 792 |
{
|
| 793 |
+
"epoch": 4.81981981981982,
|
| 794 |
+
"grad_norm": 0.703125,
|
| 795 |
+
"learning_rate": 0.00012352032438794902,
|
| 796 |
+
"loss": 0.6534,
|
| 797 |
"step": 535
|
| 798 |
},
|
| 799 |
{
|
| 800 |
+
"epoch": 4.864864864864865,
|
| 801 |
+
"grad_norm": 0.74609375,
|
| 802 |
+
"learning_rate": 0.00012198922211075778,
|
| 803 |
+
"loss": 0.6482,
|
| 804 |
"step": 540
|
| 805 |
},
|
| 806 |
{
|
| 807 |
+
"epoch": 4.90990990990991,
|
| 808 |
+
"grad_norm": 0.94140625,
|
| 809 |
+
"learning_rate": 0.00012045268345494511,
|
| 810 |
+
"loss": 0.6595,
|
| 811 |
"step": 545
|
| 812 |
},
|
| 813 |
{
|
| 814 |
+
"epoch": 4.954954954954955,
|
| 815 |
+
"grad_norm": 0.59765625,
|
| 816 |
+
"learning_rate": 0.00011891108829779165,
|
| 817 |
+
"loss": 0.6624,
|
| 818 |
+
"step": 550
|
|
|
|
| 819 |
},
|
| 820 |
{
|
| 821 |
+
"epoch": 5.0,
|
| 822 |
+
"grad_norm": 0.578125,
|
| 823 |
+
"learning_rate": 0.00011736481776669306,
|
| 824 |
+
"loss": 0.6613,
|
| 825 |
+
"step": 555
|
| 826 |
},
|
| 827 |
{
|
| 828 |
+
"epoch": 5.0,
|
| 829 |
+
"eval_loss": 2.3182225227355957,
|
| 830 |
+
"eval_runtime": 1.0028,
|
| 831 |
+
"eval_samples_per_second": 4.986,
|
| 832 |
+
"eval_steps_per_second": 1.994,
|
| 833 |
"step": 555
|
| 834 |
},
|
| 835 |
{
|
| 836 |
+
"epoch": 5.045045045045045,
|
| 837 |
+
"grad_norm": 0.98046875,
|
| 838 |
+
"learning_rate": 0.0001158142541449341,
|
| 839 |
+
"loss": 0.5564,
|
| 840 |
"step": 560
|
| 841 |
},
|
| 842 |
{
|
| 843 |
+
"epoch": 5.09009009009009,
|
| 844 |
+
"grad_norm": 0.69140625,
|
| 845 |
+
"learning_rate": 0.00011425978077717709,
|
| 846 |
+
"loss": 0.5273,
|
| 847 |
"step": 565
|
| 848 |
},
|
| 849 |
{
|
| 850 |
+
"epoch": 5.135135135135135,
|
| 851 |
+
"grad_norm": 0.69921875,
|
| 852 |
+
"learning_rate": 0.00011270178197468789,
|
| 853 |
+
"loss": 0.5589,
|
| 854 |
"step": 570
|
| 855 |
},
|
| 856 |
{
|
| 857 |
+
"epoch": 5.18018018018018,
|
| 858 |
+
"grad_norm": 0.6171875,
|
| 859 |
+
"learning_rate": 0.00011114064292032282,
|
| 860 |
+
"loss": 0.5593,
|
| 861 |
"step": 575
|
| 862 |
},
|
| 863 |
{
|
| 864 |
+
"epoch": 5.225225225225225,
|
| 865 |
+
"grad_norm": 0.69921875,
|
| 866 |
+
"learning_rate": 0.00010957674957330042,
|
| 867 |
+
"loss": 0.5672,
|
| 868 |
"step": 580
|
| 869 |
},
|
| 870 |
{
|
| 871 |
+
"epoch": 5.27027027027027,
|
| 872 |
+
"grad_norm": 0.69140625,
|
| 873 |
+
"learning_rate": 0.00010801048857378071,
|
| 874 |
+
"loss": 0.5444,
|
| 875 |
"step": 585
|
| 876 |
},
|
| 877 |
{
|
| 878 |
+
"epoch": 5.315315315315315,
|
| 879 |
+
"grad_norm": 0.66796875,
|
| 880 |
+
"learning_rate": 0.00010644224714727681,
|
| 881 |
+
"loss": 0.5747,
|
| 882 |
"step": 590
|
| 883 |
},
|
| 884 |
{
|
| 885 |
+
"epoch": 5.36036036036036,
|
| 886 |
+
"grad_norm": 0.68359375,
|
| 887 |
+
"learning_rate": 0.0001048724130089212,
|
| 888 |
+
"loss": 0.5609,
|
| 889 |
"step": 595
|
| 890 |
},
|
| 891 |
{
|
| 892 |
+
"epoch": 5.405405405405405,
|
| 893 |
+
"grad_norm": 0.8984375,
|
| 894 |
+
"learning_rate": 0.00010330137426761135,
|
| 895 |
+
"loss": 0.5625,
|
| 896 |
"step": 600
|
| 897 |
},
|
| 898 |
{
|
| 899 |
+
"epoch": 5.45045045045045,
|
| 900 |
+
"grad_norm": 0.76171875,
|
| 901 |
+
"learning_rate": 0.00010172951933005775,
|
| 902 |
+
"loss": 0.5671,
|
| 903 |
"step": 605
|
| 904 |
},
|
| 905 |
{
|
| 906 |
+
"epoch": 5.495495495495495,
|
| 907 |
+
"grad_norm": 0.80859375,
|
| 908 |
+
"learning_rate": 0.00010015723680475846,
|
| 909 |
+
"loss": 0.564,
|
| 910 |
"step": 610
|
| 911 |
},
|
| 912 |
{
|
| 913 |
+
"epoch": 5.54054054054054,
|
| 914 |
+
"grad_norm": 0.76171875,
|
| 915 |
+
"learning_rate": 9.858491540592382e-05,
|
| 916 |
+
"loss": 0.5784,
|
| 917 |
"step": 615
|
| 918 |
},
|
| 919 |
{
|
| 920 |
+
"epoch": 5.585585585585585,
|
| 921 |
+
"grad_norm": 0.7265625,
|
| 922 |
+
"learning_rate": 9.70129438573747e-05,
|
| 923 |
+
"loss": 0.5672,
|
| 924 |
"step": 620
|
| 925 |
},
|
| 926 |
{
|
| 927 |
+
"epoch": 5.63063063063063,
|
| 928 |
+
"grad_norm": 0.75390625,
|
| 929 |
+
"learning_rate": 9.54417107964389e-05,
|
| 930 |
+
"loss": 0.5592,
|
| 931 |
"step": 625
|
| 932 |
},
|
| 933 |
{
|
| 934 |
+
"epoch": 5.675675675675675,
|
| 935 |
+
"grad_norm": 0.734375,
|
| 936 |
+
"learning_rate": 9.38716046778684e-05,
|
| 937 |
+
"loss": 0.5634,
|
| 938 |
"step": 630
|
| 939 |
},
|
| 940 |
{
|
| 941 |
+
"epoch": 5.7207207207207205,
|
| 942 |
+
"grad_norm": 0.6640625,
|
| 943 |
+
"learning_rate": 9.230301367780208e-05,
|
| 944 |
+
"loss": 0.5691,
|
| 945 |
"step": 635
|
| 946 |
},
|
| 947 |
{
|
| 948 |
+
"epoch": 5.7657657657657655,
|
| 949 |
+
"grad_norm": 0.6875,
|
| 950 |
+
"learning_rate": 9.07363255977973e-05,
|
| 951 |
+
"loss": 0.5722,
|
| 952 |
"step": 640
|
| 953 |
},
|
| 954 |
{
|
| 955 |
+
"epoch": 5.8108108108108105,
|
| 956 |
+
"grad_norm": 0.76953125,
|
| 957 |
+
"learning_rate": 8.917192776895382e-05,
|
| 958 |
+
"loss": 0.5827,
|
| 959 |
"step": 645
|
| 960 |
},
|
| 961 |
{
|
| 962 |
+
"epoch": 5.8558558558558556,
|
| 963 |
+
"grad_norm": 0.83203125,
|
| 964 |
+
"learning_rate": 8.76102069561545e-05,
|
| 965 |
+
"loss": 0.5745,
|
| 966 |
"step": 650
|
| 967 |
},
|
| 968 |
{
|
| 969 |
+
"epoch": 5.900900900900901,
|
| 970 |
+
"grad_norm": 0.7265625,
|
| 971 |
+
"learning_rate": 8.605154926244543e-05,
|
| 972 |
+
"loss": 0.5614,
|
| 973 |
"step": 655
|
| 974 |
},
|
| 975 |
{
|
| 976 |
+
"epoch": 5.945945945945946,
|
| 977 |
+
"grad_norm": 0.65625,
|
| 978 |
+
"learning_rate": 8.449634003358022e-05,
|
| 979 |
+
"loss": 0.5731,
|
| 980 |
"step": 660
|
| 981 |
},
|
| 982 |
{
|
| 983 |
+
"epoch": 5.990990990990991,
|
| 984 |
+
"grad_norm": 0.8828125,
|
| 985 |
+
"learning_rate": 8.294496376275104e-05,
|
| 986 |
+
"loss": 0.577,
|
| 987 |
"step": 665
|
| 988 |
},
|
| 989 |
{
|
| 990 |
+
"epoch": 6.0,
|
| 991 |
+
"eval_loss": 2.4773526191711426,
|
| 992 |
+
"eval_runtime": 1.0034,
|
| 993 |
+
"eval_samples_per_second": 4.983,
|
| 994 |
+
"eval_steps_per_second": 1.993,
|
| 995 |
+
"step": 666
|
| 996 |
+
},
|
| 997 |
+
{
|
| 998 |
+
"epoch": 6.036036036036036,
|
| 999 |
+
"grad_norm": 0.8984375,
|
| 1000 |
+
"learning_rate": 8.13978039955308e-05,
|
| 1001 |
+
"loss": 0.5142,
|
| 1002 |
"step": 670
|
| 1003 |
},
|
| 1004 |
{
|
| 1005 |
+
"epoch": 6.081081081081081,
|
| 1006 |
+
"grad_norm": 0.8359375,
|
| 1007 |
+
"learning_rate": 7.985524323504948e-05,
|
| 1008 |
+
"loss": 0.4725,
|
| 1009 |
"step": 675
|
| 1010 |
},
|
| 1011 |
{
|
| 1012 |
+
"epoch": 6.126126126126126,
|
| 1013 |
+
"grad_norm": 0.7734375,
|
| 1014 |
+
"learning_rate": 7.831766284742807e-05,
|
| 1015 |
+
"loss": 0.4671,
|
| 1016 |
"step": 680
|
| 1017 |
},
|
| 1018 |
{
|
| 1019 |
+
"epoch": 6.171171171171171,
|
| 1020 |
+
"grad_norm": 0.7578125,
|
| 1021 |
+
"learning_rate": 7.678544296749384e-05,
|
| 1022 |
+
"loss": 0.4804,
|
| 1023 |
"step": 685
|
| 1024 |
},
|
| 1025 |
{
|
| 1026 |
+
"epoch": 6.216216216216216,
|
| 1027 |
+
"grad_norm": 0.82421875,
|
| 1028 |
+
"learning_rate": 7.525896240479976e-05,
|
| 1029 |
+
"loss": 0.4704,
|
| 1030 |
+
"step": 690
|
|
|
|
| 1031 |
},
|
| 1032 |
{
|
| 1033 |
+
"epoch": 6.261261261261261,
|
| 1034 |
+
"grad_norm": 0.75,
|
| 1035 |
+
"learning_rate": 7.37385985499718e-05,
|
| 1036 |
+
"loss": 0.4659,
|
| 1037 |
+
"step": 695
|
| 1038 |
+
},
|
| 1039 |
+
{
|
| 1040 |
+
"epoch": 6.306306306306306,
|
| 1041 |
+
"grad_norm": 0.71484375,
|
| 1042 |
+
"learning_rate": 7.222472728140695e-05,
|
| 1043 |
+
"loss": 0.4697,
|
| 1044 |
+
"step": 700
|
| 1045 |
+
},
|
| 1046 |
+
{
|
| 1047 |
+
"epoch": 6.351351351351352,
|
| 1048 |
+
"grad_norm": 0.79296875,
|
| 1049 |
+
"learning_rate": 7.071772287234497e-05,
|
| 1050 |
+
"loss": 0.4912,
|
| 1051 |
+
"step": 705
|
| 1052 |
+
},
|
| 1053 |
+
{
|
| 1054 |
+
"epoch": 6.396396396396397,
|
| 1055 |
+
"grad_norm": 0.76953125,
|
| 1056 |
+
"learning_rate": 6.921795789833723e-05,
|
| 1057 |
+
"loss": 0.4689,
|
| 1058 |
+
"step": 710
|
| 1059 |
+
},
|
| 1060 |
+
{
|
| 1061 |
+
"epoch": 6.441441441441442,
|
| 1062 |
+
"grad_norm": 0.66796875,
|
| 1063 |
+
"learning_rate": 6.772580314513508e-05,
|
| 1064 |
+
"loss": 0.4753,
|
| 1065 |
+
"step": 715
|
| 1066 |
+
},
|
| 1067 |
+
{
|
| 1068 |
+
"epoch": 6.486486486486487,
|
| 1069 |
+
"grad_norm": 0.75,
|
| 1070 |
+
"learning_rate": 6.624162751702076e-05,
|
| 1071 |
+
"loss": 0.4759,
|
| 1072 |
+
"step": 720
|
| 1073 |
+
},
|
| 1074 |
+
{
|
| 1075 |
+
"epoch": 6.531531531531532,
|
| 1076 |
+
"grad_norm": 0.70703125,
|
| 1077 |
+
"learning_rate": 6.476579794560356e-05,
|
| 1078 |
+
"loss": 0.489,
|
| 1079 |
+
"step": 725
|
| 1080 |
+
},
|
| 1081 |
+
{
|
| 1082 |
+
"epoch": 6.576576576576577,
|
| 1083 |
+
"grad_norm": 0.7265625,
|
| 1084 |
+
"learning_rate": 6.329867929910347e-05,
|
| 1085 |
+
"loss": 0.473,
|
| 1086 |
+
"step": 730
|
| 1087 |
+
},
|
| 1088 |
+
{
|
| 1089 |
+
"epoch": 6.621621621621622,
|
| 1090 |
+
"grad_norm": 0.7109375,
|
| 1091 |
+
"learning_rate": 6.184063429214515e-05,
|
| 1092 |
+
"loss": 0.4793,
|
| 1093 |
+
"step": 735
|
| 1094 |
+
},
|
| 1095 |
+
{
|
| 1096 |
+
"epoch": 6.666666666666667,
|
| 1097 |
+
"grad_norm": 0.76171875,
|
| 1098 |
+
"learning_rate": 6.039202339608432e-05,
|
| 1099 |
+
"loss": 0.5071,
|
| 1100 |
+
"step": 740
|
| 1101 |
+
},
|
| 1102 |
+
{
|
| 1103 |
+
"epoch": 6.711711711711712,
|
| 1104 |
+
"grad_norm": 0.69921875,
|
| 1105 |
+
"learning_rate": 5.895320474988864e-05,
|
| 1106 |
+
"loss": 0.4741,
|
| 1107 |
+
"step": 745
|
| 1108 |
+
},
|
| 1109 |
+
{
|
| 1110 |
+
"epoch": 6.756756756756757,
|
| 1111 |
+
"grad_norm": 0.69921875,
|
| 1112 |
+
"learning_rate": 5.752453407159522e-05,
|
| 1113 |
+
"loss": 0.4799,
|
| 1114 |
+
"step": 750
|
| 1115 |
+
},
|
| 1116 |
+
{
|
| 1117 |
+
"epoch": 6.801801801801802,
|
| 1118 |
+
"grad_norm": 0.7578125,
|
| 1119 |
+
"learning_rate": 5.610636457036693e-05,
|
| 1120 |
+
"loss": 0.4901,
|
| 1121 |
+
"step": 755
|
| 1122 |
+
},
|
| 1123 |
+
{
|
| 1124 |
+
"epoch": 6.846846846846847,
|
| 1125 |
+
"grad_norm": 0.6953125,
|
| 1126 |
+
"learning_rate": 5.469904685916861e-05,
|
| 1127 |
+
"loss": 0.4858,
|
| 1128 |
+
"step": 760
|
| 1129 |
+
},
|
| 1130 |
+
{
|
| 1131 |
+
"epoch": 6.891891891891892,
|
| 1132 |
+
"grad_norm": 0.76953125,
|
| 1133 |
+
"learning_rate": 5.33029288680852e-05,
|
| 1134 |
+
"loss": 0.4895,
|
| 1135 |
+
"step": 765
|
| 1136 |
+
},
|
| 1137 |
+
{
|
| 1138 |
+
"epoch": 6.936936936936937,
|
| 1139 |
+
"grad_norm": 0.70703125,
|
| 1140 |
+
"learning_rate": 5.191835575830352e-05,
|
| 1141 |
+
"loss": 0.4935,
|
| 1142 |
+
"step": 770
|
| 1143 |
+
},
|
| 1144 |
+
{
|
| 1145 |
+
"epoch": 6.981981981981982,
|
| 1146 |
+
"grad_norm": 0.69921875,
|
| 1147 |
+
"learning_rate": 5.0545669836778144e-05,
|
| 1148 |
+
"loss": 0.4958,
|
| 1149 |
+
"step": 775
|
| 1150 |
+
},
|
| 1151 |
+
{
|
| 1152 |
+
"epoch": 7.0,
|
| 1153 |
+
"eval_loss": 2.7035882472991943,
|
| 1154 |
+
"eval_runtime": 1.0058,
|
| 1155 |
+
"eval_samples_per_second": 4.971,
|
| 1156 |
+
"eval_steps_per_second": 1.988,
|
| 1157 |
+
"step": 777
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 7.027027027027027,
|
| 1161 |
+
"grad_norm": 0.6875,
|
| 1162 |
+
"learning_rate": 4.918521047160308e-05,
|
| 1163 |
+
"loss": 0.4443,
|
| 1164 |
+
"step": 780
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 7.072072072072072,
|
| 1168 |
+
"grad_norm": 0.7734375,
|
| 1169 |
+
"learning_rate": 4.783731400811022e-05,
|
| 1170 |
+
"loss": 0.4139,
|
| 1171 |
+
"step": 785
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 7.117117117117117,
|
| 1175 |
+
"grad_norm": 0.734375,
|
| 1176 |
+
"learning_rate": 4.650231368571486e-05,
|
| 1177 |
+
"loss": 0.41,
|
| 1178 |
+
"step": 790
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 7.162162162162162,
|
| 1182 |
+
"grad_norm": 0.90625,
|
| 1183 |
+
"learning_rate": 4.518053955552903e-05,
|
| 1184 |
+
"loss": 0.4291,
|
| 1185 |
+
"step": 795
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 7.207207207207207,
|
| 1189 |
+
"grad_norm": 0.71875,
|
| 1190 |
+
"learning_rate": 4.387231839876349e-05,
|
| 1191 |
+
"loss": 0.4141,
|
| 1192 |
+
"step": 800
|
| 1193 |
+
},
|
| 1194 |
+
{
|
| 1195 |
+
"epoch": 7.252252252252252,
|
| 1196 |
+
"grad_norm": 0.7265625,
|
| 1197 |
+
"learning_rate": 4.2577973645937674e-05,
|
| 1198 |
+
"loss": 0.4139,
|
| 1199 |
+
"step": 805
|
| 1200 |
+
},
|
| 1201 |
+
{
|
| 1202 |
+
"epoch": 7.297297297297297,
|
| 1203 |
+
"grad_norm": 0.76171875,
|
| 1204 |
+
"learning_rate": 4.129782529691815e-05,
|
| 1205 |
+
"loss": 0.4278,
|
| 1206 |
+
"step": 810
|
| 1207 |
+
},
|
| 1208 |
+
{
|
| 1209 |
+
"epoch": 7.342342342342342,
|
| 1210 |
+
"grad_norm": 0.73046875,
|
| 1211 |
+
"learning_rate": 4.003218984180552e-05,
|
| 1212 |
+
"loss": 0.4148,
|
| 1213 |
+
"step": 815
|
| 1214 |
+
},
|
| 1215 |
+
{
|
| 1216 |
+
"epoch": 7.387387387387387,
|
| 1217 |
+
"grad_norm": 0.79296875,
|
| 1218 |
+
"learning_rate": 3.878138018268866e-05,
|
| 1219 |
+
"loss": 0.4168,
|
| 1220 |
+
"step": 820
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 7.4324324324324325,
|
| 1224 |
+
"grad_norm": 0.82421875,
|
| 1225 |
+
"learning_rate": 3.7545705556286126e-05,
|
| 1226 |
+
"loss": 0.4182,
|
| 1227 |
+
"step": 825
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"epoch": 7.4774774774774775,
|
| 1231 |
+
"grad_norm": 0.70703125,
|
| 1232 |
+
"learning_rate": 3.632547145749395e-05,
|
| 1233 |
+
"loss": 0.4239,
|
| 1234 |
+
"step": 830
|
| 1235 |
+
},
|
| 1236 |
+
{
|
| 1237 |
+
"epoch": 7.5225225225225225,
|
| 1238 |
+
"grad_norm": 0.78515625,
|
| 1239 |
+
"learning_rate": 3.5120979563858266e-05,
|
| 1240 |
+
"loss": 0.4137,
|
| 1241 |
+
"step": 835
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 7.5675675675675675,
|
| 1245 |
+
"grad_norm": 0.73828125,
|
| 1246 |
+
"learning_rate": 3.393252766099187e-05,
|
| 1247 |
+
"loss": 0.4111,
|
| 1248 |
+
"step": 840
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 7.612612612612613,
|
| 1252 |
+
"grad_norm": 0.7421875,
|
| 1253 |
+
"learning_rate": 3.2760409568952766e-05,
|
| 1254 |
+
"loss": 0.4179,
|
| 1255 |
+
"step": 845
|
| 1256 |
+
},
|
| 1257 |
+
{
|
| 1258 |
+
"epoch": 7.657657657657658,
|
| 1259 |
+
"grad_norm": 0.76171875,
|
| 1260 |
+
"learning_rate": 3.1604915069603436e-05,
|
| 1261 |
+
"loss": 0.429,
|
| 1262 |
+
"step": 850
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 7.702702702702703,
|
| 1266 |
+
"grad_norm": 0.75,
|
| 1267 |
+
"learning_rate": 3.0466329834968233e-05,
|
| 1268 |
+
"loss": 0.4118,
|
| 1269 |
+
"step": 855
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 7.747747747747748,
|
| 1273 |
+
"grad_norm": 0.71484375,
|
| 1274 |
+
"learning_rate": 2.9344935356606773e-05,
|
| 1275 |
+
"loss": 0.4049,
|
| 1276 |
+
"step": 860
|
| 1277 |
+
},
|
| 1278 |
+
{
|
| 1279 |
+
"epoch": 7.792792792792793,
|
| 1280 |
+
"grad_norm": 0.74609375,
|
| 1281 |
+
"learning_rate": 2.8241008876021215e-05,
|
| 1282 |
+
"loss": 0.413,
|
| 1283 |
+
"step": 865
|
| 1284 |
+
},
|
| 1285 |
+
{
|
| 1286 |
+
"epoch": 7.837837837837838,
|
| 1287 |
+
"grad_norm": 0.72265625,
|
| 1288 |
+
"learning_rate": 2.7154823316113932e-05,
|
| 1289 |
+
"loss": 0.4071,
|
| 1290 |
+
"step": 870
|
| 1291 |
+
},
|
| 1292 |
+
{
|
| 1293 |
+
"epoch": 7.882882882882883,
|
| 1294 |
+
"grad_norm": 0.734375,
|
| 1295 |
+
"learning_rate": 2.60866472137129e-05,
|
| 1296 |
+
"loss": 0.4073,
|
| 1297 |
+
"step": 875
|
| 1298 |
+
},
|
| 1299 |
+
{
|
| 1300 |
+
"epoch": 7.927927927927928,
|
| 1301 |
+
"grad_norm": 0.71875,
|
| 1302 |
+
"learning_rate": 2.5036744653181753e-05,
|
| 1303 |
+
"loss": 0.4124,
|
| 1304 |
+
"step": 880
|
| 1305 |
+
},
|
| 1306 |
+
{
|
| 1307 |
+
"epoch": 7.972972972972973,
|
| 1308 |
+
"grad_norm": 0.7578125,
|
| 1309 |
+
"learning_rate": 2.4005375201130274e-05,
|
| 1310 |
+
"loss": 0.4205,
|
| 1311 |
+
"step": 885
|
| 1312 |
+
},
|
| 1313 |
+
{
|
| 1314 |
+
"epoch": 8.0,
|
| 1315 |
+
"eval_loss": 2.9689488410949707,
|
| 1316 |
+
"eval_runtime": 1.0053,
|
| 1317 |
+
"eval_samples_per_second": 4.973,
|
| 1318 |
+
"eval_steps_per_second": 1.989,
|
| 1319 |
+
"step": 888
|
| 1320 |
+
},
|
| 1321 |
+
{
|
| 1322 |
+
"epoch": 8.018018018018019,
|
| 1323 |
+
"grad_norm": 0.68359375,
|
| 1324 |
+
"learning_rate": 2.29927938422419e-05,
|
| 1325 |
+
"loss": 0.4012,
|
| 1326 |
+
"step": 890
|
| 1327 |
+
},
|
| 1328 |
+
{
|
| 1329 |
+
"epoch": 8.063063063063064,
|
| 1330 |
+
"grad_norm": 1.0078125,
|
| 1331 |
+
"learning_rate": 2.199925091623418e-05,
|
| 1332 |
+
"loss": 0.3781,
|
| 1333 |
+
"step": 895
|
| 1334 |
+
},
|
| 1335 |
+
{
|
| 1336 |
+
"epoch": 8.108108108108109,
|
| 1337 |
+
"grad_norm": 0.8671875,
|
| 1338 |
+
"learning_rate": 2.102499205596743e-05,
|
| 1339 |
+
"loss": 0.3809,
|
| 1340 |
+
"step": 900
|
| 1341 |
+
},
|
| 1342 |
+
{
|
| 1343 |
+
"epoch": 8.153153153153154,
|
| 1344 |
+
"grad_norm": 0.70703125,
|
| 1345 |
+
"learning_rate": 2.0070258126717e-05,
|
| 1346 |
+
"loss": 0.3699,
|
| 1347 |
+
"step": 905
|
| 1348 |
+
},
|
| 1349 |
+
{
|
| 1350 |
+
"epoch": 8.198198198198199,
|
| 1351 |
+
"grad_norm": 0.6875,
|
| 1352 |
+
"learning_rate": 1.913528516662452e-05,
|
| 1353 |
+
"loss": 0.3742,
|
| 1354 |
+
"step": 910
|
| 1355 |
+
},
|
| 1356 |
+
{
|
| 1357 |
+
"epoch": 8.243243243243244,
|
| 1358 |
+
"grad_norm": 0.70703125,
|
| 1359 |
+
"learning_rate": 1.8220304328342252e-05,
|
| 1360 |
+
"loss": 0.378,
|
| 1361 |
+
"step": 915
|
| 1362 |
+
},
|
| 1363 |
+
{
|
| 1364 |
+
"epoch": 8.288288288288289,
|
| 1365 |
+
"grad_norm": 0.70703125,
|
| 1366 |
+
"learning_rate": 1.7325541821885384e-05,
|
| 1367 |
+
"loss": 0.3842,
|
| 1368 |
+
"step": 920
|
| 1369 |
+
},
|
| 1370 |
+
{
|
| 1371 |
+
"epoch": 8.333333333333334,
|
| 1372 |
+
"grad_norm": 0.75390625,
|
| 1373 |
+
"learning_rate": 1.6451218858706374e-05,
|
| 1374 |
+
"loss": 0.3894,
|
| 1375 |
+
"step": 925
|
| 1376 |
+
},
|
| 1377 |
+
{
|
| 1378 |
+
"epoch": 8.378378378378379,
|
| 1379 |
+
"grad_norm": 0.71875,
|
| 1380 |
+
"learning_rate": 1.5597551597004966e-05,
|
| 1381 |
+
"loss": 0.3758,
|
| 1382 |
+
"step": 930
|
| 1383 |
+
},
|
| 1384 |
+
{
|
| 1385 |
+
"epoch": 8.423423423423424,
|
| 1386 |
+
"grad_norm": 0.671875,
|
| 1387 |
+
"learning_rate": 1.476475108828762e-05,
|
| 1388 |
+
"loss": 0.3717,
|
| 1389 |
+
"step": 935
|
| 1390 |
+
},
|
| 1391 |
+
{
|
| 1392 |
+
"epoch": 8.468468468468469,
|
| 1393 |
+
"grad_norm": 0.703125,
|
| 1394 |
+
"learning_rate": 1.3953023225189243e-05,
|
| 1395 |
+
"loss": 0.3771,
|
| 1396 |
+
"step": 940
|
| 1397 |
+
},
|
| 1398 |
+
{
|
| 1399 |
+
"epoch": 8.513513513513514,
|
| 1400 |
+
"grad_norm": 0.71875,
|
| 1401 |
+
"learning_rate": 1.3162568690570743e-05,
|
| 1402 |
+
"loss": 0.3759,
|
| 1403 |
+
"step": 945
|
| 1404 |
+
},
|
| 1405 |
+
{
|
| 1406 |
+
"epoch": 8.558558558558559,
|
| 1407 |
+
"grad_norm": 0.74609375,
|
| 1408 |
+
"learning_rate": 1.23935829079042e-05,
|
| 1409 |
+
"loss": 0.3786,
|
| 1410 |
+
"step": 950
|
| 1411 |
+
},
|
| 1412 |
+
{
|
| 1413 |
+
"epoch": 8.603603603603604,
|
| 1414 |
+
"grad_norm": 0.7109375,
|
| 1415 |
+
"learning_rate": 1.1646255992958466e-05,
|
| 1416 |
+
"loss": 0.3734,
|
| 1417 |
+
"step": 955
|
| 1418 |
+
},
|
| 1419 |
+
{
|
| 1420 |
+
"epoch": 8.64864864864865,
|
| 1421 |
+
"grad_norm": 0.7265625,
|
| 1422 |
+
"learning_rate": 1.0920772706797167e-05,
|
| 1423 |
+
"loss": 0.3809,
|
| 1424 |
+
"step": 960
|
| 1425 |
+
},
|
| 1426 |
+
{
|
| 1427 |
+
"epoch": 8.693693693693694,
|
| 1428 |
+
"grad_norm": 0.7109375,
|
| 1429 |
+
"learning_rate": 1.0217312410100089e-05,
|
| 1430 |
+
"loss": 0.3767,
|
| 1431 |
+
"step": 965
|
| 1432 |
+
},
|
| 1433 |
+
{
|
| 1434 |
+
"epoch": 8.73873873873874,
|
| 1435 |
+
"grad_norm": 0.68359375,
|
| 1436 |
+
"learning_rate": 9.536049018820192e-06,
|
| 1437 |
+
"loss": 0.3786,
|
| 1438 |
+
"step": 970
|
| 1439 |
+
},
|
| 1440 |
+
{
|
| 1441 |
+
"epoch": 8.783783783783784,
|
| 1442 |
+
"grad_norm": 0.71875,
|
| 1443 |
+
"learning_rate": 8.87715096118642e-06,
|
| 1444 |
+
"loss": 0.3786,
|
| 1445 |
+
"step": 975
|
| 1446 |
+
},
|
| 1447 |
+
{
|
| 1448 |
+
"epoch": 8.82882882882883,
|
| 1449 |
+
"grad_norm": 0.74609375,
|
| 1450 |
+
"learning_rate": 8.240781136063346e-06,
|
| 1451 |
+
"loss": 0.3868,
|
| 1452 |
+
"step": 980
|
| 1453 |
+
},
|
| 1454 |
+
{
|
| 1455 |
+
"epoch": 8.873873873873874,
|
| 1456 |
+
"grad_norm": 0.72265625,
|
| 1457 |
+
"learning_rate": 7.6270968726777414e-06,
|
| 1458 |
+
"loss": 0.3767,
|
| 1459 |
+
"step": 985
|
| 1460 |
+
},
|
| 1461 |
+
{
|
| 1462 |
+
"epoch": 8.91891891891892,
|
| 1463 |
+
"grad_norm": 0.7578125,
|
| 1464 |
+
"learning_rate": 7.03624989172228e-06,
|
| 1465 |
+
"loss": 0.3791,
|
| 1466 |
+
"step": 990
|
| 1467 |
+
},
|
| 1468 |
+
{
|
| 1469 |
+
"epoch": 8.963963963963964,
|
| 1470 |
+
"grad_norm": 0.71875,
|
| 1471 |
+
"learning_rate": 6.468386267845717e-06,
|
| 1472 |
+
"loss": 0.382,
|
| 1473 |
+
"step": 995
|
| 1474 |
+
},
|
| 1475 |
+
{
|
| 1476 |
+
"epoch": 9.0,
|
| 1477 |
+
"eval_loss": 3.2251663208007812,
|
| 1478 |
+
"eval_runtime": 1.0069,
|
| 1479 |
+
"eval_samples_per_second": 4.966,
|
| 1480 |
+
"eval_steps_per_second": 1.986,
|
| 1481 |
+
"step": 999
|
| 1482 |
+
},
|
| 1483 |
+
{
|
| 1484 |
+
"epoch": 9.00900900900901,
|
| 1485 |
+
"grad_norm": 0.71484375,
|
| 1486 |
+
"learning_rate": 5.9236463935389065e-06,
|
| 1487 |
+
"loss": 0.3794,
|
| 1488 |
+
"step": 1000
|
| 1489 |
+
},
|
| 1490 |
+
{
|
| 1491 |
+
"epoch": 9.054054054054054,
|
| 1492 |
+
"grad_norm": 0.671875,
|
| 1493 |
+
"learning_rate": 5.402164944425758e-06,
|
| 1494 |
+
"loss": 0.3777,
|
| 1495 |
+
"step": 1005
|
| 1496 |
+
},
|
| 1497 |
+
{
|
| 1498 |
+
"epoch": 9.0990990990991,
|
| 1499 |
+
"grad_norm": 0.734375,
|
| 1500 |
+
"learning_rate": 4.904070845967468e-06,
|
| 1501 |
+
"loss": 0.3779,
|
| 1502 |
+
"step": 1010
|
| 1503 |
+
},
|
| 1504 |
+
{
|
| 1505 |
+
"epoch": 9.144144144144144,
|
| 1506 |
+
"grad_norm": 0.703125,
|
| 1507 |
+
"learning_rate": 4.429487241588304e-06,
|
| 1508 |
+
"loss": 0.3744,
|
| 1509 |
+
"step": 1015
|
| 1510 |
+
},
|
| 1511 |
+
{
|
| 1512 |
+
"epoch": 9.18918918918919,
|
| 1513 |
+
"grad_norm": 0.6953125,
|
| 1514 |
+
"learning_rate": 3.9785314622310495e-06,
|
| 1515 |
+
"loss": 0.3694,
|
| 1516 |
+
"step": 1020
|
| 1517 |
+
},
|
| 1518 |
+
{
|
| 1519 |
+
"epoch": 9.234234234234235,
|
| 1520 |
+
"grad_norm": 0.71875,
|
| 1521 |
+
"learning_rate": 3.5513149973492976e-06,
|
| 1522 |
+
"loss": 0.3751,
|
| 1523 |
+
"step": 1025
|
| 1524 |
+
},
|
| 1525 |
+
{
|
| 1526 |
+
"epoch": 9.27927927927928,
|
| 1527 |
+
"grad_norm": 0.7265625,
|
| 1528 |
+
"learning_rate": 3.1479434673440167e-06,
|
| 1529 |
+
"loss": 0.3685,
|
| 1530 |
+
"step": 1030
|
| 1531 |
+
},
|
| 1532 |
+
{
|
| 1533 |
+
"epoch": 9.324324324324325,
|
| 1534 |
+
"grad_norm": 0.68359375,
|
| 1535 |
+
"learning_rate": 2.7685165974510986e-06,
|
| 1536 |
+
"loss": 0.3653,
|
| 1537 |
+
"step": 1035
|
| 1538 |
+
},
|
| 1539 |
+
{
|
| 1540 |
+
"epoch": 9.36936936936937,
|
| 1541 |
+
"grad_norm": 0.71484375,
|
| 1542 |
+
"learning_rate": 2.4131281930864002e-06,
|
| 1543 |
+
"loss": 0.3728,
|
| 1544 |
+
"step": 1040
|
| 1545 |
+
},
|
| 1546 |
+
{
|
| 1547 |
+
"epoch": 9.414414414414415,
|
| 1548 |
+
"grad_norm": 0.734375,
|
| 1549 |
+
"learning_rate": 2.0818661166542074e-06,
|
| 1550 |
+
"loss": 0.3693,
|
| 1551 |
+
"step": 1045
|
| 1552 |
+
},
|
| 1553 |
+
{
|
| 1554 |
+
"epoch": 9.45945945945946,
|
| 1555 |
+
"grad_norm": 0.6875,
|
| 1556 |
+
"learning_rate": 1.7748122658251876e-06,
|
| 1557 |
+
"loss": 0.3764,
|
| 1558 |
+
"step": 1050
|
| 1559 |
+
},
|
| 1560 |
+
{
|
| 1561 |
+
"epoch": 9.504504504504505,
|
| 1562 |
+
"grad_norm": 0.7265625,
|
| 1563 |
+
"learning_rate": 1.4920425532888526e-06,
|
| 1564 |
+
"loss": 0.3654,
|
| 1565 |
+
"step": 1055
|
| 1566 |
+
},
|
| 1567 |
+
{
|
| 1568 |
+
"epoch": 9.54954954954955,
|
| 1569 |
+
"grad_norm": 0.66796875,
|
| 1570 |
+
"learning_rate": 1.2336268879856727e-06,
|
| 1571 |
+
"loss": 0.3747,
|
| 1572 |
+
"step": 1060
|
| 1573 |
+
},
|
| 1574 |
+
{
|
| 1575 |
+
"epoch": 9.594594594594595,
|
| 1576 |
+
"grad_norm": 0.69140625,
|
| 1577 |
+
"learning_rate": 9.996291578236228e-07,
|
| 1578 |
+
"loss": 0.3711,
|
| 1579 |
+
"step": 1065
|
| 1580 |
+
},
|
| 1581 |
+
{
|
| 1582 |
+
"epoch": 9.63963963963964,
|
| 1583 |
+
"grad_norm": 0.71484375,
|
| 1584 |
+
"learning_rate": 7.901072138831511e-07,
|
| 1585 |
+
"loss": 0.3722,
|
| 1586 |
+
"step": 1070
|
| 1587 |
+
},
|
| 1588 |
+
{
|
| 1589 |
+
"epoch": 9.684684684684685,
|
| 1590 |
+
"grad_norm": 0.7109375,
|
| 1591 |
+
"learning_rate": 6.051128561147756e-07,
|
| 1592 |
+
"loss": 0.3612,
|
| 1593 |
+
"step": 1075
|
| 1594 |
+
},
|
| 1595 |
+
{
|
| 1596 |
+
"epoch": 9.72972972972973,
|
| 1597 |
+
"grad_norm": 0.74609375,
|
| 1598 |
+
"learning_rate": 4.44691820532539e-07,
|
| 1599 |
+
"loss": 0.3647,
|
| 1600 |
+
"step": 1080
|
| 1601 |
+
},
|
| 1602 |
+
{
|
| 1603 |
+
"epoch": 9.774774774774775,
|
| 1604 |
+
"grad_norm": 0.6875,
|
| 1605 |
+
"learning_rate": 3.0888376790679795e-07,
|
| 1606 |
+
"loss": 0.3672,
|
| 1607 |
+
"step": 1085
|
| 1608 |
+
},
|
| 1609 |
+
{
|
| 1610 |
+
"epoch": 9.81981981981982,
|
| 1611 |
+
"grad_norm": 0.6484375,
|
| 1612 |
+
"learning_rate": 1.977222739588891e-07,
|
| 1613 |
+
"loss": 0.3659,
|
| 1614 |
+
"step": 1090
|
| 1615 |
+
},
|
| 1616 |
+
{
|
| 1617 |
+
"epoch": 9.864864864864865,
|
| 1618 |
+
"grad_norm": 0.67578125,
|
| 1619 |
+
"learning_rate": 1.1123482106021322e-07,
|
| 1620 |
+
"loss": 0.3692,
|
| 1621 |
+
"step": 1095
|
| 1622 |
+
},
|
| 1623 |
+
{
|
| 1624 |
+
"epoch": 9.90990990990991,
|
| 1625 |
+
"grad_norm": 0.6875,
|
| 1626 |
+
"learning_rate": 4.9442791437848136e-08,
|
| 1627 |
+
"loss": 0.3663,
|
| 1628 |
+
"step": 1100
|
| 1629 |
+
},
|
| 1630 |
+
{
|
| 1631 |
+
"epoch": 9.954954954954955,
|
| 1632 |
+
"grad_norm": 0.6875,
|
| 1633 |
+
"learning_rate": 1.2361461888166226e-08,
|
| 1634 |
+
"loss": 0.3673,
|
| 1635 |
+
"step": 1105
|
| 1636 |
+
},
|
| 1637 |
+
{
|
| 1638 |
+
"epoch": 10.0,
|
| 1639 |
+
"grad_norm": 0.6640625,
|
| 1640 |
+
"learning_rate": 0.0,
|
| 1641 |
+
"loss": 0.372,
|
| 1642 |
+
"step": 1110
|
| 1643 |
+
},
|
| 1644 |
+
{
|
| 1645 |
+
"epoch": 10.0,
|
| 1646 |
+
"eval_loss": 3.242992401123047,
|
| 1647 |
+
"eval_runtime": 1.0031,
|
| 1648 |
+
"eval_samples_per_second": 4.984,
|
| 1649 |
+
"eval_steps_per_second": 1.994,
|
| 1650 |
+
"step": 1110
|
| 1651 |
+
},
|
| 1652 |
+
{
|
| 1653 |
+
"epoch": 10.0,
|
| 1654 |
+
"step": 1110,
|
| 1655 |
+
"total_flos": 1.697049221804327e+18,
|
| 1656 |
+
"train_loss": 1.8630313719715084,
|
| 1657 |
+
"train_runtime": 9058.6901,
|
| 1658 |
+
"train_samples_per_second": 1.957,
|
| 1659 |
+
"train_steps_per_second": 0.123
|
| 1660 |
}
|
| 1661 |
],
|
| 1662 |
"logging_steps": 5,
|
| 1663 |
+
"max_steps": 1110,
|
| 1664 |
"num_input_tokens_seen": 0,
|
| 1665 |
+
"num_train_epochs": 10,
|
| 1666 |
"save_steps": 100,
|
| 1667 |
"stateful_callbacks": {
|
| 1668 |
"TrainerControl": {
|
|
|
|
| 1676 |
"attributes": {}
|
| 1677 |
}
|
| 1678 |
},
|
| 1679 |
+
"total_flos": 1.697049221804327e+18,
|
| 1680 |
"train_batch_size": 4,
|
| 1681 |
"trial_name": null,
|
| 1682 |
"trial_params": null
|