Model save
Browse files- README.md +11 -15
- adapter_config.json +3 -3
- adapter_model.safetensors +1 -1
- all_results.json +8 -8
- eval_results.json +4 -4
- train_results.json +4 -4
- trainer_state.json +345 -345
- training_args.bin +1 -1
README.md
CHANGED
|
@@ -2,13 +2,9 @@
|
|
| 2 |
license: other
|
| 3 |
library_name: peft
|
| 4 |
tags:
|
| 5 |
-
- alignment-handbook
|
| 6 |
-
- generated_from_trainer
|
| 7 |
- trl
|
| 8 |
- sft
|
| 9 |
- generated_from_trainer
|
| 10 |
-
datasets:
|
| 11 |
-
- ruozhiba
|
| 12 |
base_model: 01-ai/Yi-6B
|
| 13 |
model-index:
|
| 14 |
- name: Yi-6B-ruozhiba-1e-5-50
|
|
@@ -20,9 +16,9 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 20 |
|
| 21 |
# Yi-6B-ruozhiba-1e-5-50
|
| 22 |
|
| 23 |
-
This model is a fine-tuned version of [01-ai/Yi-6B](https://huggingface.co/01-ai/Yi-6B) on the
|
| 24 |
It achieves the following results on the evaluation set:
|
| 25 |
-
- Loss:
|
| 26 |
|
| 27 |
## Model description
|
| 28 |
|
|
@@ -41,7 +37,7 @@ More information needed
|
|
| 41 |
### Training hyperparameters
|
| 42 |
|
| 43 |
The following hyperparameters were used during training:
|
| 44 |
-
- learning_rate:
|
| 45 |
- train_batch_size: 4
|
| 46 |
- eval_batch_size: 4
|
| 47 |
- seed: 42
|
|
@@ -54,14 +50,14 @@ The following hyperparameters were used during training:
|
|
| 54 |
|
| 55 |
| Training Loss | Epoch | Step | Validation Loss |
|
| 56 |
|:-------------:|:-----:|:----:|:---------------:|
|
| 57 |
-
|
|
| 58 |
-
|
|
| 59 |
-
|
|
| 60 |
-
|
|
| 61 |
-
|
|
| 62 |
-
|
|
| 63 |
-
|
|
| 64 |
-
|
|
| 65 |
|
| 66 |
|
| 67 |
### Framework versions
|
|
|
|
| 2 |
license: other
|
| 3 |
library_name: peft
|
| 4 |
tags:
|
|
|
|
|
|
|
| 5 |
- trl
|
| 6 |
- sft
|
| 7 |
- generated_from_trainer
|
|
|
|
|
|
|
| 8 |
base_model: 01-ai/Yi-6B
|
| 9 |
model-index:
|
| 10 |
- name: Yi-6B-ruozhiba-1e-5-50
|
|
|
|
| 16 |
|
| 17 |
# Yi-6B-ruozhiba-1e-5-50
|
| 18 |
|
| 19 |
+
This model is a fine-tuned version of [01-ai/Yi-6B](https://huggingface.co/01-ai/Yi-6B) on the None dataset.
|
| 20 |
It achieves the following results on the evaluation set:
|
| 21 |
+
- Loss: 1.9851
|
| 22 |
|
| 23 |
## Model description
|
| 24 |
|
|
|
|
| 37 |
### Training hyperparameters
|
| 38 |
|
| 39 |
The following hyperparameters were used during training:
|
| 40 |
+
- learning_rate: 1e-05
|
| 41 |
- train_batch_size: 4
|
| 42 |
- eval_batch_size: 4
|
| 43 |
- seed: 42
|
|
|
|
| 50 |
|
| 51 |
| Training Loss | Epoch | Step | Validation Loss |
|
| 52 |
|:-------------:|:-----:|:----:|:---------------:|
|
| 53 |
+
| 2.6749 | 1.0 | 55 | 2.5387 |
|
| 54 |
+
| 2.0502 | 3.0 | 165 | 2.0924 |
|
| 55 |
+
| 1.9287 | 5.0 | 275 | 2.0103 |
|
| 56 |
+
| 2.0021 | 6.0 | 330 | 1.9935 |
|
| 57 |
+
| 1.8199 | 7.0 | 385 | 1.9878 |
|
| 58 |
+
| 1.9559 | 8.0 | 440 | 1.9858 |
|
| 59 |
+
| 1.8229 | 9.0 | 495 | 1.9853 |
|
| 60 |
+
| 1.8178 | 10.0 | 550 | 1.9851 |
|
| 61 |
|
| 62 |
|
| 63 |
### Framework versions
|
adapter_config.json
CHANGED
|
@@ -19,12 +19,12 @@
|
|
| 19 |
"rank_pattern": {},
|
| 20 |
"revision": null,
|
| 21 |
"target_modules": [
|
| 22 |
-
"gate_proj",
|
| 23 |
-
"q_proj",
|
| 24 |
-
"up_proj",
|
| 25 |
"v_proj",
|
| 26 |
"k_proj",
|
| 27 |
"o_proj",
|
|
|
|
|
|
|
|
|
|
| 28 |
"down_proj"
|
| 29 |
],
|
| 30 |
"task_type": "CAUSAL_LM"
|
|
|
|
| 19 |
"rank_pattern": {},
|
| 20 |
"revision": null,
|
| 21 |
"target_modules": [
|
|
|
|
|
|
|
|
|
|
| 22 |
"v_proj",
|
| 23 |
"k_proj",
|
| 24 |
"o_proj",
|
| 25 |
+
"gate_proj",
|
| 26 |
+
"q_proj",
|
| 27 |
+
"up_proj",
|
| 28 |
"down_proj"
|
| 29 |
],
|
| 30 |
"task_type": "CAUSAL_LM"
|
adapter_model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 145287696
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4a7400933ae66031ef5dd58be0dff5a4913af8976f677bd919c53f2f056545d7
|
| 3 |
size 145287696
|
all_results.json
CHANGED
|
@@ -1,13 +1,13 @@
|
|
| 1 |
{
|
| 2 |
"epoch": 10.0,
|
| 3 |
-
"eval_loss":
|
| 4 |
-
"eval_runtime": 4.
|
| 5 |
"eval_samples": 23,
|
| 6 |
-
"eval_samples_per_second": 4.
|
| 7 |
-
"eval_steps_per_second": 1.
|
| 8 |
-
"train_loss":
|
| 9 |
-
"train_runtime":
|
| 10 |
"train_samples": 217,
|
| 11 |
-
"train_samples_per_second": 0.
|
| 12 |
-
"train_steps_per_second": 0.
|
| 13 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"epoch": 10.0,
|
| 3 |
+
"eval_loss": 1.9850994348526,
|
| 4 |
+
"eval_runtime": 4.9118,
|
| 5 |
"eval_samples": 23,
|
| 6 |
+
"eval_samples_per_second": 4.683,
|
| 7 |
+
"eval_steps_per_second": 1.222,
|
| 8 |
+
"train_loss": 1.1392531923814253,
|
| 9 |
+
"train_runtime": 18256.4954,
|
| 10 |
"train_samples": 217,
|
| 11 |
+
"train_samples_per_second": 0.119,
|
| 12 |
+
"train_steps_per_second": 0.03
|
| 13 |
}
|
eval_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
"epoch": 10.0,
|
| 3 |
-
"eval_loss":
|
| 4 |
-
"eval_runtime": 4.
|
| 5 |
"eval_samples": 23,
|
| 6 |
-
"eval_samples_per_second": 4.
|
| 7 |
-
"eval_steps_per_second": 1.
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"epoch": 10.0,
|
| 3 |
+
"eval_loss": 1.9850994348526,
|
| 4 |
+
"eval_runtime": 4.9118,
|
| 5 |
"eval_samples": 23,
|
| 6 |
+
"eval_samples_per_second": 4.683,
|
| 7 |
+
"eval_steps_per_second": 1.222
|
| 8 |
}
|
train_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
"epoch": 10.0,
|
| 3 |
-
"train_loss":
|
| 4 |
-
"train_runtime":
|
| 5 |
"train_samples": 217,
|
| 6 |
-
"train_samples_per_second": 0.
|
| 7 |
-
"train_steps_per_second": 0.
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"epoch": 10.0,
|
| 3 |
+
"train_loss": 1.1392531923814253,
|
| 4 |
+
"train_runtime": 18256.4954,
|
| 5 |
"train_samples": 217,
|
| 6 |
+
"train_samples_per_second": 0.119,
|
| 7 |
+
"train_steps_per_second": 0.03
|
| 8 |
}
|
trainer_state.json
CHANGED
|
@@ -10,952 +10,952 @@
|
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
"epoch": 0.02,
|
| 13 |
-
"learning_rate":
|
| 14 |
"loss": 2.7431,
|
| 15 |
"step": 1
|
| 16 |
},
|
| 17 |
{
|
| 18 |
"epoch": 0.07,
|
| 19 |
-
"learning_rate":
|
| 20 |
-
"loss": 2.
|
| 21 |
"step": 4
|
| 22 |
},
|
| 23 |
{
|
| 24 |
"epoch": 0.15,
|
| 25 |
-
"learning_rate":
|
| 26 |
-
"loss": 2.
|
| 27 |
"step": 8
|
| 28 |
},
|
| 29 |
{
|
| 30 |
"epoch": 0.22,
|
| 31 |
-
"learning_rate":
|
| 32 |
-
"loss": 2.
|
| 33 |
"step": 12
|
| 34 |
},
|
| 35 |
{
|
| 36 |
"epoch": 0.29,
|
| 37 |
-
"learning_rate":
|
| 38 |
-
"loss": 2.
|
| 39 |
"step": 16
|
| 40 |
},
|
| 41 |
{
|
| 42 |
"epoch": 0.36,
|
| 43 |
-
"learning_rate":
|
| 44 |
-
"loss": 2.
|
| 45 |
"step": 20
|
| 46 |
},
|
| 47 |
{
|
| 48 |
"epoch": 0.44,
|
| 49 |
-
"learning_rate":
|
| 50 |
-
"loss": 2.
|
| 51 |
"step": 24
|
| 52 |
},
|
| 53 |
{
|
| 54 |
"epoch": 0.51,
|
| 55 |
-
"learning_rate":
|
| 56 |
-
"loss":
|
| 57 |
"step": 28
|
| 58 |
},
|
| 59 |
{
|
| 60 |
"epoch": 0.58,
|
| 61 |
-
"learning_rate":
|
| 62 |
-
"loss": 2.
|
| 63 |
"step": 32
|
| 64 |
},
|
| 65 |
{
|
| 66 |
"epoch": 0.65,
|
| 67 |
-
"learning_rate":
|
| 68 |
-
"loss": 2.
|
| 69 |
"step": 36
|
| 70 |
},
|
| 71 |
{
|
| 72 |
"epoch": 0.73,
|
| 73 |
-
"learning_rate":
|
| 74 |
-
"loss":
|
| 75 |
"step": 40
|
| 76 |
},
|
| 77 |
{
|
| 78 |
"epoch": 0.8,
|
| 79 |
-
"learning_rate":
|
| 80 |
-
"loss":
|
| 81 |
"step": 44
|
| 82 |
},
|
| 83 |
{
|
| 84 |
"epoch": 0.87,
|
| 85 |
-
"learning_rate":
|
| 86 |
-
"loss":
|
| 87 |
"step": 48
|
| 88 |
},
|
| 89 |
{
|
| 90 |
"epoch": 0.95,
|
| 91 |
-
"learning_rate":
|
| 92 |
-
"loss": 2.
|
| 93 |
"step": 52
|
| 94 |
},
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 95 |
{
|
| 96 |
"epoch": 1.02,
|
| 97 |
-
"learning_rate":
|
| 98 |
-
"loss":
|
| 99 |
"step": 56
|
| 100 |
},
|
| 101 |
{
|
| 102 |
"epoch": 1.09,
|
| 103 |
-
"learning_rate":
|
| 104 |
-
"loss":
|
| 105 |
"step": 60
|
| 106 |
},
|
| 107 |
{
|
| 108 |
"epoch": 1.16,
|
| 109 |
-
"learning_rate":
|
| 110 |
-
"loss":
|
| 111 |
"step": 64
|
| 112 |
},
|
| 113 |
{
|
| 114 |
"epoch": 1.24,
|
| 115 |
-
"learning_rate":
|
| 116 |
-
"loss":
|
| 117 |
"step": 68
|
| 118 |
},
|
| 119 |
{
|
| 120 |
"epoch": 1.31,
|
| 121 |
-
"learning_rate":
|
| 122 |
-
"loss":
|
| 123 |
"step": 72
|
| 124 |
},
|
| 125 |
{
|
| 126 |
"epoch": 1.38,
|
| 127 |
-
"learning_rate":
|
| 128 |
-
"loss":
|
| 129 |
"step": 76
|
| 130 |
},
|
| 131 |
{
|
| 132 |
"epoch": 1.45,
|
| 133 |
-
"learning_rate":
|
| 134 |
-
"loss":
|
| 135 |
"step": 80
|
| 136 |
},
|
| 137 |
{
|
| 138 |
"epoch": 1.53,
|
| 139 |
-
"learning_rate":
|
| 140 |
-
"loss":
|
| 141 |
"step": 84
|
| 142 |
},
|
| 143 |
{
|
| 144 |
"epoch": 1.6,
|
| 145 |
-
"learning_rate":
|
| 146 |
-
"loss":
|
| 147 |
"step": 88
|
| 148 |
},
|
| 149 |
{
|
| 150 |
"epoch": 1.67,
|
| 151 |
-
"learning_rate":
|
| 152 |
-
"loss":
|
| 153 |
"step": 92
|
| 154 |
},
|
| 155 |
{
|
| 156 |
"epoch": 1.75,
|
| 157 |
-
"learning_rate":
|
| 158 |
-
"loss":
|
| 159 |
"step": 96
|
| 160 |
},
|
| 161 |
{
|
| 162 |
"epoch": 1.82,
|
| 163 |
-
"learning_rate":
|
| 164 |
-
"loss":
|
| 165 |
"step": 100
|
| 166 |
},
|
| 167 |
{
|
| 168 |
"epoch": 1.89,
|
| 169 |
-
"learning_rate":
|
| 170 |
-
"loss":
|
| 171 |
"step": 104
|
| 172 |
},
|
| 173 |
{
|
| 174 |
"epoch": 1.96,
|
| 175 |
-
"learning_rate":
|
| 176 |
-
"loss":
|
| 177 |
"step": 108
|
| 178 |
},
|
| 179 |
-
{
|
| 180 |
-
"epoch": 2.0,
|
| 181 |
-
"pls_score": 62.8,
|
| 182 |
-
"std": 4.561052510112109,
|
| 183 |
-
"step": 110
|
| 184 |
-
},
|
| 185 |
-
{
|
| 186 |
-
"epoch": 2.0,
|
| 187 |
-
"eval_loss": 2.0346944332122803,
|
| 188 |
-
"eval_runtime": 4.9191,
|
| 189 |
-
"eval_samples_per_second": 4.676,
|
| 190 |
-
"eval_steps_per_second": 1.22,
|
| 191 |
-
"step": 110
|
| 192 |
-
},
|
| 193 |
{
|
| 194 |
"epoch": 2.04,
|
| 195 |
-
"learning_rate":
|
| 196 |
-
"loss":
|
| 197 |
"step": 112
|
| 198 |
},
|
| 199 |
{
|
| 200 |
"epoch": 2.11,
|
| 201 |
-
"learning_rate":
|
| 202 |
-
"loss":
|
| 203 |
"step": 116
|
| 204 |
},
|
| 205 |
{
|
| 206 |
"epoch": 2.18,
|
| 207 |
-
"learning_rate":
|
| 208 |
-
"loss":
|
| 209 |
"step": 120
|
| 210 |
},
|
| 211 |
{
|
| 212 |
"epoch": 2.25,
|
| 213 |
-
"learning_rate":
|
| 214 |
-
"loss":
|
| 215 |
"step": 124
|
| 216 |
},
|
| 217 |
{
|
| 218 |
"epoch": 2.33,
|
| 219 |
-
"learning_rate":
|
| 220 |
-
"loss":
|
| 221 |
"step": 128
|
| 222 |
},
|
| 223 |
{
|
| 224 |
"epoch": 2.4,
|
| 225 |
-
"learning_rate":
|
| 226 |
-
"loss":
|
| 227 |
"step": 132
|
| 228 |
},
|
| 229 |
{
|
| 230 |
"epoch": 2.47,
|
| 231 |
-
"learning_rate":
|
| 232 |
-
"loss":
|
| 233 |
"step": 136
|
| 234 |
},
|
| 235 |
{
|
| 236 |
"epoch": 2.55,
|
| 237 |
-
"learning_rate":
|
| 238 |
-
"loss":
|
| 239 |
"step": 140
|
| 240 |
},
|
| 241 |
{
|
| 242 |
"epoch": 2.62,
|
| 243 |
-
"learning_rate":
|
| 244 |
-
"loss":
|
| 245 |
"step": 144
|
| 246 |
},
|
| 247 |
{
|
| 248 |
"epoch": 2.69,
|
| 249 |
-
"learning_rate":
|
| 250 |
-
"loss":
|
| 251 |
"step": 148
|
| 252 |
},
|
| 253 |
{
|
| 254 |
"epoch": 2.76,
|
| 255 |
-
"learning_rate":
|
| 256 |
-
"loss":
|
| 257 |
"step": 152
|
| 258 |
},
|
| 259 |
{
|
| 260 |
"epoch": 2.84,
|
| 261 |
-
"learning_rate":
|
| 262 |
-
"loss":
|
| 263 |
"step": 156
|
| 264 |
},
|
| 265 |
{
|
| 266 |
"epoch": 2.91,
|
| 267 |
-
"learning_rate":
|
| 268 |
-
"loss":
|
| 269 |
"step": 160
|
| 270 |
},
|
| 271 |
{
|
| 272 |
"epoch": 2.98,
|
| 273 |
-
"learning_rate":
|
| 274 |
-
"loss":
|
| 275 |
"step": 164
|
| 276 |
},
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 277 |
{
|
| 278 |
"epoch": 3.05,
|
| 279 |
-
"learning_rate":
|
| 280 |
-
"loss":
|
| 281 |
"step": 168
|
| 282 |
},
|
| 283 |
{
|
| 284 |
"epoch": 3.13,
|
| 285 |
-
"learning_rate":
|
| 286 |
-
"loss":
|
| 287 |
"step": 172
|
| 288 |
},
|
| 289 |
{
|
| 290 |
"epoch": 3.2,
|
| 291 |
-
"learning_rate":
|
| 292 |
-
"loss":
|
| 293 |
"step": 176
|
| 294 |
},
|
| 295 |
{
|
| 296 |
"epoch": 3.27,
|
| 297 |
-
"learning_rate":
|
| 298 |
-
"loss":
|
| 299 |
"step": 180
|
| 300 |
},
|
| 301 |
{
|
| 302 |
"epoch": 3.35,
|
| 303 |
-
"learning_rate":
|
| 304 |
-
"loss":
|
| 305 |
"step": 184
|
| 306 |
},
|
| 307 |
{
|
| 308 |
"epoch": 3.42,
|
| 309 |
-
"learning_rate":
|
| 310 |
-
"loss":
|
| 311 |
"step": 188
|
| 312 |
},
|
| 313 |
{
|
| 314 |
"epoch": 3.49,
|
| 315 |
-
"learning_rate":
|
| 316 |
-
"loss":
|
| 317 |
"step": 192
|
| 318 |
},
|
| 319 |
{
|
| 320 |
"epoch": 3.56,
|
| 321 |
-
"learning_rate":
|
| 322 |
-
"loss":
|
| 323 |
"step": 196
|
| 324 |
},
|
| 325 |
{
|
| 326 |
"epoch": 3.64,
|
| 327 |
-
"learning_rate":
|
| 328 |
-
"loss":
|
| 329 |
"step": 200
|
| 330 |
},
|
| 331 |
{
|
| 332 |
"epoch": 3.71,
|
| 333 |
-
"learning_rate":
|
| 334 |
-
"loss":
|
| 335 |
"step": 204
|
| 336 |
},
|
| 337 |
{
|
| 338 |
"epoch": 3.78,
|
| 339 |
-
"learning_rate":
|
| 340 |
-
"loss":
|
| 341 |
"step": 208
|
| 342 |
},
|
| 343 |
{
|
| 344 |
"epoch": 3.85,
|
| 345 |
-
"learning_rate":
|
| 346 |
-
"loss":
|
| 347 |
"step": 212
|
| 348 |
},
|
| 349 |
{
|
| 350 |
"epoch": 3.93,
|
| 351 |
-
"learning_rate":
|
| 352 |
-
"loss":
|
| 353 |
"step": 216
|
| 354 |
},
|
| 355 |
{
|
| 356 |
"epoch": 4.0,
|
| 357 |
-
"learning_rate":
|
| 358 |
-
"loss":
|
| 359 |
-
"step": 220
|
| 360 |
-
},
|
| 361 |
-
{
|
| 362 |
-
"epoch": 4.0,
|
| 363 |
-
"pls_score": 63.6,
|
| 364 |
-
"std": 4.067038234391212,
|
| 365 |
-
"step": 220
|
| 366 |
-
},
|
| 367 |
-
{
|
| 368 |
-
"epoch": 4.0,
|
| 369 |
-
"eval_loss": 2.783196449279785,
|
| 370 |
-
"eval_runtime": 4.9791,
|
| 371 |
-
"eval_samples_per_second": 4.619,
|
| 372 |
-
"eval_steps_per_second": 1.205,
|
| 373 |
"step": 220
|
| 374 |
},
|
| 375 |
{
|
| 376 |
"epoch": 4.07,
|
| 377 |
-
"learning_rate":
|
| 378 |
-
"loss":
|
| 379 |
"step": 224
|
| 380 |
},
|
| 381 |
{
|
| 382 |
"epoch": 4.15,
|
| 383 |
-
"learning_rate":
|
| 384 |
-
"loss":
|
| 385 |
"step": 228
|
| 386 |
},
|
| 387 |
{
|
| 388 |
"epoch": 4.22,
|
| 389 |
-
"learning_rate":
|
| 390 |
-
"loss":
|
| 391 |
"step": 232
|
| 392 |
},
|
| 393 |
{
|
| 394 |
"epoch": 4.29,
|
| 395 |
-
"learning_rate":
|
| 396 |
-
"loss":
|
| 397 |
"step": 236
|
| 398 |
},
|
| 399 |
{
|
| 400 |
"epoch": 4.36,
|
| 401 |
-
"learning_rate":
|
| 402 |
-
"loss":
|
| 403 |
"step": 240
|
| 404 |
},
|
| 405 |
{
|
| 406 |
"epoch": 4.44,
|
| 407 |
-
"learning_rate":
|
| 408 |
-
"loss":
|
| 409 |
"step": 244
|
| 410 |
},
|
| 411 |
{
|
| 412 |
"epoch": 4.51,
|
| 413 |
-
"learning_rate":
|
| 414 |
-
"loss":
|
| 415 |
"step": 248
|
| 416 |
},
|
| 417 |
{
|
| 418 |
"epoch": 4.58,
|
| 419 |
-
"learning_rate":
|
| 420 |
-
"loss":
|
| 421 |
"step": 252
|
| 422 |
},
|
| 423 |
{
|
| 424 |
"epoch": 4.65,
|
| 425 |
-
"learning_rate":
|
| 426 |
-
"loss":
|
| 427 |
"step": 256
|
| 428 |
},
|
| 429 |
{
|
| 430 |
"epoch": 4.73,
|
| 431 |
-
"learning_rate":
|
| 432 |
-
"loss":
|
| 433 |
"step": 260
|
| 434 |
},
|
| 435 |
{
|
| 436 |
"epoch": 4.8,
|
| 437 |
-
"learning_rate":
|
| 438 |
-
"loss":
|
| 439 |
"step": 264
|
| 440 |
},
|
| 441 |
{
|
| 442 |
"epoch": 4.87,
|
| 443 |
-
"learning_rate":
|
| 444 |
-
"loss":
|
| 445 |
"step": 268
|
| 446 |
},
|
| 447 |
{
|
| 448 |
"epoch": 4.95,
|
| 449 |
-
"learning_rate":
|
| 450 |
-
"loss":
|
| 451 |
"step": 272
|
| 452 |
},
|
| 453 |
{
|
| 454 |
"epoch": 5.0,
|
| 455 |
-
"pls_score":
|
| 456 |
-
"std": 3.
|
| 457 |
"step": 275
|
| 458 |
},
|
| 459 |
{
|
| 460 |
"epoch": 5.0,
|
| 461 |
-
"eval_loss":
|
| 462 |
-
"eval_runtime": 4.
|
| 463 |
-
"eval_samples_per_second": 4.
|
| 464 |
-
"eval_steps_per_second": 1.
|
| 465 |
"step": 275
|
| 466 |
},
|
| 467 |
{
|
| 468 |
"epoch": 5.02,
|
| 469 |
-
"learning_rate":
|
| 470 |
-
"loss":
|
| 471 |
"step": 276
|
| 472 |
},
|
| 473 |
{
|
| 474 |
"epoch": 5.09,
|
| 475 |
-
"learning_rate":
|
| 476 |
-
"loss":
|
| 477 |
"step": 280
|
| 478 |
},
|
| 479 |
{
|
| 480 |
"epoch": 5.16,
|
| 481 |
-
"learning_rate":
|
| 482 |
-
"loss":
|
| 483 |
"step": 284
|
| 484 |
},
|
| 485 |
{
|
| 486 |
"epoch": 5.24,
|
| 487 |
-
"learning_rate":
|
| 488 |
-
"loss":
|
| 489 |
"step": 288
|
| 490 |
},
|
| 491 |
{
|
| 492 |
"epoch": 5.31,
|
| 493 |
-
"learning_rate":
|
| 494 |
-
"loss":
|
| 495 |
"step": 292
|
| 496 |
},
|
| 497 |
{
|
| 498 |
"epoch": 5.38,
|
| 499 |
-
"learning_rate":
|
| 500 |
-
"loss":
|
| 501 |
"step": 296
|
| 502 |
},
|
| 503 |
{
|
| 504 |
"epoch": 5.45,
|
| 505 |
-
"learning_rate":
|
| 506 |
-
"loss":
|
| 507 |
"step": 300
|
| 508 |
},
|
| 509 |
{
|
| 510 |
"epoch": 5.53,
|
| 511 |
-
"learning_rate":
|
| 512 |
-
"loss":
|
| 513 |
"step": 304
|
| 514 |
},
|
| 515 |
{
|
| 516 |
"epoch": 5.6,
|
| 517 |
-
"learning_rate":
|
| 518 |
-
"loss":
|
| 519 |
"step": 308
|
| 520 |
},
|
| 521 |
{
|
| 522 |
"epoch": 5.67,
|
| 523 |
-
"learning_rate":
|
| 524 |
-
"loss":
|
| 525 |
"step": 312
|
| 526 |
},
|
| 527 |
{
|
| 528 |
"epoch": 5.75,
|
| 529 |
-
"learning_rate":
|
| 530 |
-
"loss":
|
| 531 |
"step": 316
|
| 532 |
},
|
| 533 |
{
|
| 534 |
"epoch": 5.82,
|
| 535 |
-
"learning_rate":
|
| 536 |
-
"loss":
|
| 537 |
"step": 320
|
| 538 |
},
|
| 539 |
{
|
| 540 |
"epoch": 5.89,
|
| 541 |
-
"learning_rate":
|
| 542 |
-
"loss":
|
| 543 |
"step": 324
|
| 544 |
},
|
| 545 |
{
|
| 546 |
"epoch": 5.96,
|
| 547 |
-
"learning_rate":
|
| 548 |
-
"loss":
|
| 549 |
"step": 328
|
| 550 |
},
|
| 551 |
{
|
| 552 |
"epoch": 6.0,
|
| 553 |
-
"pls_score":
|
| 554 |
-
"std": 4.
|
| 555 |
"step": 330
|
| 556 |
},
|
| 557 |
{
|
| 558 |
"epoch": 6.0,
|
| 559 |
-
"eval_loss":
|
| 560 |
-
"eval_runtime": 4.
|
| 561 |
-
"eval_samples_per_second": 4.
|
| 562 |
-
"eval_steps_per_second": 1.
|
| 563 |
"step": 330
|
| 564 |
},
|
| 565 |
{
|
| 566 |
"epoch": 6.04,
|
| 567 |
-
"learning_rate":
|
| 568 |
-
"loss":
|
| 569 |
"step": 332
|
| 570 |
},
|
| 571 |
{
|
| 572 |
"epoch": 6.11,
|
| 573 |
-
"learning_rate":
|
| 574 |
-
"loss":
|
| 575 |
"step": 336
|
| 576 |
},
|
| 577 |
{
|
| 578 |
"epoch": 6.18,
|
| 579 |
-
"learning_rate":
|
| 580 |
-
"loss":
|
| 581 |
"step": 340
|
| 582 |
},
|
| 583 |
{
|
| 584 |
"epoch": 6.25,
|
| 585 |
-
"learning_rate":
|
| 586 |
-
"loss":
|
| 587 |
"step": 344
|
| 588 |
},
|
| 589 |
{
|
| 590 |
"epoch": 6.33,
|
| 591 |
-
"learning_rate":
|
| 592 |
-
"loss":
|
| 593 |
"step": 348
|
| 594 |
},
|
| 595 |
{
|
| 596 |
"epoch": 6.4,
|
| 597 |
-
"learning_rate":
|
| 598 |
-
"loss":
|
| 599 |
"step": 352
|
| 600 |
},
|
| 601 |
{
|
| 602 |
"epoch": 6.47,
|
| 603 |
-
"learning_rate":
|
| 604 |
-
"loss":
|
| 605 |
"step": 356
|
| 606 |
},
|
| 607 |
{
|
| 608 |
"epoch": 6.55,
|
| 609 |
-
"learning_rate":
|
| 610 |
-
"loss":
|
| 611 |
"step": 360
|
| 612 |
},
|
| 613 |
{
|
| 614 |
"epoch": 6.62,
|
| 615 |
-
"learning_rate":
|
| 616 |
-
"loss":
|
| 617 |
"step": 364
|
| 618 |
},
|
| 619 |
{
|
| 620 |
"epoch": 6.69,
|
| 621 |
-
"learning_rate":
|
| 622 |
-
"loss":
|
| 623 |
"step": 368
|
| 624 |
},
|
| 625 |
{
|
| 626 |
"epoch": 6.76,
|
| 627 |
-
"learning_rate":
|
| 628 |
-
"loss":
|
| 629 |
"step": 372
|
| 630 |
},
|
| 631 |
{
|
| 632 |
"epoch": 6.84,
|
| 633 |
-
"learning_rate":
|
| 634 |
-
"loss":
|
| 635 |
"step": 376
|
| 636 |
},
|
| 637 |
{
|
| 638 |
"epoch": 6.91,
|
| 639 |
-
"learning_rate":
|
| 640 |
-
"loss":
|
| 641 |
"step": 380
|
| 642 |
},
|
| 643 |
{
|
| 644 |
"epoch": 6.98,
|
| 645 |
-
"learning_rate":
|
| 646 |
-
"loss":
|
| 647 |
"step": 384
|
| 648 |
},
|
| 649 |
{
|
| 650 |
"epoch": 7.0,
|
| 651 |
-
"pls_score": 59.
|
| 652 |
-
"std": 4.
|
| 653 |
"step": 385
|
| 654 |
},
|
| 655 |
{
|
| 656 |
"epoch": 7.0,
|
| 657 |
-
"eval_loss":
|
| 658 |
-
"eval_runtime": 4.
|
| 659 |
-
"eval_samples_per_second": 4.
|
| 660 |
-
"eval_steps_per_second": 1.
|
| 661 |
"step": 385
|
| 662 |
},
|
| 663 |
{
|
| 664 |
"epoch": 7.05,
|
| 665 |
-
"learning_rate":
|
| 666 |
-
"loss":
|
| 667 |
"step": 388
|
| 668 |
},
|
| 669 |
{
|
| 670 |
"epoch": 7.13,
|
| 671 |
-
"learning_rate":
|
| 672 |
-
"loss":
|
| 673 |
"step": 392
|
| 674 |
},
|
| 675 |
{
|
| 676 |
"epoch": 7.2,
|
| 677 |
-
"learning_rate":
|
| 678 |
-
"loss":
|
| 679 |
"step": 396
|
| 680 |
},
|
| 681 |
{
|
| 682 |
"epoch": 7.27,
|
| 683 |
-
"learning_rate":
|
| 684 |
-
"loss":
|
| 685 |
"step": 400
|
| 686 |
},
|
| 687 |
{
|
| 688 |
"epoch": 7.35,
|
| 689 |
-
"learning_rate":
|
| 690 |
-
"loss":
|
| 691 |
"step": 404
|
| 692 |
},
|
| 693 |
{
|
| 694 |
"epoch": 7.42,
|
| 695 |
-
"learning_rate":
|
| 696 |
-
"loss":
|
| 697 |
"step": 408
|
| 698 |
},
|
| 699 |
{
|
| 700 |
"epoch": 7.49,
|
| 701 |
-
"learning_rate":
|
| 702 |
-
"loss":
|
| 703 |
"step": 412
|
| 704 |
},
|
| 705 |
{
|
| 706 |
"epoch": 7.56,
|
| 707 |
-
"learning_rate":
|
| 708 |
-
"loss":
|
| 709 |
"step": 416
|
| 710 |
},
|
| 711 |
{
|
| 712 |
"epoch": 7.64,
|
| 713 |
-
"learning_rate":
|
| 714 |
-
"loss":
|
| 715 |
"step": 420
|
| 716 |
},
|
| 717 |
{
|
| 718 |
"epoch": 7.71,
|
| 719 |
-
"learning_rate":
|
| 720 |
-
"loss":
|
| 721 |
"step": 424
|
| 722 |
},
|
| 723 |
{
|
| 724 |
"epoch": 7.78,
|
| 725 |
-
"learning_rate":
|
| 726 |
-
"loss":
|
| 727 |
"step": 428
|
| 728 |
},
|
| 729 |
{
|
| 730 |
"epoch": 7.85,
|
| 731 |
-
"learning_rate":
|
| 732 |
-
"loss":
|
| 733 |
"step": 432
|
| 734 |
},
|
| 735 |
{
|
| 736 |
"epoch": 7.93,
|
| 737 |
-
"learning_rate":
|
| 738 |
-
"loss":
|
| 739 |
"step": 436
|
| 740 |
},
|
| 741 |
{
|
| 742 |
"epoch": 8.0,
|
| 743 |
-
"learning_rate":
|
| 744 |
-
"loss":
|
| 745 |
"step": 440
|
| 746 |
},
|
| 747 |
{
|
| 748 |
"epoch": 8.0,
|
| 749 |
-
"pls_score":
|
| 750 |
-
"std":
|
| 751 |
"step": 440
|
| 752 |
},
|
| 753 |
{
|
| 754 |
"epoch": 8.0,
|
| 755 |
-
"eval_loss":
|
| 756 |
-
"eval_runtime": 4.
|
| 757 |
-
"eval_samples_per_second": 4.
|
| 758 |
-
"eval_steps_per_second": 1.
|
| 759 |
"step": 440
|
| 760 |
},
|
| 761 |
{
|
| 762 |
"epoch": 8.07,
|
| 763 |
-
"learning_rate":
|
| 764 |
-
"loss":
|
| 765 |
"step": 444
|
| 766 |
},
|
| 767 |
{
|
| 768 |
"epoch": 8.15,
|
| 769 |
-
"learning_rate":
|
| 770 |
-
"loss":
|
| 771 |
"step": 448
|
| 772 |
},
|
| 773 |
{
|
| 774 |
"epoch": 8.22,
|
| 775 |
-
"learning_rate":
|
| 776 |
-
"loss":
|
| 777 |
"step": 452
|
| 778 |
},
|
| 779 |
{
|
| 780 |
"epoch": 8.29,
|
| 781 |
-
"learning_rate":
|
| 782 |
-
"loss":
|
| 783 |
"step": 456
|
| 784 |
},
|
| 785 |
{
|
| 786 |
"epoch": 8.36,
|
| 787 |
-
"learning_rate":
|
| 788 |
-
"loss":
|
| 789 |
"step": 460
|
| 790 |
},
|
| 791 |
{
|
| 792 |
"epoch": 8.44,
|
| 793 |
-
"learning_rate":
|
| 794 |
-
"loss":
|
| 795 |
"step": 464
|
| 796 |
},
|
| 797 |
{
|
| 798 |
"epoch": 8.51,
|
| 799 |
-
"learning_rate":
|
| 800 |
-
"loss":
|
| 801 |
"step": 468
|
| 802 |
},
|
| 803 |
{
|
| 804 |
"epoch": 8.58,
|
| 805 |
-
"learning_rate":
|
| 806 |
-
"loss":
|
| 807 |
"step": 472
|
| 808 |
},
|
| 809 |
{
|
| 810 |
"epoch": 8.65,
|
| 811 |
-
"learning_rate":
|
| 812 |
-
"loss":
|
| 813 |
"step": 476
|
| 814 |
},
|
| 815 |
{
|
| 816 |
"epoch": 8.73,
|
| 817 |
-
"learning_rate":
|
| 818 |
-
"loss":
|
| 819 |
"step": 480
|
| 820 |
},
|
| 821 |
{
|
| 822 |
"epoch": 8.8,
|
| 823 |
-
"learning_rate":
|
| 824 |
-
"loss":
|
| 825 |
"step": 484
|
| 826 |
},
|
| 827 |
{
|
| 828 |
"epoch": 8.87,
|
| 829 |
-
"learning_rate":
|
| 830 |
-
"loss":
|
| 831 |
"step": 488
|
| 832 |
},
|
| 833 |
{
|
| 834 |
"epoch": 8.95,
|
| 835 |
-
"learning_rate":
|
| 836 |
-
"loss":
|
| 837 |
"step": 492
|
| 838 |
},
|
| 839 |
{
|
| 840 |
"epoch": 9.0,
|
| 841 |
-
"pls_score":
|
| 842 |
-
"std": 4.
|
| 843 |
"step": 495
|
| 844 |
},
|
| 845 |
{
|
| 846 |
"epoch": 9.0,
|
| 847 |
-
"eval_loss":
|
| 848 |
-
"eval_runtime": 4.
|
| 849 |
-
"eval_samples_per_second": 4.
|
| 850 |
-
"eval_steps_per_second": 1.
|
| 851 |
"step": 495
|
| 852 |
},
|
| 853 |
{
|
| 854 |
"epoch": 9.02,
|
| 855 |
-
"learning_rate":
|
| 856 |
-
"loss":
|
| 857 |
"step": 496
|
| 858 |
},
|
| 859 |
{
|
| 860 |
"epoch": 9.09,
|
| 861 |
-
"learning_rate":
|
| 862 |
-
"loss":
|
| 863 |
"step": 500
|
| 864 |
},
|
| 865 |
{
|
| 866 |
"epoch": 9.16,
|
| 867 |
-
"learning_rate":
|
| 868 |
-
"loss":
|
| 869 |
"step": 504
|
| 870 |
},
|
| 871 |
{
|
| 872 |
"epoch": 9.24,
|
| 873 |
-
"learning_rate":
|
| 874 |
-
"loss":
|
| 875 |
"step": 508
|
| 876 |
},
|
| 877 |
{
|
| 878 |
"epoch": 9.31,
|
| 879 |
-
"learning_rate":
|
| 880 |
-
"loss":
|
| 881 |
"step": 512
|
| 882 |
},
|
| 883 |
{
|
| 884 |
"epoch": 9.38,
|
| 885 |
-
"learning_rate":
|
| 886 |
-
"loss":
|
| 887 |
"step": 516
|
| 888 |
},
|
| 889 |
{
|
| 890 |
"epoch": 9.45,
|
| 891 |
-
"learning_rate":
|
| 892 |
-
"loss":
|
| 893 |
"step": 520
|
| 894 |
},
|
| 895 |
{
|
| 896 |
"epoch": 9.53,
|
| 897 |
-
"learning_rate":
|
| 898 |
-
"loss":
|
| 899 |
"step": 524
|
| 900 |
},
|
| 901 |
{
|
| 902 |
"epoch": 9.6,
|
| 903 |
-
"learning_rate":
|
| 904 |
-
"loss":
|
| 905 |
"step": 528
|
| 906 |
},
|
| 907 |
{
|
| 908 |
"epoch": 9.67,
|
| 909 |
-
"learning_rate":
|
| 910 |
-
"loss":
|
| 911 |
"step": 532
|
| 912 |
},
|
| 913 |
{
|
| 914 |
"epoch": 9.75,
|
| 915 |
-
"learning_rate":
|
| 916 |
-
"loss":
|
| 917 |
"step": 536
|
| 918 |
},
|
| 919 |
{
|
| 920 |
"epoch": 9.82,
|
| 921 |
-
"learning_rate":
|
| 922 |
-
"loss":
|
| 923 |
"step": 540
|
| 924 |
},
|
| 925 |
{
|
| 926 |
"epoch": 9.89,
|
| 927 |
-
"learning_rate":
|
| 928 |
-
"loss":
|
| 929 |
"step": 544
|
| 930 |
},
|
| 931 |
{
|
| 932 |
"epoch": 9.96,
|
| 933 |
-
"learning_rate":
|
| 934 |
-
"loss":
|
| 935 |
"step": 548
|
| 936 |
},
|
| 937 |
{
|
| 938 |
"epoch": 10.0,
|
| 939 |
-
"pls_score":
|
| 940 |
-
"std": 4.
|
| 941 |
"step": 550
|
| 942 |
},
|
| 943 |
{
|
| 944 |
"epoch": 10.0,
|
| 945 |
-
"eval_loss":
|
| 946 |
-
"eval_runtime": 4.
|
| 947 |
-
"eval_samples_per_second": 4.
|
| 948 |
-
"eval_steps_per_second": 1.
|
| 949 |
"step": 550
|
| 950 |
},
|
| 951 |
{
|
| 952 |
"epoch": 10.0,
|
| 953 |
"step": 550,
|
| 954 |
-
"total_flos": 1.
|
| 955 |
-
"train_loss":
|
| 956 |
-
"train_runtime":
|
| 957 |
-
"train_samples_per_second": 0.
|
| 958 |
-
"train_steps_per_second": 0.
|
| 959 |
}
|
| 960 |
],
|
| 961 |
"logging_steps": 4,
|
|
@@ -963,7 +963,7 @@
|
|
| 963 |
"num_input_tokens_seen": 0,
|
| 964 |
"num_train_epochs": 10,
|
| 965 |
"save_steps": 55,
|
| 966 |
-
"total_flos": 1.
|
| 967 |
"train_batch_size": 4,
|
| 968 |
"trial_name": null,
|
| 969 |
"trial_params": null
|
|
|
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
"epoch": 0.02,
|
| 13 |
+
"learning_rate": 1.8181818181818183e-07,
|
| 14 |
"loss": 2.7431,
|
| 15 |
"step": 1
|
| 16 |
},
|
| 17 |
{
|
| 18 |
"epoch": 0.07,
|
| 19 |
+
"learning_rate": 7.272727272727273e-07,
|
| 20 |
+
"loss": 2.89,
|
| 21 |
"step": 4
|
| 22 |
},
|
| 23 |
{
|
| 24 |
"epoch": 0.15,
|
| 25 |
+
"learning_rate": 1.4545454545454546e-06,
|
| 26 |
+
"loss": 2.7283,
|
| 27 |
"step": 8
|
| 28 |
},
|
| 29 |
{
|
| 30 |
"epoch": 0.22,
|
| 31 |
+
"learning_rate": 2.181818181818182e-06,
|
| 32 |
+
"loss": 2.6956,
|
| 33 |
"step": 12
|
| 34 |
},
|
| 35 |
{
|
| 36 |
"epoch": 0.29,
|
| 37 |
+
"learning_rate": 2.9090909090909093e-06,
|
| 38 |
+
"loss": 2.6684,
|
| 39 |
"step": 16
|
| 40 |
},
|
| 41 |
{
|
| 42 |
"epoch": 0.36,
|
| 43 |
+
"learning_rate": 3.6363636363636366e-06,
|
| 44 |
+
"loss": 2.7545,
|
| 45 |
"step": 20
|
| 46 |
},
|
| 47 |
{
|
| 48 |
"epoch": 0.44,
|
| 49 |
+
"learning_rate": 4.363636363636364e-06,
|
| 50 |
+
"loss": 2.8797,
|
| 51 |
"step": 24
|
| 52 |
},
|
| 53 |
{
|
| 54 |
"epoch": 0.51,
|
| 55 |
+
"learning_rate": 5.090909090909091e-06,
|
| 56 |
+
"loss": 3.0007,
|
| 57 |
"step": 28
|
| 58 |
},
|
| 59 |
{
|
| 60 |
"epoch": 0.58,
|
| 61 |
+
"learning_rate": 5.8181818181818185e-06,
|
| 62 |
+
"loss": 2.9172,
|
| 63 |
"step": 32
|
| 64 |
},
|
| 65 |
{
|
| 66 |
"epoch": 0.65,
|
| 67 |
+
"learning_rate": 6.545454545454546e-06,
|
| 68 |
+
"loss": 2.8011,
|
| 69 |
"step": 36
|
| 70 |
},
|
| 71 |
{
|
| 72 |
"epoch": 0.73,
|
| 73 |
+
"learning_rate": 7.272727272727273e-06,
|
| 74 |
+
"loss": 2.6401,
|
| 75 |
"step": 40
|
| 76 |
},
|
| 77 |
{
|
| 78 |
"epoch": 0.8,
|
| 79 |
+
"learning_rate": 8.000000000000001e-06,
|
| 80 |
+
"loss": 2.7162,
|
| 81 |
"step": 44
|
| 82 |
},
|
| 83 |
{
|
| 84 |
"epoch": 0.87,
|
| 85 |
+
"learning_rate": 8.727272727272728e-06,
|
| 86 |
+
"loss": 2.6036,
|
| 87 |
"step": 48
|
| 88 |
},
|
| 89 |
{
|
| 90 |
"epoch": 0.95,
|
| 91 |
+
"learning_rate": 9.454545454545456e-06,
|
| 92 |
+
"loss": 2.6749,
|
| 93 |
"step": 52
|
| 94 |
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 1.0,
|
| 97 |
+
"pls_score": 26.262626262626263,
|
| 98 |
+
"std": 2.1312214954012014,
|
| 99 |
+
"step": 55
|
| 100 |
+
},
|
| 101 |
+
{
|
| 102 |
+
"epoch": 1.0,
|
| 103 |
+
"eval_loss": 2.5386734008789062,
|
| 104 |
+
"eval_runtime": 4.9212,
|
| 105 |
+
"eval_samples_per_second": 4.674,
|
| 106 |
+
"eval_steps_per_second": 1.219,
|
| 107 |
+
"step": 55
|
| 108 |
+
},
|
| 109 |
{
|
| 110 |
"epoch": 1.02,
|
| 111 |
+
"learning_rate": 9.999899300364534e-06,
|
| 112 |
+
"loss": 2.6648,
|
| 113 |
"step": 56
|
| 114 |
},
|
| 115 |
{
|
| 116 |
"epoch": 1.09,
|
| 117 |
+
"learning_rate": 9.997482711915926e-06,
|
| 118 |
+
"loss": 2.5038,
|
| 119 |
"step": 60
|
| 120 |
},
|
| 121 |
{
|
| 122 |
"epoch": 1.16,
|
| 123 |
+
"learning_rate": 9.991845519630679e-06,
|
| 124 |
+
"loss": 2.424,
|
| 125 |
"step": 64
|
| 126 |
},
|
| 127 |
{
|
| 128 |
"epoch": 1.24,
|
| 129 |
+
"learning_rate": 9.982991356370404e-06,
|
| 130 |
+
"loss": 2.492,
|
| 131 |
"step": 68
|
| 132 |
},
|
| 133 |
{
|
| 134 |
"epoch": 1.31,
|
| 135 |
+
"learning_rate": 9.970925928158275e-06,
|
| 136 |
+
"loss": 2.5985,
|
| 137 |
"step": 72
|
| 138 |
},
|
| 139 |
{
|
| 140 |
"epoch": 1.38,
|
| 141 |
+
"learning_rate": 9.955657010501807e-06,
|
| 142 |
+
"loss": 2.2698,
|
| 143 |
"step": 76
|
| 144 |
},
|
| 145 |
{
|
| 146 |
"epoch": 1.45,
|
| 147 |
+
"learning_rate": 9.937194443381972e-06,
|
| 148 |
+
"loss": 2.3577,
|
| 149 |
"step": 80
|
| 150 |
},
|
| 151 |
{
|
| 152 |
"epoch": 1.53,
|
| 153 |
+
"learning_rate": 9.915550124911866e-06,
|
| 154 |
+
"loss": 2.4562,
|
| 155 |
"step": 84
|
| 156 |
},
|
| 157 |
{
|
| 158 |
"epoch": 1.6,
|
| 159 |
+
"learning_rate": 9.890738003669029e-06,
|
| 160 |
+
"loss": 2.266,
|
| 161 |
"step": 88
|
| 162 |
},
|
| 163 |
{
|
| 164 |
"epoch": 1.67,
|
| 165 |
+
"learning_rate": 9.862774069706346e-06,
|
| 166 |
+
"loss": 2.2448,
|
| 167 |
"step": 92
|
| 168 |
},
|
| 169 |
{
|
| 170 |
"epoch": 1.75,
|
| 171 |
+
"learning_rate": 9.831676344247343e-06,
|
| 172 |
+
"loss": 2.2374,
|
| 173 |
"step": 96
|
| 174 |
},
|
| 175 |
{
|
| 176 |
"epoch": 1.82,
|
| 177 |
+
"learning_rate": 9.797464868072489e-06,
|
| 178 |
+
"loss": 2.2036,
|
| 179 |
"step": 100
|
| 180 |
},
|
| 181 |
{
|
| 182 |
"epoch": 1.89,
|
| 183 |
+
"learning_rate": 9.760161688604008e-06,
|
| 184 |
+
"loss": 2.1167,
|
| 185 |
"step": 104
|
| 186 |
},
|
| 187 |
{
|
| 188 |
"epoch": 1.96,
|
| 189 |
+
"learning_rate": 9.719790845697534e-06,
|
| 190 |
+
"loss": 2.133,
|
| 191 |
"step": 108
|
| 192 |
},
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 193 |
{
|
| 194 |
"epoch": 2.04,
|
| 195 |
+
"learning_rate": 9.676378356149733e-06,
|
| 196 |
+
"loss": 2.2253,
|
| 197 |
"step": 112
|
| 198 |
},
|
| 199 |
{
|
| 200 |
"epoch": 2.11,
|
| 201 |
+
"learning_rate": 9.629952196931902e-06,
|
| 202 |
+
"loss": 2.1681,
|
| 203 |
"step": 116
|
| 204 |
},
|
| 205 |
{
|
| 206 |
"epoch": 2.18,
|
| 207 |
+
"learning_rate": 9.580542287160348e-06,
|
| 208 |
+
"loss": 2.3558,
|
| 209 |
"step": 120
|
| 210 |
},
|
| 211 |
{
|
| 212 |
"epoch": 2.25,
|
| 213 |
+
"learning_rate": 9.528180468815155e-06,
|
| 214 |
+
"loss": 2.1262,
|
| 215 |
"step": 124
|
| 216 |
},
|
| 217 |
{
|
| 218 |
"epoch": 2.33,
|
| 219 |
+
"learning_rate": 9.47290048621977e-06,
|
| 220 |
+
"loss": 2.0675,
|
| 221 |
"step": 128
|
| 222 |
},
|
| 223 |
{
|
| 224 |
"epoch": 2.4,
|
| 225 |
+
"learning_rate": 9.414737964294636e-06,
|
| 226 |
+
"loss": 2.0984,
|
| 227 |
"step": 132
|
| 228 |
},
|
| 229 |
{
|
| 230 |
"epoch": 2.47,
|
| 231 |
+
"learning_rate": 9.353730385598887e-06,
|
| 232 |
+
"loss": 2.0763,
|
| 233 |
"step": 136
|
| 234 |
},
|
| 235 |
{
|
| 236 |
"epoch": 2.55,
|
| 237 |
+
"learning_rate": 9.289917066174887e-06,
|
| 238 |
+
"loss": 2.0899,
|
| 239 |
"step": 140
|
| 240 |
},
|
| 241 |
{
|
| 242 |
"epoch": 2.62,
|
| 243 |
+
"learning_rate": 9.223339130211194e-06,
|
| 244 |
+
"loss": 1.9647,
|
| 245 |
"step": 144
|
| 246 |
},
|
| 247 |
{
|
| 248 |
"epoch": 2.69,
|
| 249 |
+
"learning_rate": 9.154039483540273e-06,
|
| 250 |
+
"loss": 2.0805,
|
| 251 |
"step": 148
|
| 252 |
},
|
| 253 |
{
|
| 254 |
"epoch": 2.76,
|
| 255 |
+
"learning_rate": 9.08206278598805e-06,
|
| 256 |
+
"loss": 1.996,
|
| 257 |
"step": 152
|
| 258 |
},
|
| 259 |
{
|
| 260 |
"epoch": 2.84,
|
| 261 |
+
"learning_rate": 9.007455422593077e-06,
|
| 262 |
+
"loss": 2.1735,
|
| 263 |
"step": 156
|
| 264 |
},
|
| 265 |
{
|
| 266 |
"epoch": 2.91,
|
| 267 |
+
"learning_rate": 8.930265473713939e-06,
|
| 268 |
+
"loss": 2.197,
|
| 269 |
"step": 160
|
| 270 |
},
|
| 271 |
{
|
| 272 |
"epoch": 2.98,
|
| 273 |
+
"learning_rate": 8.850542684044078e-06,
|
| 274 |
+
"loss": 2.0502,
|
| 275 |
"step": 164
|
| 276 |
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 3.0,
|
| 279 |
+
"pls_score": 60.4,
|
| 280 |
+
"std": 4.326291714621197,
|
| 281 |
+
"step": 165
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 3.0,
|
| 285 |
+
"eval_loss": 2.0924267768859863,
|
| 286 |
+
"eval_runtime": 4.9011,
|
| 287 |
+
"eval_samples_per_second": 4.693,
|
| 288 |
+
"eval_steps_per_second": 1.224,
|
| 289 |
+
"step": 165
|
| 290 |
+
},
|
| 291 |
{
|
| 292 |
"epoch": 3.05,
|
| 293 |
+
"learning_rate": 8.768338430554083e-06,
|
| 294 |
+
"loss": 1.8989,
|
| 295 |
"step": 168
|
| 296 |
},
|
| 297 |
{
|
| 298 |
"epoch": 3.13,
|
| 299 |
+
"learning_rate": 8.683705689382025e-06,
|
| 300 |
+
"loss": 2.149,
|
| 301 |
"step": 172
|
| 302 |
},
|
| 303 |
{
|
| 304 |
"epoch": 3.2,
|
| 305 |
+
"learning_rate": 8.596699001693257e-06,
|
| 306 |
+
"loss": 2.0518,
|
| 307 |
"step": 176
|
| 308 |
},
|
| 309 |
{
|
| 310 |
"epoch": 3.27,
|
| 311 |
+
"learning_rate": 8.507374438531606e-06,
|
| 312 |
+
"loss": 2.1007,
|
| 313 |
"step": 180
|
| 314 |
},
|
| 315 |
{
|
| 316 |
"epoch": 3.35,
|
| 317 |
+
"learning_rate": 8.415789564684673e-06,
|
| 318 |
+
"loss": 2.0187,
|
| 319 |
"step": 184
|
| 320 |
},
|
| 321 |
{
|
| 322 |
"epoch": 3.42,
|
| 323 |
+
"learning_rate": 8.322003401586463e-06,
|
| 324 |
+
"loss": 1.9728,
|
| 325 |
"step": 188
|
| 326 |
},
|
| 327 |
{
|
| 328 |
"epoch": 3.49,
|
| 329 |
+
"learning_rate": 8.226076389281316e-06,
|
| 330 |
+
"loss": 1.9981,
|
| 331 |
"step": 192
|
| 332 |
},
|
| 333 |
{
|
| 334 |
"epoch": 3.56,
|
| 335 |
+
"learning_rate": 8.128070347473609e-06,
|
| 336 |
+
"loss": 2.0167,
|
| 337 |
"step": 196
|
| 338 |
},
|
| 339 |
{
|
| 340 |
"epoch": 3.64,
|
| 341 |
+
"learning_rate": 8.028048435688333e-06,
|
| 342 |
+
"loss": 2.023,
|
| 343 |
"step": 200
|
| 344 |
},
|
| 345 |
{
|
| 346 |
"epoch": 3.71,
|
| 347 |
+
"learning_rate": 7.92607511256826e-06,
|
| 348 |
+
"loss": 2.0933,
|
| 349 |
"step": 204
|
| 350 |
},
|
| 351 |
{
|
| 352 |
"epoch": 3.78,
|
| 353 |
+
"learning_rate": 7.822216094333847e-06,
|
| 354 |
+
"loss": 2.0343,
|
| 355 |
"step": 208
|
| 356 |
},
|
| 357 |
{
|
| 358 |
"epoch": 3.85,
|
| 359 |
+
"learning_rate": 7.716538312432767e-06,
|
| 360 |
+
"loss": 2.0147,
|
| 361 |
"step": 212
|
| 362 |
},
|
| 363 |
{
|
| 364 |
"epoch": 3.93,
|
| 365 |
+
"learning_rate": 7.60910987040623e-06,
|
| 366 |
+
"loss": 2.0385,
|
| 367 |
"step": 216
|
| 368 |
},
|
| 369 |
{
|
| 370 |
"epoch": 4.0,
|
| 371 |
+
"learning_rate": 7.500000000000001e-06,
|
| 372 |
+
"loss": 1.953,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 373 |
"step": 220
|
| 374 |
},
|
| 375 |
{
|
| 376 |
"epoch": 4.07,
|
| 377 |
+
"learning_rate": 7.3892790165483164e-06,
|
| 378 |
+
"loss": 1.9079,
|
| 379 |
"step": 224
|
| 380 |
},
|
| 381 |
{
|
| 382 |
"epoch": 4.15,
|
| 383 |
+
"learning_rate": 7.2770182736595164e-06,
|
| 384 |
+
"loss": 2.0103,
|
| 385 |
"step": 228
|
| 386 |
},
|
| 387 |
{
|
| 388 |
"epoch": 4.22,
|
| 389 |
+
"learning_rate": 7.163290117232542e-06,
|
| 390 |
+
"loss": 2.0291,
|
| 391 |
"step": 232
|
| 392 |
},
|
| 393 |
{
|
| 394 |
"epoch": 4.29,
|
| 395 |
+
"learning_rate": 7.048167838833977e-06,
|
| 396 |
+
"loss": 1.9614,
|
| 397 |
"step": 236
|
| 398 |
},
|
| 399 |
{
|
| 400 |
"epoch": 4.36,
|
| 401 |
+
"learning_rate": 6.931725628465643e-06,
|
| 402 |
+
"loss": 1.9674,
|
| 403 |
"step": 240
|
| 404 |
},
|
| 405 |
{
|
| 406 |
"epoch": 4.44,
|
| 407 |
+
"learning_rate": 6.814038526753205e-06,
|
| 408 |
+
"loss": 1.8811,
|
| 409 |
"step": 244
|
| 410 |
},
|
| 411 |
{
|
| 412 |
"epoch": 4.51,
|
| 413 |
+
"learning_rate": 6.695182376586603e-06,
|
| 414 |
+
"loss": 1.9888,
|
| 415 |
"step": 248
|
| 416 |
},
|
| 417 |
{
|
| 418 |
"epoch": 4.58,
|
| 419 |
+
"learning_rate": 6.5752337742434644e-06,
|
| 420 |
+
"loss": 1.8913,
|
| 421 |
"step": 252
|
| 422 |
},
|
| 423 |
{
|
| 424 |
"epoch": 4.65,
|
| 425 |
+
"learning_rate": 6.454270020026996e-06,
|
| 426 |
+
"loss": 2.0448,
|
| 427 |
"step": 256
|
| 428 |
},
|
| 429 |
{
|
| 430 |
"epoch": 4.73,
|
| 431 |
+
"learning_rate": 6.332369068450175e-06,
|
| 432 |
+
"loss": 1.9602,
|
| 433 |
"step": 260
|
| 434 |
},
|
| 435 |
{
|
| 436 |
"epoch": 4.8,
|
| 437 |
+
"learning_rate": 6.209609477998339e-06,
|
| 438 |
+
"loss": 2.0635,
|
| 439 |
"step": 264
|
| 440 |
},
|
| 441 |
{
|
| 442 |
"epoch": 4.87,
|
| 443 |
+
"learning_rate": 6.08607036050254e-06,
|
| 444 |
+
"loss": 1.9178,
|
| 445 |
"step": 268
|
| 446 |
},
|
| 447 |
{
|
| 448 |
"epoch": 4.95,
|
| 449 |
+
"learning_rate": 5.961831330156306e-06,
|
| 450 |
+
"loss": 1.9287,
|
| 451 |
"step": 272
|
| 452 |
},
|
| 453 |
{
|
| 454 |
"epoch": 5.0,
|
| 455 |
+
"pls_score": 57.4,
|
| 456 |
+
"std": 3.9880822458921283,
|
| 457 |
"step": 275
|
| 458 |
},
|
| 459 |
{
|
| 460 |
"epoch": 5.0,
|
| 461 |
+
"eval_loss": 2.0102691650390625,
|
| 462 |
+
"eval_runtime": 4.9168,
|
| 463 |
+
"eval_samples_per_second": 4.678,
|
| 464 |
+
"eval_steps_per_second": 1.22,
|
| 465 |
"step": 275
|
| 466 |
},
|
| 467 |
{
|
| 468 |
"epoch": 5.02,
|
| 469 |
+
"learning_rate": 5.8369724522086545e-06,
|
| 470 |
+
"loss": 1.8716,
|
| 471 |
"step": 276
|
| 472 |
},
|
| 473 |
{
|
| 474 |
"epoch": 5.09,
|
| 475 |
+
"learning_rate": 5.711574191366427e-06,
|
| 476 |
+
"loss": 1.9052,
|
| 477 |
"step": 280
|
| 478 |
},
|
| 479 |
{
|
| 480 |
"epoch": 5.16,
|
| 481 |
+
"learning_rate": 5.585717359939192e-06,
|
| 482 |
+
"loss": 1.8104,
|
| 483 |
"step": 284
|
| 484 |
},
|
| 485 |
{
|
| 486 |
"epoch": 5.24,
|
| 487 |
+
"learning_rate": 5.459483065760138e-06,
|
| 488 |
+
"loss": 1.9212,
|
| 489 |
"step": 288
|
| 490 |
},
|
| 491 |
{
|
| 492 |
"epoch": 5.31,
|
| 493 |
+
"learning_rate": 5.33295265991652e-06,
|
| 494 |
+
"loss": 1.9339,
|
| 495 |
"step": 292
|
| 496 |
},
|
| 497 |
{
|
| 498 |
"epoch": 5.38,
|
| 499 |
+
"learning_rate": 5.206207684323337e-06,
|
| 500 |
+
"loss": 1.9414,
|
| 501 |
"step": 296
|
| 502 |
},
|
| 503 |
{
|
| 504 |
"epoch": 5.45,
|
| 505 |
+
"learning_rate": 5.07932981917404e-06,
|
| 506 |
+
"loss": 1.8935,
|
| 507 |
"step": 300
|
| 508 |
},
|
| 509 |
{
|
| 510 |
"epoch": 5.53,
|
| 511 |
+
"learning_rate": 4.952400830302117e-06,
|
| 512 |
+
"loss": 2.0557,
|
| 513 |
"step": 304
|
| 514 |
},
|
| 515 |
{
|
| 516 |
"epoch": 5.6,
|
| 517 |
+
"learning_rate": 4.825502516487497e-06,
|
| 518 |
+
"loss": 1.8989,
|
| 519 |
"step": 308
|
| 520 |
},
|
| 521 |
{
|
| 522 |
"epoch": 5.67,
|
| 523 |
+
"learning_rate": 4.6987166567417085e-06,
|
| 524 |
+
"loss": 1.8834,
|
| 525 |
"step": 312
|
| 526 |
},
|
| 527 |
{
|
| 528 |
"epoch": 5.75,
|
| 529 |
+
"learning_rate": 4.572124957605803e-06,
|
| 530 |
+
"loss": 1.8287,
|
| 531 |
"step": 316
|
| 532 |
},
|
| 533 |
{
|
| 534 |
"epoch": 5.82,
|
| 535 |
+
"learning_rate": 4.445809000494945e-06,
|
| 536 |
+
"loss": 1.944,
|
| 537 |
"step": 320
|
| 538 |
},
|
| 539 |
{
|
| 540 |
"epoch": 5.89,
|
| 541 |
+
"learning_rate": 4.319850189123681e-06,
|
| 542 |
+
"loss": 1.9083,
|
| 543 |
"step": 324
|
| 544 |
},
|
| 545 |
{
|
| 546 |
"epoch": 5.96,
|
| 547 |
+
"learning_rate": 4.194329697045681e-06,
|
| 548 |
+
"loss": 2.0021,
|
| 549 |
"step": 328
|
| 550 |
},
|
| 551 |
{
|
| 552 |
"epoch": 6.0,
|
| 553 |
+
"pls_score": 58.4,
|
| 554 |
+
"std": 4.227150340359329,
|
| 555 |
"step": 330
|
| 556 |
},
|
| 557 |
{
|
| 558 |
"epoch": 6.0,
|
| 559 |
+
"eval_loss": 1.9934663772583008,
|
| 560 |
+
"eval_runtime": 4.9133,
|
| 561 |
+
"eval_samples_per_second": 4.681,
|
| 562 |
+
"eval_steps_per_second": 1.221,
|
| 563 |
"step": 330
|
| 564 |
},
|
| 565 |
{
|
| 566 |
"epoch": 6.04,
|
| 567 |
+
"learning_rate": 4.06932841534185e-06,
|
| 568 |
+
"loss": 2.0032,
|
| 569 |
"step": 332
|
| 570 |
},
|
| 571 |
{
|
| 572 |
"epoch": 6.11,
|
| 573 |
+
"learning_rate": 3.944926900490452e-06,
|
| 574 |
+
"loss": 1.968,
|
| 575 |
"step": 336
|
| 576 |
},
|
| 577 |
{
|
| 578 |
"epoch": 6.18,
|
| 579 |
+
"learning_rate": 3.821205322452863e-06,
|
| 580 |
+
"loss": 1.8771,
|
| 581 |
"step": 340
|
| 582 |
},
|
| 583 |
{
|
| 584 |
"epoch": 6.25,
|
| 585 |
+
"learning_rate": 3.69824341300844e-06,
|
| 586 |
+
"loss": 1.948,
|
| 587 |
"step": 344
|
| 588 |
},
|
| 589 |
{
|
| 590 |
"epoch": 6.33,
|
| 591 |
+
"learning_rate": 3.5761204143717387e-06,
|
| 592 |
+
"loss": 1.9014,
|
| 593 |
"step": 348
|
| 594 |
},
|
| 595 |
{
|
| 596 |
"epoch": 6.4,
|
| 597 |
+
"learning_rate": 3.4549150281252635e-06,
|
| 598 |
+
"loss": 1.9815,
|
| 599 |
"step": 352
|
| 600 |
},
|
| 601 |
{
|
| 602 |
"epoch": 6.47,
|
| 603 |
+
"learning_rate": 3.3347053645005965e-06,
|
| 604 |
+
"loss": 1.9389,
|
| 605 |
"step": 356
|
| 606 |
},
|
| 607 |
{
|
| 608 |
"epoch": 6.55,
|
| 609 |
+
"learning_rate": 3.2155688920406415e-06,
|
| 610 |
+
"loss": 1.7294,
|
| 611 |
"step": 360
|
| 612 |
},
|
| 613 |
{
|
| 614 |
"epoch": 6.62,
|
| 615 |
+
"learning_rate": 3.097582387675385e-06,
|
| 616 |
+
"loss": 1.8354,
|
| 617 |
"step": 364
|
| 618 |
},
|
| 619 |
{
|
| 620 |
"epoch": 6.69,
|
| 621 |
+
"learning_rate": 2.980821887243377e-06,
|
| 622 |
+
"loss": 1.9155,
|
| 623 |
"step": 368
|
| 624 |
},
|
| 625 |
{
|
| 626 |
"epoch": 6.76,
|
| 627 |
+
"learning_rate": 2.8653626364907918e-06,
|
| 628 |
+
"loss": 1.9047,
|
| 629 |
"step": 372
|
| 630 |
},
|
| 631 |
{
|
| 632 |
"epoch": 6.84,
|
| 633 |
+
"learning_rate": 2.751279042579672e-06,
|
| 634 |
+
"loss": 1.949,
|
| 635 |
"step": 376
|
| 636 |
},
|
| 637 |
{
|
| 638 |
"epoch": 6.91,
|
| 639 |
+
"learning_rate": 2.6386446261365874e-06,
|
| 640 |
+
"loss": 1.8363,
|
| 641 |
"step": 380
|
| 642 |
},
|
| 643 |
{
|
| 644 |
"epoch": 6.98,
|
| 645 |
+
"learning_rate": 2.527531973872617e-06,
|
| 646 |
+
"loss": 1.8199,
|
| 647 |
"step": 384
|
| 648 |
},
|
| 649 |
{
|
| 650 |
"epoch": 7.0,
|
| 651 |
+
"pls_score": 59.6,
|
| 652 |
+
"std": 4.098389927764319,
|
| 653 |
"step": 385
|
| 654 |
},
|
| 655 |
{
|
| 656 |
"epoch": 7.0,
|
| 657 |
+
"eval_loss": 1.9878090620040894,
|
| 658 |
+
"eval_runtime": 4.9106,
|
| 659 |
+
"eval_samples_per_second": 4.684,
|
| 660 |
+
"eval_steps_per_second": 1.222,
|
| 661 |
"step": 385
|
| 662 |
},
|
| 663 |
{
|
| 664 |
"epoch": 7.05,
|
| 665 |
+
"learning_rate": 2.418012691805191e-06,
|
| 666 |
+
"loss": 2.0762,
|
| 667 |
"step": 388
|
| 668 |
},
|
| 669 |
{
|
| 670 |
"epoch": 7.13,
|
| 671 |
+
"learning_rate": 2.310157359111938e-06,
|
| 672 |
+
"loss": 1.8485,
|
| 673 |
"step": 392
|
| 674 |
},
|
| 675 |
{
|
| 676 |
"epoch": 7.2,
|
| 677 |
+
"learning_rate": 2.204035482646267e-06,
|
| 678 |
+
"loss": 1.9232,
|
| 679 |
"step": 396
|
| 680 |
},
|
| 681 |
{
|
| 682 |
"epoch": 7.27,
|
| 683 |
+
"learning_rate": 2.09971545214401e-06,
|
| 684 |
+
"loss": 1.8724,
|
| 685 |
"step": 400
|
| 686 |
},
|
| 687 |
{
|
| 688 |
"epoch": 7.35,
|
| 689 |
+
"learning_rate": 1.9972644961499853e-06,
|
| 690 |
+
"loss": 1.8075,
|
| 691 |
"step": 404
|
| 692 |
},
|
| 693 |
{
|
| 694 |
"epoch": 7.42,
|
| 695 |
+
"learning_rate": 1.8967486386928819e-06,
|
| 696 |
+
"loss": 1.8526,
|
| 697 |
"step": 408
|
| 698 |
},
|
| 699 |
{
|
| 700 |
"epoch": 7.49,
|
| 701 |
+
"learning_rate": 1.798232656736389e-06,
|
| 702 |
+
"loss": 1.899,
|
| 703 |
"step": 412
|
| 704 |
},
|
| 705 |
{
|
| 706 |
"epoch": 7.56,
|
| 707 |
+
"learning_rate": 1.7017800384339928e-06,
|
| 708 |
+
"loss": 1.7487,
|
| 709 |
"step": 416
|
| 710 |
},
|
| 711 |
{
|
| 712 |
"epoch": 7.64,
|
| 713 |
+
"learning_rate": 1.6074529422143398e-06,
|
| 714 |
+
"loss": 1.9218,
|
| 715 |
"step": 420
|
| 716 |
},
|
| 717 |
{
|
| 718 |
"epoch": 7.71,
|
| 719 |
+
"learning_rate": 1.5153121567235334e-06,
|
| 720 |
+
"loss": 1.8776,
|
| 721 |
"step": 424
|
| 722 |
},
|
| 723 |
{
|
| 724 |
"epoch": 7.78,
|
| 725 |
+
"learning_rate": 1.4254170616501828e-06,
|
| 726 |
+
"loss": 1.761,
|
| 727 |
"step": 428
|
| 728 |
},
|
| 729 |
{
|
| 730 |
"epoch": 7.85,
|
| 731 |
+
"learning_rate": 1.3378255894584463e-06,
|
| 732 |
+
"loss": 1.9254,
|
| 733 |
"step": 432
|
| 734 |
},
|
| 735 |
{
|
| 736 |
"epoch": 7.93,
|
| 737 |
+
"learning_rate": 1.2525941880537307e-06,
|
| 738 |
+
"loss": 1.8895,
|
| 739 |
"step": 436
|
| 740 |
},
|
| 741 |
{
|
| 742 |
"epoch": 8.0,
|
| 743 |
+
"learning_rate": 1.1697777844051105e-06,
|
| 744 |
+
"loss": 1.9559,
|
| 745 |
"step": 440
|
| 746 |
},
|
| 747 |
{
|
| 748 |
"epoch": 8.0,
|
| 749 |
+
"pls_score": 60.0,
|
| 750 |
+
"std": 4.233202097703344,
|
| 751 |
"step": 440
|
| 752 |
},
|
| 753 |
{
|
| 754 |
"epoch": 8.0,
|
| 755 |
+
"eval_loss": 1.9857509136199951,
|
| 756 |
+
"eval_runtime": 4.912,
|
| 757 |
+
"eval_samples_per_second": 4.682,
|
| 758 |
+
"eval_steps_per_second": 1.222,
|
| 759 |
"step": 440
|
| 760 |
},
|
| 761 |
{
|
| 762 |
"epoch": 8.07,
|
| 763 |
+
"learning_rate": 1.0894297491479044e-06,
|
| 764 |
+
"loss": 1.8595,
|
| 765 |
"step": 444
|
| 766 |
},
|
| 767 |
{
|
| 768 |
"epoch": 8.15,
|
| 769 |
+
"learning_rate": 1.0116018621892237e-06,
|
| 770 |
+
"loss": 1.7881,
|
| 771 |
"step": 448
|
| 772 |
},
|
| 773 |
{
|
| 774 |
"epoch": 8.22,
|
| 775 |
+
"learning_rate": 9.363442793386606e-07,
|
| 776 |
+
"loss": 1.8575,
|
| 777 |
"step": 452
|
| 778 |
},
|
| 779 |
{
|
| 780 |
"epoch": 8.29,
|
| 781 |
+
"learning_rate": 8.637054999856148e-07,
|
| 782 |
+
"loss": 1.8457,
|
| 783 |
"step": 456
|
| 784 |
},
|
| 785 |
{
|
| 786 |
"epoch": 8.36,
|
| 787 |
+
"learning_rate": 7.937323358440935e-07,
|
| 788 |
+
"loss": 1.8717,
|
| 789 |
"step": 460
|
| 790 |
},
|
| 791 |
{
|
| 792 |
"epoch": 8.44,
|
| 793 |
+
"learning_rate": 7.264698807851328e-07,
|
| 794 |
+
"loss": 1.887,
|
| 795 |
"step": 464
|
| 796 |
},
|
| 797 |
{
|
| 798 |
"epoch": 8.51,
|
| 799 |
+
"learning_rate": 6.619614817762537e-07,
|
| 800 |
+
"loss": 1.8931,
|
| 801 |
"step": 468
|
| 802 |
},
|
| 803 |
{
|
| 804 |
"epoch": 8.58,
|
| 805 |
+
"learning_rate": 6.002487109467347e-07,
|
| 806 |
+
"loss": 1.8454,
|
| 807 |
"step": 472
|
| 808 |
},
|
| 809 |
{
|
| 810 |
"epoch": 8.65,
|
| 811 |
+
"learning_rate": 5.413713387966329e-07,
|
| 812 |
+
"loss": 1.9584,
|
| 813 |
"step": 476
|
| 814 |
},
|
| 815 |
{
|
| 816 |
"epoch": 8.73,
|
| 817 |
+
"learning_rate": 4.853673085668947e-07,
|
| 818 |
+
"loss": 1.9234,
|
| 819 |
"step": 480
|
| 820 |
},
|
| 821 |
{
|
| 822 |
"epoch": 8.8,
|
| 823 |
+
"learning_rate": 4.322727117869951e-07,
|
| 824 |
+
"loss": 1.8912,
|
| 825 |
"step": 484
|
| 826 |
},
|
| 827 |
{
|
| 828 |
"epoch": 8.87,
|
| 829 |
+
"learning_rate": 3.821217650159453e-07,
|
| 830 |
+
"loss": 1.8707,
|
| 831 |
"step": 488
|
| 832 |
},
|
| 833 |
{
|
| 834 |
"epoch": 8.95,
|
| 835 |
+
"learning_rate": 3.3494678779157464e-07,
|
| 836 |
+
"loss": 1.8229,
|
| 837 |
"step": 492
|
| 838 |
},
|
| 839 |
{
|
| 840 |
"epoch": 9.0,
|
| 841 |
+
"pls_score": 59.2,
|
| 842 |
+
"std": 4.343639027359432,
|
| 843 |
"step": 495
|
| 844 |
},
|
| 845 |
{
|
| 846 |
"epoch": 9.0,
|
| 847 |
+
"eval_loss": 1.9853376150131226,
|
| 848 |
+
"eval_runtime": 4.916,
|
| 849 |
+
"eval_samples_per_second": 4.679,
|
| 850 |
+
"eval_steps_per_second": 1.221,
|
| 851 |
"step": 495
|
| 852 |
},
|
| 853 |
{
|
| 854 |
"epoch": 9.02,
|
| 855 |
+
"learning_rate": 2.9077818180237693e-07,
|
| 856 |
+
"loss": 1.7843,
|
| 857 |
"step": 496
|
| 858 |
},
|
| 859 |
{
|
| 860 |
"epoch": 9.09,
|
| 861 |
+
"learning_rate": 2.4964441129527337e-07,
|
| 862 |
+
"loss": 1.9182,
|
| 863 |
"step": 500
|
| 864 |
},
|
| 865 |
{
|
| 866 |
"epoch": 9.16,
|
| 867 |
+
"learning_rate": 2.1157198473197417e-07,
|
| 868 |
+
"loss": 1.9246,
|
| 869 |
"step": 504
|
| 870 |
},
|
| 871 |
{
|
| 872 |
"epoch": 9.24,
|
| 873 |
+
"learning_rate": 1.765854377057219e-07,
|
| 874 |
+
"loss": 1.8744,
|
| 875 |
"step": 508
|
| 876 |
},
|
| 877 |
{
|
| 878 |
"epoch": 9.31,
|
| 879 |
+
"learning_rate": 1.4470731712944885e-07,
|
| 880 |
+
"loss": 1.872,
|
| 881 |
"step": 512
|
| 882 |
},
|
| 883 |
{
|
| 884 |
"epoch": 9.38,
|
| 885 |
+
"learning_rate": 1.1595816670552429e-07,
|
| 886 |
+
"loss": 1.8631,
|
| 887 |
"step": 516
|
| 888 |
},
|
| 889 |
{
|
| 890 |
"epoch": 9.45,
|
| 891 |
+
"learning_rate": 9.035651368646647e-08,
|
| 892 |
+
"loss": 1.889,
|
| 893 |
"step": 520
|
| 894 |
},
|
| 895 |
{
|
| 896 |
"epoch": 9.53,
|
| 897 |
+
"learning_rate": 6.791885693514134e-08,
|
| 898 |
+
"loss": 1.9402,
|
| 899 |
"step": 524
|
| 900 |
},
|
| 901 |
{
|
| 902 |
"epoch": 9.6,
|
| 903 |
+
"learning_rate": 4.865965629214819e-08,
|
| 904 |
+
"loss": 1.7325,
|
| 905 |
"step": 528
|
| 906 |
},
|
| 907 |
{
|
| 908 |
"epoch": 9.67,
|
| 909 |
+
"learning_rate": 3.25913232572489e-08,
|
| 910 |
+
"loss": 1.8187,
|
| 911 |
"step": 532
|
| 912 |
},
|
| 913 |
{
|
| 914 |
"epoch": 9.75,
|
| 915 |
+
"learning_rate": 1.9724212990830938e-08,
|
| 916 |
+
"loss": 1.877,
|
| 917 |
"step": 536
|
| 918 |
},
|
| 919 |
{
|
| 920 |
"epoch": 9.82,
|
| 921 |
+
"learning_rate": 1.006661764057837e-08,
|
| 922 |
+
"loss": 1.9245,
|
| 923 |
"step": 540
|
| 924 |
},
|
| 925 |
{
|
| 926 |
"epoch": 9.89,
|
| 927 |
+
"learning_rate": 3.6247609976319818e-09,
|
| 928 |
+
"loss": 1.8485,
|
| 929 |
"step": 544
|
| 930 |
},
|
| 931 |
{
|
| 932 |
"epoch": 9.96,
|
| 933 |
+
"learning_rate": 4.027944857032395e-10,
|
| 934 |
+
"loss": 1.8178,
|
| 935 |
"step": 548
|
| 936 |
},
|
| 937 |
{
|
| 938 |
"epoch": 10.0,
|
| 939 |
+
"pls_score": 56.326530612244895,
|
| 940 |
+
"std": 4.189332932171676,
|
| 941 |
"step": 550
|
| 942 |
},
|
| 943 |
{
|
| 944 |
"epoch": 10.0,
|
| 945 |
+
"eval_loss": 1.9850994348526,
|
| 946 |
+
"eval_runtime": 4.916,
|
| 947 |
+
"eval_samples_per_second": 4.679,
|
| 948 |
+
"eval_steps_per_second": 1.22,
|
| 949 |
"step": 550
|
| 950 |
},
|
| 951 |
{
|
| 952 |
"epoch": 10.0,
|
| 953 |
"step": 550,
|
| 954 |
+
"total_flos": 1.8777166830084096e+16,
|
| 955 |
+
"train_loss": 1.1392531923814253,
|
| 956 |
+
"train_runtime": 18256.4954,
|
| 957 |
+
"train_samples_per_second": 0.119,
|
| 958 |
+
"train_steps_per_second": 0.03
|
| 959 |
}
|
| 960 |
],
|
| 961 |
"logging_steps": 4,
|
|
|
|
| 963 |
"num_input_tokens_seen": 0,
|
| 964 |
"num_train_epochs": 10,
|
| 965 |
"save_steps": 55,
|
| 966 |
+
"total_flos": 1.8777166830084096e+16,
|
| 967 |
"train_batch_size": 4,
|
| 968 |
"trial_name": null,
|
| 969 |
"trial_params": null
|
training_args.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 4792
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7ba76d65b0f816a12acd6deb020064c70467d4ba65c899f758b8c99c2c6e5f2d
|
| 3 |
size 4792
|