Model save
Browse files- README.md +22 -26
- all_results.json +8 -8
- eval_results.json +4 -4
- train_results.json +4 -4
- trainer_state.json +141 -141
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
|
|
@@ -20,9 +16,9 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 20 |
|
| 21 |
# Yi-6B-ruozhiba
|
| 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: 4.
|
| 26 |
|
| 27 |
## Model description
|
| 28 |
|
|
@@ -54,26 +50,26 @@ The following hyperparameters were used during training:
|
|
| 54 |
|
| 55 |
| Training Loss | Epoch | Step | Validation Loss |
|
| 56 |
|:-------------:|:-----:|:----:|:---------------:|
|
| 57 |
-
| 2.
|
| 58 |
-
|
|
| 59 |
-
| 1.
|
| 60 |
-
| 1.
|
| 61 |
-
| 1.
|
| 62 |
-
| 0.
|
| 63 |
-
| 0.
|
| 64 |
-
| 0.
|
| 65 |
-
| 0.
|
| 66 |
-
| 0.
|
| 67 |
-
| 0.
|
| 68 |
-
| 0.
|
| 69 |
-
| 0.
|
| 70 |
-
| 0.
|
| 71 |
-
| 0.
|
| 72 |
-
| 0.
|
| 73 |
-
| 0.
|
| 74 |
-
| 0.
|
| 75 |
-
| 0.
|
| 76 |
-
| 0.
|
| 77 |
|
| 78 |
|
| 79 |
### 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
|
|
|
|
| 16 |
|
| 17 |
# Yi-6B-ruozhiba
|
| 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: 4.2052
|
| 22 |
|
| 23 |
## Model description
|
| 24 |
|
|
|
|
| 50 |
|
| 51 |
| Training Loss | Epoch | Step | Validation Loss |
|
| 52 |
|:-------------:|:-----:|:----:|:---------------:|
|
| 53 |
+
| 2.7672 | 1.0 | 55 | 2.3123 |
|
| 54 |
+
| 2.0304 | 2.0 | 110 | 1.9674 |
|
| 55 |
+
| 1.7981 | 3.0 | 165 | 1.9422 |
|
| 56 |
+
| 1.5825 | 4.0 | 220 | 2.0125 |
|
| 57 |
+
| 1.2811 | 5.0 | 275 | 2.2657 |
|
| 58 |
+
| 0.928 | 6.0 | 330 | 2.5271 |
|
| 59 |
+
| 0.6688 | 7.0 | 385 | 2.7800 |
|
| 60 |
+
| 0.4595 | 8.0 | 440 | 3.1072 |
|
| 61 |
+
| 0.3117 | 9.0 | 495 | 3.3261 |
|
| 62 |
+
| 0.242 | 10.0 | 550 | 3.5036 |
|
| 63 |
+
| 0.1718 | 11.0 | 605 | 3.6742 |
|
| 64 |
+
| 0.157 | 12.0 | 660 | 3.7895 |
|
| 65 |
+
| 0.1188 | 13.0 | 715 | 3.9184 |
|
| 66 |
+
| 0.1296 | 14.0 | 770 | 3.9965 |
|
| 67 |
+
| 0.1219 | 15.0 | 825 | 4.1147 |
|
| 68 |
+
| 0.1 | 16.0 | 880 | 4.1376 |
|
| 69 |
+
| 0.1155 | 17.0 | 935 | 4.1758 |
|
| 70 |
+
| 0.1083 | 18.0 | 990 | 4.2016 |
|
| 71 |
+
| 0.0961 | 19.0 | 1045 | 4.2047 |
|
| 72 |
+
| 0.0917 | 20.0 | 1100 | 4.2052 |
|
| 73 |
|
| 74 |
|
| 75 |
### Framework versions
|
all_results.json
CHANGED
|
@@ -1,13 +1,13 @@
|
|
| 1 |
{
|
| 2 |
"epoch": 20.0,
|
| 3 |
-
"eval_loss": 4.
|
| 4 |
-
"eval_runtime": 4.
|
| 5 |
"eval_samples": 23,
|
| 6 |
-
"eval_samples_per_second": 4.
|
| 7 |
-
"eval_steps_per_second": 1.
|
| 8 |
-
"train_loss": 0.
|
| 9 |
-
"train_runtime":
|
| 10 |
"train_samples": 217,
|
| 11 |
-
"train_samples_per_second": 1.
|
| 12 |
-
"train_steps_per_second": 0.
|
| 13 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"epoch": 20.0,
|
| 3 |
+
"eval_loss": 4.205177307128906,
|
| 4 |
+
"eval_runtime": 4.9959,
|
| 5 |
"eval_samples": 23,
|
| 6 |
+
"eval_samples_per_second": 4.604,
|
| 7 |
+
"eval_steps_per_second": 1.201,
|
| 8 |
+
"train_loss": 0.6597742699493061,
|
| 9 |
+
"train_runtime": 3169.8576,
|
| 10 |
"train_samples": 217,
|
| 11 |
+
"train_samples_per_second": 1.369,
|
| 12 |
+
"train_steps_per_second": 0.347
|
| 13 |
}
|
eval_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
"epoch": 20.0,
|
| 3 |
-
"eval_loss": 4.
|
| 4 |
-
"eval_runtime": 4.
|
| 5 |
"eval_samples": 23,
|
| 6 |
-
"eval_samples_per_second": 4.
|
| 7 |
-
"eval_steps_per_second": 1.
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"epoch": 20.0,
|
| 3 |
+
"eval_loss": 4.205177307128906,
|
| 4 |
+
"eval_runtime": 4.9959,
|
| 5 |
"eval_samples": 23,
|
| 6 |
+
"eval_samples_per_second": 4.604,
|
| 7 |
+
"eval_steps_per_second": 1.201
|
| 8 |
}
|
train_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
"epoch": 20.0,
|
| 3 |
-
"train_loss": 0.
|
| 4 |
-
"train_runtime":
|
| 5 |
"train_samples": 217,
|
| 6 |
-
"train_samples_per_second": 1.
|
| 7 |
-
"train_steps_per_second": 0.
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"epoch": 20.0,
|
| 3 |
+
"train_loss": 0.6597742699493061,
|
| 4 |
+
"train_runtime": 3169.8576,
|
| 5 |
"train_samples": 217,
|
| 6 |
+
"train_samples_per_second": 1.369,
|
| 7 |
+
"train_steps_per_second": 0.347
|
| 8 |
}
|
trainer_state.json
CHANGED
|
@@ -11,19 +11,19 @@
|
|
| 11 |
{
|
| 12 |
"epoch": 0.02,
|
| 13 |
"learning_rate": 4.545454545454545e-07,
|
| 14 |
-
"loss": 2.
|
| 15 |
"step": 1
|
| 16 |
},
|
| 17 |
{
|
| 18 |
"epoch": 0.36,
|
| 19 |
"learning_rate": 9.090909090909091e-06,
|
| 20 |
-
"loss": 2.
|
| 21 |
"step": 20
|
| 22 |
},
|
| 23 |
{
|
| 24 |
"epoch": 0.73,
|
| 25 |
"learning_rate": 1.8181818181818182e-05,
|
| 26 |
-
"loss": 2.
|
| 27 |
"step": 40
|
| 28 |
},
|
| 29 |
{
|
|
@@ -33,28 +33,28 @@
|
|
| 33 |
},
|
| 34 |
{
|
| 35 |
"epoch": 1.0,
|
| 36 |
-
"eval_loss": 2.
|
| 37 |
-
"eval_runtime": 4.
|
| 38 |
-
"eval_samples_per_second": 4.
|
| 39 |
-
"eval_steps_per_second": 1.
|
| 40 |
"step": 55
|
| 41 |
},
|
| 42 |
{
|
| 43 |
"epoch": 1.09,
|
| 44 |
"learning_rate": 2.7272727272727273e-05,
|
| 45 |
-
"loss": 2.
|
| 46 |
"step": 60
|
| 47 |
},
|
| 48 |
{
|
| 49 |
"epoch": 1.45,
|
| 50 |
"learning_rate": 3.6363636363636364e-05,
|
| 51 |
-
"loss": 2.
|
| 52 |
"step": 80
|
| 53 |
},
|
| 54 |
{
|
| 55 |
"epoch": 1.82,
|
| 56 |
"learning_rate": 4.545454545454546e-05,
|
| 57 |
-
"loss":
|
| 58 |
"step": 100
|
| 59 |
},
|
| 60 |
{
|
|
@@ -64,28 +64,28 @@
|
|
| 64 |
},
|
| 65 |
{
|
| 66 |
"epoch": 2.0,
|
| 67 |
-
"eval_loss": 1.
|
| 68 |
-
"eval_runtime": 4.
|
| 69 |
-
"eval_samples_per_second": 4.
|
| 70 |
-
"eval_steps_per_second": 1.
|
| 71 |
"step": 110
|
| 72 |
},
|
| 73 |
{
|
| 74 |
"epoch": 2.18,
|
| 75 |
"learning_rate": 4.9987413559579636e-05,
|
| 76 |
-
"loss": 1.
|
| 77 |
"step": 120
|
| 78 |
},
|
| 79 |
{
|
| 80 |
"epoch": 2.55,
|
| 81 |
"learning_rate": 4.988679806432712e-05,
|
| 82 |
-
"loss": 1.
|
| 83 |
"step": 140
|
| 84 |
},
|
| 85 |
{
|
| 86 |
"epoch": 2.91,
|
| 87 |
"learning_rate": 4.968597221690986e-05,
|
| 88 |
-
"loss": 1.
|
| 89 |
"step": 160
|
| 90 |
},
|
| 91 |
{
|
|
@@ -95,10 +95,10 @@
|
|
| 95 |
},
|
| 96 |
{
|
| 97 |
"epoch": 3.0,
|
| 98 |
-
"eval_loss": 1.
|
| 99 |
-
"eval_runtime": 4.
|
| 100 |
-
"eval_samples_per_second": 4.
|
| 101 |
-
"eval_steps_per_second": 1.
|
| 102 |
"step": 165
|
| 103 |
},
|
| 104 |
{
|
|
@@ -110,13 +110,13 @@
|
|
| 110 |
{
|
| 111 |
"epoch": 3.64,
|
| 112 |
"learning_rate": 4.898732434036244e-05,
|
| 113 |
-
"loss": 1.
|
| 114 |
"step": 200
|
| 115 |
},
|
| 116 |
{
|
| 117 |
"epoch": 4.0,
|
| 118 |
"learning_rate": 4.849231551964771e-05,
|
| 119 |
-
"loss": 1.
|
| 120 |
"step": 220
|
| 121 |
},
|
| 122 |
{
|
|
@@ -126,22 +126,22 @@
|
|
| 126 |
},
|
| 127 |
{
|
| 128 |
"epoch": 4.0,
|
| 129 |
-
"eval_loss":
|
| 130 |
-
"eval_runtime": 4.
|
| 131 |
-
"eval_samples_per_second": 4.
|
| 132 |
-
"eval_steps_per_second": 1.
|
| 133 |
"step": 220
|
| 134 |
},
|
| 135 |
{
|
| 136 |
"epoch": 4.36,
|
| 137 |
"learning_rate": 4.790271143580174e-05,
|
| 138 |
-
"loss": 1.
|
| 139 |
"step": 240
|
| 140 |
},
|
| 141 |
{
|
| 142 |
"epoch": 4.73,
|
| 143 |
"learning_rate": 4.722088621637309e-05,
|
| 144 |
-
"loss": 1.
|
| 145 |
"step": 260
|
| 146 |
},
|
| 147 |
{
|
|
@@ -151,28 +151,28 @@
|
|
| 151 |
},
|
| 152 |
{
|
| 153 |
"epoch": 5.0,
|
| 154 |
-
"eval_loss": 2.
|
| 155 |
-
"eval_runtime": 4.
|
| 156 |
-
"eval_samples_per_second": 4.
|
| 157 |
-
"eval_steps_per_second": 1.
|
| 158 |
"step": 275
|
| 159 |
},
|
| 160 |
{
|
| 161 |
"epoch": 5.09,
|
| 162 |
"learning_rate": 4.644958533087443e-05,
|
| 163 |
-
"loss": 1.
|
| 164 |
"step": 280
|
| 165 |
},
|
| 166 |
{
|
| 167 |
"epoch": 5.45,
|
| 168 |
"learning_rate": 4.559191453574582e-05,
|
| 169 |
-
"loss": 0.
|
| 170 |
"step": 300
|
| 171 |
},
|
| 172 |
{
|
| 173 |
"epoch": 5.82,
|
| 174 |
"learning_rate": 4.465132736856969e-05,
|
| 175 |
-
"loss": 0.
|
| 176 |
"step": 320
|
| 177 |
},
|
| 178 |
{
|
|
@@ -182,28 +182,28 @@
|
|
| 182 |
},
|
| 183 |
{
|
| 184 |
"epoch": 6.0,
|
| 185 |
-
"eval_loss": 2.
|
| 186 |
-
"eval_runtime": 4.
|
| 187 |
-
"eval_samples_per_second": 4.
|
| 188 |
-
"eval_steps_per_second": 1.
|
| 189 |
"step": 330
|
| 190 |
},
|
| 191 |
{
|
| 192 |
"epoch": 6.18,
|
| 193 |
"learning_rate": 4.3631611241893874e-05,
|
| 194 |
-
"loss": 0.
|
| 195 |
"step": 340
|
| 196 |
},
|
| 197 |
{
|
| 198 |
"epoch": 6.55,
|
| 199 |
"learning_rate": 4.2536872192658036e-05,
|
| 200 |
-
"loss": 0.
|
| 201 |
"step": 360
|
| 202 |
},
|
| 203 |
{
|
| 204 |
"epoch": 6.91,
|
| 205 |
"learning_rate": 4.137151834863213e-05,
|
| 206 |
-
"loss": 0.
|
| 207 |
"step": 380
|
| 208 |
},
|
| 209 |
{
|
|
@@ -213,28 +213,28 @@
|
|
| 213 |
},
|
| 214 |
{
|
| 215 |
"epoch": 7.0,
|
| 216 |
-
"eval_loss": 2.
|
| 217 |
-
"eval_runtime":
|
| 218 |
-
"eval_samples_per_second": 4.
|
| 219 |
-
"eval_steps_per_second": 1.
|
| 220 |
"step": 385
|
| 221 |
},
|
| 222 |
{
|
| 223 |
"epoch": 7.27,
|
| 224 |
"learning_rate": 4.014024217844167e-05,
|
| 225 |
-
"loss": 0.
|
| 226 |
"step": 400
|
| 227 |
},
|
| 228 |
{
|
| 229 |
"epoch": 7.64,
|
| 230 |
"learning_rate": 3.884800159665276e-05,
|
| 231 |
-
"loss": 0.
|
| 232 |
"step": 420
|
| 233 |
},
|
| 234 |
{
|
| 235 |
"epoch": 8.0,
|
| 236 |
"learning_rate": 3.7500000000000003e-05,
|
| 237 |
-
"loss": 0.
|
| 238 |
"step": 440
|
| 239 |
},
|
| 240 |
{
|
|
@@ -244,22 +244,22 @@
|
|
| 244 |
},
|
| 245 |
{
|
| 246 |
"epoch": 8.0,
|
| 247 |
-
"eval_loss": 3.
|
| 248 |
-
"eval_runtime": 4.
|
| 249 |
-
"eval_samples_per_second": 4.
|
| 250 |
-
"eval_steps_per_second": 1.
|
| 251 |
"step": 440
|
| 252 |
},
|
| 253 |
{
|
| 254 |
"epoch": 8.36,
|
| 255 |
"learning_rate": 3.610166531514436e-05,
|
| 256 |
-
"loss": 0.
|
| 257 |
"step": 460
|
| 258 |
},
|
| 259 |
{
|
| 260 |
"epoch": 8.73,
|
| 261 |
"learning_rate": 3.465862814232822e-05,
|
| 262 |
-
"loss": 0.
|
| 263 |
"step": 480
|
| 264 |
},
|
| 265 |
{
|
|
@@ -269,28 +269,28 @@
|
|
| 269 |
},
|
| 270 |
{
|
| 271 |
"epoch": 9.0,
|
| 272 |
-
"eval_loss": 3.
|
| 273 |
-
"eval_runtime": 4.
|
| 274 |
-
"eval_samples_per_second": 4.
|
| 275 |
-
"eval_steps_per_second": 1.
|
| 276 |
"step": 495
|
| 277 |
},
|
| 278 |
{
|
| 279 |
"epoch": 9.09,
|
| 280 |
"learning_rate": 3.3176699082935545e-05,
|
| 281 |
-
"loss": 0.
|
| 282 |
"step": 500
|
| 283 |
},
|
| 284 |
{
|
| 285 |
"epoch": 9.45,
|
| 286 |
"learning_rate": 3.166184534225087e-05,
|
| 287 |
-
"loss": 0.
|
| 288 |
"step": 520
|
| 289 |
},
|
| 290 |
{
|
| 291 |
"epoch": 9.82,
|
| 292 |
"learning_rate": 3.012016670162977e-05,
|
| 293 |
-
"loss": 0.
|
| 294 |
"step": 540
|
| 295 |
},
|
| 296 |
{
|
|
@@ -300,28 +300,28 @@
|
|
| 300 |
},
|
| 301 |
{
|
| 302 |
"epoch": 10.0,
|
| 303 |
-
"eval_loss": 3.
|
| 304 |
-
"eval_runtime": 4.
|
| 305 |
-
"eval_samples_per_second": 4.
|
| 306 |
-
"eval_steps_per_second": 1.
|
| 307 |
"step": 550
|
| 308 |
},
|
| 309 |
{
|
| 310 |
"epoch": 10.18,
|
| 311 |
"learning_rate": 2.8557870956832132e-05,
|
| 312 |
-
"loss": 0.
|
| 313 |
"step": 560
|
| 314 |
},
|
| 315 |
{
|
| 316 |
"epoch": 10.55,
|
| 317 |
"learning_rate": 2.698124892141971e-05,
|
| 318 |
-
"loss": 0.
|
| 319 |
"step": 580
|
| 320 |
},
|
| 321 |
{
|
| 322 |
"epoch": 10.91,
|
| 323 |
"learning_rate": 2.5396649095870202e-05,
|
| 324 |
-
"loss": 0.
|
| 325 |
"step": 600
|
| 326 |
},
|
| 327 |
{
|
|
@@ -331,28 +331,28 @@
|
|
| 331 |
},
|
| 332 |
{
|
| 333 |
"epoch": 11.0,
|
| 334 |
-
"eval_loss": 3.
|
| 335 |
-
"eval_runtime": 4.
|
| 336 |
-
"eval_samples_per_second": 4.
|
| 337 |
-
"eval_steps_per_second": 1.
|
| 338 |
"step": 605
|
| 339 |
},
|
| 340 |
{
|
| 341 |
"epoch": 11.27,
|
| 342 |
"learning_rate": 2.3810452104406444e-05,
|
| 343 |
-
"loss": 0.
|
| 344 |
"step": 620
|
| 345 |
},
|
| 346 |
{
|
| 347 |
"epoch": 11.64,
|
| 348 |
"learning_rate": 2.222904500247473e-05,
|
| 349 |
-
"loss": 0.
|
| 350 |
"step": 640
|
| 351 |
},
|
| 352 |
{
|
| 353 |
"epoch": 12.0,
|
| 354 |
"learning_rate": 2.0658795558326743e-05,
|
| 355 |
-
"loss": 0.
|
| 356 |
"step": 660
|
| 357 |
},
|
| 358 |
{
|
|
@@ -362,22 +362,22 @@
|
|
| 362 |
},
|
| 363 |
{
|
| 364 |
"epoch": 12.0,
|
| 365 |
-
"eval_loss": 3.
|
| 366 |
-
"eval_runtime": 4.
|
| 367 |
-
"eval_samples_per_second": 4.
|
| 368 |
-
"eval_steps_per_second": 1.
|
| 369 |
"step": 660
|
| 370 |
},
|
| 371 |
{
|
| 372 |
"epoch": 12.36,
|
| 373 |
"learning_rate": 1.9106026612264316e-05,
|
| 374 |
-
"loss": 0.
|
| 375 |
"step": 680
|
| 376 |
},
|
| 377 |
{
|
| 378 |
"epoch": 12.73,
|
| 379 |
"learning_rate": 1.7576990616793137e-05,
|
| 380 |
-
"loss": 0.
|
| 381 |
"step": 700
|
| 382 |
},
|
| 383 |
{
|
|
@@ -387,28 +387,28 @@
|
|
| 387 |
},
|
| 388 |
{
|
| 389 |
"epoch": 13.0,
|
| 390 |
-
"eval_loss": 3.
|
| 391 |
-
"eval_runtime": 4.
|
| 392 |
-
"eval_samples_per_second": 4.
|
| 393 |
-
"eval_steps_per_second": 1.
|
| 394 |
"step": 715
|
| 395 |
},
|
| 396 |
{
|
| 397 |
"epoch": 13.09,
|
| 398 |
"learning_rate": 1.6077844460203206e-05,
|
| 399 |
-
"loss": 0.
|
| 400 |
"step": 720
|
| 401 |
},
|
| 402 |
{
|
| 403 |
"epoch": 13.45,
|
| 404 |
"learning_rate": 1.4614624674952842e-05,
|
| 405 |
-
"loss": 0.
|
| 406 |
"step": 740
|
| 407 |
},
|
| 408 |
{
|
| 409 |
"epoch": 13.82,
|
| 410 |
"learning_rate": 1.3193223130682936e-05,
|
| 411 |
-
"loss": 0.
|
| 412 |
"step": 760
|
| 413 |
},
|
| 414 |
{
|
|
@@ -418,28 +418,28 @@
|
|
| 418 |
},
|
| 419 |
{
|
| 420 |
"epoch": 14.0,
|
| 421 |
-
"eval_loss":
|
| 422 |
-
"eval_runtime": 4.
|
| 423 |
-
"eval_samples_per_second": 4.
|
| 424 |
-
"eval_steps_per_second": 1.
|
| 425 |
"step": 770
|
| 426 |
},
|
| 427 |
{
|
| 428 |
"epoch": 14.18,
|
| 429 |
"learning_rate": 1.181936330973744e-05,
|
| 430 |
-
"loss": 0.
|
| 431 |
"step": 780
|
| 432 |
},
|
| 433 |
{
|
| 434 |
"epoch": 14.55,
|
| 435 |
"learning_rate": 1.049857726072005e-05,
|
| 436 |
-
"loss": 0.
|
| 437 |
"step": 800
|
| 438 |
},
|
| 439 |
{
|
| 440 |
"epoch": 14.91,
|
| 441 |
"learning_rate": 9.236183322886945e-06,
|
| 442 |
-
"loss": 0.
|
| 443 |
"step": 820
|
| 444 |
},
|
| 445 |
{
|
|
@@ -449,28 +449,28 @@
|
|
| 449 |
},
|
| 450 |
{
|
| 451 |
"epoch": 15.0,
|
| 452 |
-
"eval_loss": 4.
|
| 453 |
-
"eval_runtime": 4.
|
| 454 |
-
"eval_samples_per_second": 4.
|
| 455 |
-
"eval_steps_per_second": 1.
|
| 456 |
"step": 825
|
| 457 |
},
|
| 458 |
{
|
| 459 |
"epoch": 15.27,
|
| 460 |
"learning_rate": 8.0372647110717e-06,
|
| 461 |
-
"loss": 0.
|
| 462 |
"step": 840
|
| 463 |
},
|
| 464 |
{
|
| 465 |
"epoch": 15.64,
|
| 466 |
"learning_rate": 6.906649047373246e-06,
|
| 467 |
-
"loss": 0.
|
| 468 |
"step": 860
|
| 469 |
},
|
| 470 |
{
|
| 471 |
"epoch": 16.0,
|
| 472 |
"learning_rate": 5.848888922025553e-06,
|
| 473 |
-
"loss": 0.
|
| 474 |
"step": 880
|
| 475 |
},
|
| 476 |
{
|
|
@@ -480,22 +480,22 @@
|
|
| 480 |
},
|
| 481 |
{
|
| 482 |
"epoch": 16.0,
|
| 483 |
-
"eval_loss": 4.
|
| 484 |
-
"eval_runtime": 4.
|
| 485 |
-
"eval_samples_per_second": 4.
|
| 486 |
-
"eval_steps_per_second": 1.
|
| 487 |
"step": 880
|
| 488 |
},
|
| 489 |
{
|
| 490 |
"epoch": 16.36,
|
| 491 |
"learning_rate": 4.868243561723535e-06,
|
| 492 |
-
"loss": 0.
|
| 493 |
"step": 900
|
| 494 |
},
|
| 495 |
{
|
| 496 |
"epoch": 16.73,
|
| 497 |
"learning_rate": 3.968661679220468e-06,
|
| 498 |
-
"loss": 0.
|
| 499 |
"step": 920
|
| 500 |
},
|
| 501 |
{
|
|
@@ -505,28 +505,28 @@
|
|
| 505 |
},
|
| 506 |
{
|
| 507 |
"epoch": 17.0,
|
| 508 |
-
"eval_loss": 4.
|
| 509 |
-
"eval_runtime": 4.
|
| 510 |
-
"eval_samples_per_second": 4.
|
| 511 |
-
"eval_steps_per_second": 1.
|
| 512 |
"step": 935
|
| 513 |
},
|
| 514 |
{
|
| 515 |
"epoch": 17.09,
|
| 516 |
"learning_rate": 3.1537655732553768e-06,
|
| 517 |
-
"loss": 0.
|
| 518 |
"step": 940
|
| 519 |
},
|
| 520 |
{
|
| 521 |
"epoch": 17.45,
|
| 522 |
"learning_rate": 2.4268365428344736e-06,
|
| 523 |
-
"loss": 0.
|
| 524 |
"step": 960
|
| 525 |
},
|
| 526 |
{
|
| 527 |
"epoch": 17.82,
|
| 528 |
"learning_rate": 1.790801674598186e-06,
|
| 529 |
-
"loss": 0.
|
| 530 |
"step": 980
|
| 531 |
},
|
| 532 |
{
|
|
@@ -536,28 +536,28 @@
|
|
| 536 |
},
|
| 537 |
{
|
| 538 |
"epoch": 18.0,
|
| 539 |
-
"eval_loss": 4.
|
| 540 |
-
"eval_runtime": 4.
|
| 541 |
-
"eval_samples_per_second": 4.
|
| 542 |
-
"eval_steps_per_second": 1.
|
| 543 |
"step": 990
|
| 544 |
},
|
| 545 |
{
|
| 546 |
"epoch": 18.18,
|
| 547 |
"learning_rate": 1.248222056476367e-06,
|
| 548 |
-
"loss": 0.
|
| 549 |
"step": 1000
|
| 550 |
},
|
| 551 |
{
|
| 552 |
"epoch": 18.55,
|
| 553 |
"learning_rate": 8.012824650910938e-07,
|
| 554 |
-
"loss": 0.
|
| 555 |
"step": 1020
|
| 556 |
},
|
| 557 |
{
|
| 558 |
"epoch": 18.91,
|
| 559 |
"learning_rate": 4.517825684323324e-07,
|
| 560 |
-
"loss": 0.
|
| 561 |
"step": 1040
|
| 562 |
},
|
| 563 |
{
|
|
@@ -567,28 +567,28 @@
|
|
| 567 |
},
|
| 568 |
{
|
| 569 |
"epoch": 19.0,
|
| 570 |
-
"eval_loss": 4.
|
| 571 |
-
"eval_runtime": 4.
|
| 572 |
-
"eval_samples_per_second": 4.
|
| 573 |
-
"eval_steps_per_second": 1.
|
| 574 |
"step": 1045
|
| 575 |
},
|
| 576 |
{
|
| 577 |
"epoch": 19.27,
|
| 578 |
"learning_rate": 2.011296792301165e-07,
|
| 579 |
-
"loss": 0.
|
| 580 |
"step": 1060
|
| 581 |
},
|
| 582 |
{
|
| 583 |
"epoch": 19.64,
|
| 584 |
"learning_rate": 5.033308820289184e-08,
|
| 585 |
-
"loss": 0.
|
| 586 |
"step": 1080
|
| 587 |
},
|
| 588 |
{
|
| 589 |
"epoch": 20.0,
|
| 590 |
"learning_rate": 0.0,
|
| 591 |
-
"loss": 0.
|
| 592 |
"step": 1100
|
| 593 |
},
|
| 594 |
{
|
|
@@ -598,20 +598,20 @@
|
|
| 598 |
},
|
| 599 |
{
|
| 600 |
"epoch": 20.0,
|
| 601 |
-
"eval_loss": 4.
|
| 602 |
-
"eval_runtime": 4.
|
| 603 |
-
"eval_samples_per_second": 4.
|
| 604 |
-
"eval_steps_per_second": 1.
|
| 605 |
"step": 1100
|
| 606 |
},
|
| 607 |
{
|
| 608 |
"epoch": 20.0,
|
| 609 |
"step": 1100,
|
| 610 |
-
"total_flos": 3.
|
| 611 |
-
"train_loss": 0.
|
| 612 |
-
"train_runtime":
|
| 613 |
-
"train_samples_per_second": 1.
|
| 614 |
-
"train_steps_per_second": 0.
|
| 615 |
}
|
| 616 |
],
|
| 617 |
"logging_steps": 20,
|
|
@@ -619,7 +619,7 @@
|
|
| 619 |
"num_input_tokens_seen": 0,
|
| 620 |
"num_train_epochs": 20,
|
| 621 |
"save_steps": 20,
|
| 622 |
-
"total_flos": 3.
|
| 623 |
"train_batch_size": 4,
|
| 624 |
"trial_name": null,
|
| 625 |
"trial_params": null
|
|
|
|
| 11 |
{
|
| 12 |
"epoch": 0.02,
|
| 13 |
"learning_rate": 4.545454545454545e-07,
|
| 14 |
+
"loss": 2.7297,
|
| 15 |
"step": 1
|
| 16 |
},
|
| 17 |
{
|
| 18 |
"epoch": 0.36,
|
| 19 |
"learning_rate": 9.090909090909091e-06,
|
| 20 |
+
"loss": 2.7244,
|
| 21 |
"step": 20
|
| 22 |
},
|
| 23 |
{
|
| 24 |
"epoch": 0.73,
|
| 25 |
"learning_rate": 1.8181818181818182e-05,
|
| 26 |
+
"loss": 2.7672,
|
| 27 |
"step": 40
|
| 28 |
},
|
| 29 |
{
|
|
|
|
| 33 |
},
|
| 34 |
{
|
| 35 |
"epoch": 1.0,
|
| 36 |
+
"eval_loss": 2.312263250350952,
|
| 37 |
+
"eval_runtime": 4.9535,
|
| 38 |
+
"eval_samples_per_second": 4.643,
|
| 39 |
+
"eval_steps_per_second": 1.211,
|
| 40 |
"step": 55
|
| 41 |
},
|
| 42 |
{
|
| 43 |
"epoch": 1.09,
|
| 44 |
"learning_rate": 2.7272727272727273e-05,
|
| 45 |
+
"loss": 2.4172,
|
| 46 |
"step": 60
|
| 47 |
},
|
| 48 |
{
|
| 49 |
"epoch": 1.45,
|
| 50 |
"learning_rate": 3.6363636363636364e-05,
|
| 51 |
+
"loss": 2.1557,
|
| 52 |
"step": 80
|
| 53 |
},
|
| 54 |
{
|
| 55 |
"epoch": 1.82,
|
| 56 |
"learning_rate": 4.545454545454546e-05,
|
| 57 |
+
"loss": 2.0304,
|
| 58 |
"step": 100
|
| 59 |
},
|
| 60 |
{
|
|
|
|
| 64 |
},
|
| 65 |
{
|
| 66 |
"epoch": 2.0,
|
| 67 |
+
"eval_loss": 1.96739661693573,
|
| 68 |
+
"eval_runtime": 4.9491,
|
| 69 |
+
"eval_samples_per_second": 4.647,
|
| 70 |
+
"eval_steps_per_second": 1.212,
|
| 71 |
"step": 110
|
| 72 |
},
|
| 73 |
{
|
| 74 |
"epoch": 2.18,
|
| 75 |
"learning_rate": 4.9987413559579636e-05,
|
| 76 |
+
"loss": 1.8323,
|
| 77 |
"step": 120
|
| 78 |
},
|
| 79 |
{
|
| 80 |
"epoch": 2.55,
|
| 81 |
"learning_rate": 4.988679806432712e-05,
|
| 82 |
+
"loss": 1.8472,
|
| 83 |
"step": 140
|
| 84 |
},
|
| 85 |
{
|
| 86 |
"epoch": 2.91,
|
| 87 |
"learning_rate": 4.968597221690986e-05,
|
| 88 |
+
"loss": 1.7981,
|
| 89 |
"step": 160
|
| 90 |
},
|
| 91 |
{
|
|
|
|
| 95 |
},
|
| 96 |
{
|
| 97 |
"epoch": 3.0,
|
| 98 |
+
"eval_loss": 1.942185401916504,
|
| 99 |
+
"eval_runtime": 4.9566,
|
| 100 |
+
"eval_samples_per_second": 4.64,
|
| 101 |
+
"eval_steps_per_second": 1.211,
|
| 102 |
"step": 165
|
| 103 |
},
|
| 104 |
{
|
|
|
|
| 110 |
{
|
| 111 |
"epoch": 3.64,
|
| 112 |
"learning_rate": 4.898732434036244e-05,
|
| 113 |
+
"loss": 1.6171,
|
| 114 |
"step": 200
|
| 115 |
},
|
| 116 |
{
|
| 117 |
"epoch": 4.0,
|
| 118 |
"learning_rate": 4.849231551964771e-05,
|
| 119 |
+
"loss": 1.5825,
|
| 120 |
"step": 220
|
| 121 |
},
|
| 122 |
{
|
|
|
|
| 126 |
},
|
| 127 |
{
|
| 128 |
"epoch": 4.0,
|
| 129 |
+
"eval_loss": 2.0124683380126953,
|
| 130 |
+
"eval_runtime": 4.9519,
|
| 131 |
+
"eval_samples_per_second": 4.645,
|
| 132 |
+
"eval_steps_per_second": 1.212,
|
| 133 |
"step": 220
|
| 134 |
},
|
| 135 |
{
|
| 136 |
"epoch": 4.36,
|
| 137 |
"learning_rate": 4.790271143580174e-05,
|
| 138 |
+
"loss": 1.2929,
|
| 139 |
"step": 240
|
| 140 |
},
|
| 141 |
{
|
| 142 |
"epoch": 4.73,
|
| 143 |
"learning_rate": 4.722088621637309e-05,
|
| 144 |
+
"loss": 1.2811,
|
| 145 |
"step": 260
|
| 146 |
},
|
| 147 |
{
|
|
|
|
| 151 |
},
|
| 152 |
{
|
| 153 |
"epoch": 5.0,
|
| 154 |
+
"eval_loss": 2.265650510787964,
|
| 155 |
+
"eval_runtime": 4.96,
|
| 156 |
+
"eval_samples_per_second": 4.637,
|
| 157 |
+
"eval_steps_per_second": 1.21,
|
| 158 |
"step": 275
|
| 159 |
},
|
| 160 |
{
|
| 161 |
"epoch": 5.09,
|
| 162 |
"learning_rate": 4.644958533087443e-05,
|
| 163 |
+
"loss": 1.1431,
|
| 164 |
"step": 280
|
| 165 |
},
|
| 166 |
{
|
| 167 |
"epoch": 5.45,
|
| 168 |
"learning_rate": 4.559191453574582e-05,
|
| 169 |
+
"loss": 0.9042,
|
| 170 |
"step": 300
|
| 171 |
},
|
| 172 |
{
|
| 173 |
"epoch": 5.82,
|
| 174 |
"learning_rate": 4.465132736856969e-05,
|
| 175 |
+
"loss": 0.928,
|
| 176 |
"step": 320
|
| 177 |
},
|
| 178 |
{
|
|
|
|
| 182 |
},
|
| 183 |
{
|
| 184 |
"epoch": 6.0,
|
| 185 |
+
"eval_loss": 2.5270919799804688,
|
| 186 |
+
"eval_runtime": 4.9548,
|
| 187 |
+
"eval_samples_per_second": 4.642,
|
| 188 |
+
"eval_steps_per_second": 1.211,
|
| 189 |
"step": 330
|
| 190 |
},
|
| 191 |
{
|
| 192 |
"epoch": 6.18,
|
| 193 |
"learning_rate": 4.3631611241893874e-05,
|
| 194 |
+
"loss": 0.7725,
|
| 195 |
"step": 340
|
| 196 |
},
|
| 197 |
{
|
| 198 |
"epoch": 6.55,
|
| 199 |
"learning_rate": 4.2536872192658036e-05,
|
| 200 |
+
"loss": 0.6495,
|
| 201 |
"step": 360
|
| 202 |
},
|
| 203 |
{
|
| 204 |
"epoch": 6.91,
|
| 205 |
"learning_rate": 4.137151834863213e-05,
|
| 206 |
+
"loss": 0.6688,
|
| 207 |
"step": 380
|
| 208 |
},
|
| 209 |
{
|
|
|
|
| 213 |
},
|
| 214 |
{
|
| 215 |
"epoch": 7.0,
|
| 216 |
+
"eval_loss": 2.780010461807251,
|
| 217 |
+
"eval_runtime": 5.0002,
|
| 218 |
+
"eval_samples_per_second": 4.6,
|
| 219 |
+
"eval_steps_per_second": 1.2,
|
| 220 |
"step": 385
|
| 221 |
},
|
| 222 |
{
|
| 223 |
"epoch": 7.27,
|
| 224 |
"learning_rate": 4.014024217844167e-05,
|
| 225 |
+
"loss": 0.5264,
|
| 226 |
"step": 400
|
| 227 |
},
|
| 228 |
{
|
| 229 |
"epoch": 7.64,
|
| 230 |
"learning_rate": 3.884800159665276e-05,
|
| 231 |
+
"loss": 0.4558,
|
| 232 |
"step": 420
|
| 233 |
},
|
| 234 |
{
|
| 235 |
"epoch": 8.0,
|
| 236 |
"learning_rate": 3.7500000000000003e-05,
|
| 237 |
+
"loss": 0.4595,
|
| 238 |
"step": 440
|
| 239 |
},
|
| 240 |
{
|
|
|
|
| 244 |
},
|
| 245 |
{
|
| 246 |
"epoch": 8.0,
|
| 247 |
+
"eval_loss": 3.1072428226470947,
|
| 248 |
+
"eval_runtime": 4.9459,
|
| 249 |
+
"eval_samples_per_second": 4.65,
|
| 250 |
+
"eval_steps_per_second": 1.213,
|
| 251 |
"step": 440
|
| 252 |
},
|
| 253 |
{
|
| 254 |
"epoch": 8.36,
|
| 255 |
"learning_rate": 3.610166531514436e-05,
|
| 256 |
+
"loss": 0.3119,
|
| 257 |
"step": 460
|
| 258 |
},
|
| 259 |
{
|
| 260 |
"epoch": 8.73,
|
| 261 |
"learning_rate": 3.465862814232822e-05,
|
| 262 |
+
"loss": 0.3117,
|
| 263 |
"step": 480
|
| 264 |
},
|
| 265 |
{
|
|
|
|
| 269 |
},
|
| 270 |
{
|
| 271 |
"epoch": 9.0,
|
| 272 |
+
"eval_loss": 3.326101541519165,
|
| 273 |
+
"eval_runtime": 4.9468,
|
| 274 |
+
"eval_samples_per_second": 4.649,
|
| 275 |
+
"eval_steps_per_second": 1.213,
|
| 276 |
"step": 495
|
| 277 |
},
|
| 278 |
{
|
| 279 |
"epoch": 9.09,
|
| 280 |
"learning_rate": 3.3176699082935545e-05,
|
| 281 |
+
"loss": 0.3084,
|
| 282 |
"step": 500
|
| 283 |
},
|
| 284 |
{
|
| 285 |
"epoch": 9.45,
|
| 286 |
"learning_rate": 3.166184534225087e-05,
|
| 287 |
+
"loss": 0.2188,
|
| 288 |
"step": 520
|
| 289 |
},
|
| 290 |
{
|
| 291 |
"epoch": 9.82,
|
| 292 |
"learning_rate": 3.012016670162977e-05,
|
| 293 |
+
"loss": 0.242,
|
| 294 |
"step": 540
|
| 295 |
},
|
| 296 |
{
|
|
|
|
| 300 |
},
|
| 301 |
{
|
| 302 |
"epoch": 10.0,
|
| 303 |
+
"eval_loss": 3.5036461353302,
|
| 304 |
+
"eval_runtime": 4.9552,
|
| 305 |
+
"eval_samples_per_second": 4.642,
|
| 306 |
+
"eval_steps_per_second": 1.211,
|
| 307 |
"step": 550
|
| 308 |
},
|
| 309 |
{
|
| 310 |
"epoch": 10.18,
|
| 311 |
"learning_rate": 2.8557870956832132e-05,
|
| 312 |
+
"loss": 0.198,
|
| 313 |
"step": 560
|
| 314 |
},
|
| 315 |
{
|
| 316 |
"epoch": 10.55,
|
| 317 |
"learning_rate": 2.698124892141971e-05,
|
| 318 |
+
"loss": 0.1773,
|
| 319 |
"step": 580
|
| 320 |
},
|
| 321 |
{
|
| 322 |
"epoch": 10.91,
|
| 323 |
"learning_rate": 2.5396649095870202e-05,
|
| 324 |
+
"loss": 0.1718,
|
| 325 |
"step": 600
|
| 326 |
},
|
| 327 |
{
|
|
|
|
| 331 |
},
|
| 332 |
{
|
| 333 |
"epoch": 11.0,
|
| 334 |
+
"eval_loss": 3.6741881370544434,
|
| 335 |
+
"eval_runtime": 4.9476,
|
| 336 |
+
"eval_samples_per_second": 4.649,
|
| 337 |
+
"eval_steps_per_second": 1.213,
|
| 338 |
"step": 605
|
| 339 |
},
|
| 340 |
{
|
| 341 |
"epoch": 11.27,
|
| 342 |
"learning_rate": 2.3810452104406444e-05,
|
| 343 |
+
"loss": 0.14,
|
| 344 |
"step": 620
|
| 345 |
},
|
| 346 |
{
|
| 347 |
"epoch": 11.64,
|
| 348 |
"learning_rate": 2.222904500247473e-05,
|
| 349 |
+
"loss": 0.1337,
|
| 350 |
"step": 640
|
| 351 |
},
|
| 352 |
{
|
| 353 |
"epoch": 12.0,
|
| 354 |
"learning_rate": 2.0658795558326743e-05,
|
| 355 |
+
"loss": 0.157,
|
| 356 |
"step": 660
|
| 357 |
},
|
| 358 |
{
|
|
|
|
| 362 |
},
|
| 363 |
{
|
| 364 |
"epoch": 12.0,
|
| 365 |
+
"eval_loss": 3.789468288421631,
|
| 366 |
+
"eval_runtime": 4.9943,
|
| 367 |
+
"eval_samples_per_second": 4.605,
|
| 368 |
+
"eval_steps_per_second": 1.201,
|
| 369 |
"step": 660
|
| 370 |
},
|
| 371 |
{
|
| 372 |
"epoch": 12.36,
|
| 373 |
"learning_rate": 1.9106026612264316e-05,
|
| 374 |
+
"loss": 0.1428,
|
| 375 |
"step": 680
|
| 376 |
},
|
| 377 |
{
|
| 378 |
"epoch": 12.73,
|
| 379 |
"learning_rate": 1.7576990616793137e-05,
|
| 380 |
+
"loss": 0.1188,
|
| 381 |
"step": 700
|
| 382 |
},
|
| 383 |
{
|
|
|
|
| 387 |
},
|
| 388 |
{
|
| 389 |
"epoch": 13.0,
|
| 390 |
+
"eval_loss": 3.9183552265167236,
|
| 391 |
+
"eval_runtime": 4.9519,
|
| 392 |
+
"eval_samples_per_second": 4.645,
|
| 393 |
+
"eval_steps_per_second": 1.212,
|
| 394 |
"step": 715
|
| 395 |
},
|
| 396 |
{
|
| 397 |
"epoch": 13.09,
|
| 398 |
"learning_rate": 1.6077844460203206e-05,
|
| 399 |
+
"loss": 0.1176,
|
| 400 |
"step": 720
|
| 401 |
},
|
| 402 |
{
|
| 403 |
"epoch": 13.45,
|
| 404 |
"learning_rate": 1.4614624674952842e-05,
|
| 405 |
+
"loss": 0.108,
|
| 406 |
"step": 740
|
| 407 |
},
|
| 408 |
{
|
| 409 |
"epoch": 13.82,
|
| 410 |
"learning_rate": 1.3193223130682936e-05,
|
| 411 |
+
"loss": 0.1296,
|
| 412 |
"step": 760
|
| 413 |
},
|
| 414 |
{
|
|
|
|
| 418 |
},
|
| 419 |
{
|
| 420 |
"epoch": 14.0,
|
| 421 |
+
"eval_loss": 3.996537208557129,
|
| 422 |
+
"eval_runtime": 4.9512,
|
| 423 |
+
"eval_samples_per_second": 4.645,
|
| 424 |
+
"eval_steps_per_second": 1.212,
|
| 425 |
"step": 770
|
| 426 |
},
|
| 427 |
{
|
| 428 |
"epoch": 14.18,
|
| 429 |
"learning_rate": 1.181936330973744e-05,
|
| 430 |
+
"loss": 0.1085,
|
| 431 |
"step": 780
|
| 432 |
},
|
| 433 |
{
|
| 434 |
"epoch": 14.55,
|
| 435 |
"learning_rate": 1.049857726072005e-05,
|
| 436 |
+
"loss": 0.1042,
|
| 437 |
"step": 800
|
| 438 |
},
|
| 439 |
{
|
| 440 |
"epoch": 14.91,
|
| 441 |
"learning_rate": 9.236183322886945e-06,
|
| 442 |
+
"loss": 0.1219,
|
| 443 |
"step": 820
|
| 444 |
},
|
| 445 |
{
|
|
|
|
| 449 |
},
|
| 450 |
{
|
| 451 |
"epoch": 15.0,
|
| 452 |
+
"eval_loss": 4.114721775054932,
|
| 453 |
+
"eval_runtime": 4.9671,
|
| 454 |
+
"eval_samples_per_second": 4.63,
|
| 455 |
+
"eval_steps_per_second": 1.208,
|
| 456 |
"step": 825
|
| 457 |
},
|
| 458 |
{
|
| 459 |
"epoch": 15.27,
|
| 460 |
"learning_rate": 8.0372647110717e-06,
|
| 461 |
+
"loss": 0.101,
|
| 462 |
"step": 840
|
| 463 |
},
|
| 464 |
{
|
| 465 |
"epoch": 15.64,
|
| 466 |
"learning_rate": 6.906649047373246e-06,
|
| 467 |
+
"loss": 0.1184,
|
| 468 |
"step": 860
|
| 469 |
},
|
| 470 |
{
|
| 471 |
"epoch": 16.0,
|
| 472 |
"learning_rate": 5.848888922025553e-06,
|
| 473 |
+
"loss": 0.1,
|
| 474 |
"step": 880
|
| 475 |
},
|
| 476 |
{
|
|
|
|
| 480 |
},
|
| 481 |
{
|
| 482 |
"epoch": 16.0,
|
| 483 |
+
"eval_loss": 4.137594699859619,
|
| 484 |
+
"eval_runtime": 4.9557,
|
| 485 |
+
"eval_samples_per_second": 4.641,
|
| 486 |
+
"eval_steps_per_second": 1.211,
|
| 487 |
"step": 880
|
| 488 |
},
|
| 489 |
{
|
| 490 |
"epoch": 16.36,
|
| 491 |
"learning_rate": 4.868243561723535e-06,
|
| 492 |
+
"loss": 0.0975,
|
| 493 |
"step": 900
|
| 494 |
},
|
| 495 |
{
|
| 496 |
"epoch": 16.73,
|
| 497 |
"learning_rate": 3.968661679220468e-06,
|
| 498 |
+
"loss": 0.1155,
|
| 499 |
"step": 920
|
| 500 |
},
|
| 501 |
{
|
|
|
|
| 505 |
},
|
| 506 |
{
|
| 507 |
"epoch": 17.0,
|
| 508 |
+
"eval_loss": 4.1758294105529785,
|
| 509 |
+
"eval_runtime": 4.9475,
|
| 510 |
+
"eval_samples_per_second": 4.649,
|
| 511 |
+
"eval_steps_per_second": 1.213,
|
| 512 |
"step": 935
|
| 513 |
},
|
| 514 |
{
|
| 515 |
"epoch": 17.09,
|
| 516 |
"learning_rate": 3.1537655732553768e-06,
|
| 517 |
+
"loss": 0.1017,
|
| 518 |
"step": 940
|
| 519 |
},
|
| 520 |
{
|
| 521 |
"epoch": 17.45,
|
| 522 |
"learning_rate": 2.4268365428344736e-06,
|
| 523 |
+
"loss": 0.0972,
|
| 524 |
"step": 960
|
| 525 |
},
|
| 526 |
{
|
| 527 |
"epoch": 17.82,
|
| 528 |
"learning_rate": 1.790801674598186e-06,
|
| 529 |
+
"loss": 0.1083,
|
| 530 |
"step": 980
|
| 531 |
},
|
| 532 |
{
|
|
|
|
| 536 |
},
|
| 537 |
{
|
| 538 |
"epoch": 18.0,
|
| 539 |
+
"eval_loss": 4.201599597930908,
|
| 540 |
+
"eval_runtime": 4.963,
|
| 541 |
+
"eval_samples_per_second": 4.634,
|
| 542 |
+
"eval_steps_per_second": 1.209,
|
| 543 |
"step": 990
|
| 544 |
},
|
| 545 |
{
|
| 546 |
"epoch": 18.18,
|
| 547 |
"learning_rate": 1.248222056476367e-06,
|
| 548 |
+
"loss": 0.1031,
|
| 549 |
"step": 1000
|
| 550 |
},
|
| 551 |
{
|
| 552 |
"epoch": 18.55,
|
| 553 |
"learning_rate": 8.012824650910938e-07,
|
| 554 |
+
"loss": 0.109,
|
| 555 |
"step": 1020
|
| 556 |
},
|
| 557 |
{
|
| 558 |
"epoch": 18.91,
|
| 559 |
"learning_rate": 4.517825684323324e-07,
|
| 560 |
+
"loss": 0.0961,
|
| 561 |
"step": 1040
|
| 562 |
},
|
| 563 |
{
|
|
|
|
| 567 |
},
|
| 568 |
{
|
| 569 |
"epoch": 19.0,
|
| 570 |
+
"eval_loss": 4.204728603363037,
|
| 571 |
+
"eval_runtime": 4.956,
|
| 572 |
+
"eval_samples_per_second": 4.641,
|
| 573 |
+
"eval_steps_per_second": 1.211,
|
| 574 |
"step": 1045
|
| 575 |
},
|
| 576 |
{
|
| 577 |
"epoch": 19.27,
|
| 578 |
"learning_rate": 2.011296792301165e-07,
|
| 579 |
+
"loss": 0.0981,
|
| 580 |
"step": 1060
|
| 581 |
},
|
| 582 |
{
|
| 583 |
"epoch": 19.64,
|
| 584 |
"learning_rate": 5.033308820289184e-08,
|
| 585 |
+
"loss": 0.1161,
|
| 586 |
"step": 1080
|
| 587 |
},
|
| 588 |
{
|
| 589 |
"epoch": 20.0,
|
| 590 |
"learning_rate": 0.0,
|
| 591 |
+
"loss": 0.0917,
|
| 592 |
"step": 1100
|
| 593 |
},
|
| 594 |
{
|
|
|
|
| 598 |
},
|
| 599 |
{
|
| 600 |
"epoch": 20.0,
|
| 601 |
+
"eval_loss": 4.205177307128906,
|
| 602 |
+
"eval_runtime": 4.9637,
|
| 603 |
+
"eval_samples_per_second": 4.634,
|
| 604 |
+
"eval_steps_per_second": 1.209,
|
| 605 |
"step": 1100
|
| 606 |
},
|
| 607 |
{
|
| 608 |
"epoch": 20.0,
|
| 609 |
"step": 1100,
|
| 610 |
+
"total_flos": 3.836907908090266e+16,
|
| 611 |
+
"train_loss": 0.6597742699493061,
|
| 612 |
+
"train_runtime": 3169.8576,
|
| 613 |
+
"train_samples_per_second": 1.369,
|
| 614 |
+
"train_steps_per_second": 0.347
|
| 615 |
}
|
| 616 |
],
|
| 617 |
"logging_steps": 20,
|
|
|
|
| 619 |
"num_input_tokens_seen": 0,
|
| 620 |
"num_train_epochs": 20,
|
| 621 |
"save_steps": 20,
|
| 622 |
+
"total_flos": 3.836907908090266e+16,
|
| 623 |
"train_batch_size": 4,
|
| 624 |
"trial_name": null,
|
| 625 |
"trial_params": null
|