Model save
Browse files- README.md +10 -12
- adapter_config.json +5 -5
- adapter_model.safetensors +1 -1
- all_results.json +8 -8
- eval_results.json +4 -4
- train_results.json +4 -4
- trainer_state.json +216 -212
- 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-5e-5-50
|
|
@@ -20,9 +16,9 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 20 |
|
| 21 |
# Yi-6B-ruozhiba-5e-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: 3.
|
| 26 |
|
| 27 |
## Model description
|
| 28 |
|
|
@@ -54,12 +50,14 @@ The following hyperparameters were used during training:
|
|
| 54 |
|
| 55 |
| Training Loss | Epoch | Step | Validation Loss |
|
| 56 |
|:-------------:|:-----:|:----:|:---------------:|
|
| 57 |
-
| 1.
|
| 58 |
-
| 1.
|
| 59 |
-
| 1.
|
| 60 |
-
|
|
| 61 |
-
| 0.
|
| 62 |
-
| 0.
|
|
|
|
|
|
|
| 63 |
|
| 64 |
|
| 65 |
### 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-5e-5-50
|
|
|
|
| 16 |
|
| 17 |
# Yi-6B-ruozhiba-5e-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: 3.0726
|
| 22 |
|
| 23 |
## Model description
|
| 24 |
|
|
|
|
| 50 |
|
| 51 |
| Training Loss | Epoch | Step | Validation Loss |
|
| 52 |
|:-------------:|:-----:|:----:|:---------------:|
|
| 53 |
+
| 1.8473 | 2.0 | 110 | 1.9843 |
|
| 54 |
+
| 1.7645 | 3.0 | 165 | 1.9680 |
|
| 55 |
+
| 1.4795 | 4.0 | 220 | 2.0760 |
|
| 56 |
+
| 1.2467 | 5.0 | 275 | 2.2715 |
|
| 57 |
+
| 0.7614 | 7.0 | 385 | 2.7865 |
|
| 58 |
+
| 0.6237 | 8.0 | 440 | 2.9589 |
|
| 59 |
+
| 0.5711 | 9.0 | 495 | 3.0581 |
|
| 60 |
+
| 0.5884 | 10.0 | 550 | 3.0726 |
|
| 61 |
|
| 62 |
|
| 63 |
### Framework versions
|
adapter_config.json
CHANGED
|
@@ -19,13 +19,13 @@
|
|
| 19 |
"rank_pattern": {},
|
| 20 |
"revision": null,
|
| 21 |
"target_modules": [
|
| 22 |
-
"
|
| 23 |
-
"
|
| 24 |
-
"gate_proj",
|
| 25 |
"o_proj",
|
|
|
|
| 26 |
"k_proj",
|
| 27 |
-
"
|
| 28 |
-
"
|
| 29 |
],
|
| 30 |
"task_type": "CAUSAL_LM"
|
| 31 |
}
|
|
|
|
| 19 |
"rank_pattern": {},
|
| 20 |
"revision": null,
|
| 21 |
"target_modules": [
|
| 22 |
+
"q_proj",
|
| 23 |
+
"down_proj",
|
|
|
|
| 24 |
"o_proj",
|
| 25 |
+
"v_proj",
|
| 26 |
"k_proj",
|
| 27 |
+
"gate_proj",
|
| 28 |
+
"up_proj"
|
| 29 |
],
|
| 30 |
"task_type": "CAUSAL_LM"
|
| 31 |
}
|
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:c9ce552cc5fb48a9312f62598453e6872694589dbd7643cd7f32eba154c51f85
|
| 3 |
size 145287696
|
all_results.json
CHANGED
|
@@ -1,13 +1,13 @@
|
|
| 1 |
{
|
| 2 |
"epoch": 10.0,
|
| 3 |
-
"eval_loss": 3.
|
| 4 |
-
"eval_runtime":
|
| 5 |
"eval_samples": 23,
|
| 6 |
-
"eval_samples_per_second":
|
| 7 |
-
"eval_steps_per_second":
|
| 8 |
-
"train_loss": 0.
|
| 9 |
-
"train_runtime":
|
| 10 |
"train_samples": 217,
|
| 11 |
-
"train_samples_per_second":
|
| 12 |
-
"train_steps_per_second":
|
| 13 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"epoch": 10.0,
|
| 3 |
+
"eval_loss": 3.072557210922241,
|
| 4 |
+
"eval_runtime": 4.9184,
|
| 5 |
"eval_samples": 23,
|
| 6 |
+
"eval_samples_per_second": 4.676,
|
| 7 |
+
"eval_steps_per_second": 1.22,
|
| 8 |
+
"train_loss": 0.2617128190127286,
|
| 9 |
+
"train_runtime": 10832.3445,
|
| 10 |
"train_samples": 217,
|
| 11 |
+
"train_samples_per_second": 0.2,
|
| 12 |
+
"train_steps_per_second": 0.051
|
| 13 |
}
|
eval_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
"epoch": 10.0,
|
| 3 |
-
"eval_loss": 3.
|
| 4 |
-
"eval_runtime":
|
| 5 |
"eval_samples": 23,
|
| 6 |
-
"eval_samples_per_second":
|
| 7 |
-
"eval_steps_per_second":
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"epoch": 10.0,
|
| 3 |
+
"eval_loss": 3.072557210922241,
|
| 4 |
+
"eval_runtime": 4.9184,
|
| 5 |
"eval_samples": 23,
|
| 6 |
+
"eval_samples_per_second": 4.676,
|
| 7 |
+
"eval_steps_per_second": 1.22
|
| 8 |
}
|
train_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
"epoch": 10.0,
|
| 3 |
-
"train_loss": 0.
|
| 4 |
-
"train_runtime":
|
| 5 |
"train_samples": 217,
|
| 6 |
-
"train_samples_per_second":
|
| 7 |
-
"train_steps_per_second":
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"epoch": 10.0,
|
| 3 |
+
"train_loss": 0.2617128190127286,
|
| 4 |
+
"train_runtime": 10832.3445,
|
| 5 |
"train_samples": 217,
|
| 6 |
+
"train_samples_per_second": 0.2,
|
| 7 |
+
"train_steps_per_second": 0.051
|
| 8 |
}
|
trainer_state.json
CHANGED
|
@@ -11,947 +11,951 @@
|
|
| 11 |
{
|
| 12 |
"epoch": 0.02,
|
| 13 |
"learning_rate": 9.09090909090909e-07,
|
| 14 |
-
"loss": 2.
|
| 15 |
"step": 1
|
| 16 |
},
|
| 17 |
{
|
| 18 |
"epoch": 0.07,
|
| 19 |
"learning_rate": 3.636363636363636e-06,
|
| 20 |
-
"loss": 2.
|
| 21 |
"step": 4
|
| 22 |
},
|
| 23 |
{
|
| 24 |
"epoch": 0.15,
|
| 25 |
"learning_rate": 7.272727272727272e-06,
|
| 26 |
-
"loss": 2.
|
| 27 |
"step": 8
|
| 28 |
},
|
| 29 |
{
|
| 30 |
"epoch": 0.22,
|
| 31 |
"learning_rate": 1.0909090909090909e-05,
|
| 32 |
-
"loss": 2.
|
| 33 |
"step": 12
|
| 34 |
},
|
| 35 |
{
|
| 36 |
"epoch": 0.29,
|
| 37 |
"learning_rate": 1.4545454545454545e-05,
|
| 38 |
-
"loss": 2.
|
| 39 |
"step": 16
|
| 40 |
},
|
| 41 |
{
|
| 42 |
"epoch": 0.36,
|
| 43 |
"learning_rate": 1.8181818181818182e-05,
|
| 44 |
-
"loss": 2.
|
| 45 |
"step": 20
|
| 46 |
},
|
| 47 |
{
|
| 48 |
"epoch": 0.44,
|
| 49 |
"learning_rate": 2.1818181818181818e-05,
|
| 50 |
-
"loss": 2.
|
| 51 |
"step": 24
|
| 52 |
},
|
| 53 |
{
|
| 54 |
"epoch": 0.51,
|
| 55 |
"learning_rate": 2.5454545454545454e-05,
|
| 56 |
-
"loss": 2.
|
| 57 |
"step": 28
|
| 58 |
},
|
| 59 |
{
|
| 60 |
"epoch": 0.58,
|
| 61 |
"learning_rate": 2.909090909090909e-05,
|
| 62 |
-
"loss": 2.
|
| 63 |
"step": 32
|
| 64 |
},
|
| 65 |
{
|
| 66 |
"epoch": 0.65,
|
| 67 |
"learning_rate": 3.272727272727273e-05,
|
| 68 |
-
"loss": 2.
|
| 69 |
"step": 36
|
| 70 |
},
|
| 71 |
{
|
| 72 |
"epoch": 0.73,
|
| 73 |
"learning_rate": 3.6363636363636364e-05,
|
| 74 |
-
"loss":
|
| 75 |
"step": 40
|
| 76 |
},
|
| 77 |
{
|
| 78 |
"epoch": 0.8,
|
| 79 |
"learning_rate": 4e-05,
|
| 80 |
-
"loss": 2.
|
| 81 |
"step": 44
|
| 82 |
},
|
| 83 |
{
|
| 84 |
"epoch": 0.87,
|
| 85 |
"learning_rate": 4.3636363636363636e-05,
|
| 86 |
-
"loss": 2.
|
| 87 |
"step": 48
|
| 88 |
},
|
| 89 |
{
|
| 90 |
"epoch": 0.95,
|
| 91 |
"learning_rate": 4.7272727272727275e-05,
|
| 92 |
-
"loss":
|
| 93 |
"step": 52
|
| 94 |
},
|
| 95 |
{
|
| 96 |
"epoch": 1.02,
|
| 97 |
"learning_rate": 4.999949650182266e-05,
|
| 98 |
-
"loss":
|
| 99 |
"step": 56
|
| 100 |
},
|
| 101 |
{
|
| 102 |
"epoch": 1.09,
|
| 103 |
"learning_rate": 4.9987413559579636e-05,
|
| 104 |
-
"loss":
|
| 105 |
"step": 60
|
| 106 |
},
|
| 107 |
{
|
| 108 |
"epoch": 1.16,
|
| 109 |
"learning_rate": 4.995922759815339e-05,
|
| 110 |
-
"loss":
|
| 111 |
"step": 64
|
| 112 |
},
|
| 113 |
{
|
| 114 |
"epoch": 1.24,
|
| 115 |
"learning_rate": 4.991495678185202e-05,
|
| 116 |
-
"loss": 1.
|
| 117 |
"step": 68
|
| 118 |
},
|
| 119 |
{
|
| 120 |
"epoch": 1.31,
|
| 121 |
"learning_rate": 4.985462964079137e-05,
|
| 122 |
-
"loss":
|
| 123 |
"step": 72
|
| 124 |
},
|
| 125 |
{
|
| 126 |
"epoch": 1.38,
|
| 127 |
"learning_rate": 4.977828505250903e-05,
|
| 128 |
-
"loss": 1.
|
| 129 |
"step": 76
|
| 130 |
},
|
| 131 |
{
|
| 132 |
"epoch": 1.45,
|
| 133 |
"learning_rate": 4.968597221690986e-05,
|
| 134 |
-
"loss": 1.
|
| 135 |
"step": 80
|
| 136 |
},
|
| 137 |
{
|
| 138 |
"epoch": 1.53,
|
| 139 |
"learning_rate": 4.957775062455933e-05,
|
| 140 |
-
"loss": 1.
|
| 141 |
"step": 84
|
| 142 |
},
|
| 143 |
{
|
| 144 |
"epoch": 1.6,
|
| 145 |
"learning_rate": 4.9453690018345144e-05,
|
| 146 |
-
"loss": 1.
|
| 147 |
"step": 88
|
| 148 |
},
|
| 149 |
{
|
| 150 |
"epoch": 1.67,
|
| 151 |
"learning_rate": 4.931387034853173e-05,
|
| 152 |
-
"loss": 1.
|
| 153 |
"step": 92
|
| 154 |
},
|
| 155 |
{
|
| 156 |
"epoch": 1.75,
|
| 157 |
"learning_rate": 4.915838172123671e-05,
|
| 158 |
-
"loss": 1.
|
| 159 |
"step": 96
|
| 160 |
},
|
| 161 |
{
|
| 162 |
"epoch": 1.82,
|
| 163 |
"learning_rate": 4.898732434036244e-05,
|
| 164 |
-
"loss": 1.
|
| 165 |
"step": 100
|
| 166 |
},
|
| 167 |
{
|
| 168 |
"epoch": 1.89,
|
| 169 |
"learning_rate": 4.880080844302004e-05,
|
| 170 |
-
"loss": 1.
|
| 171 |
"step": 104
|
| 172 |
},
|
| 173 |
{
|
| 174 |
"epoch": 1.96,
|
| 175 |
"learning_rate": 4.859895422848767e-05,
|
| 176 |
-
"loss": 1.
|
| 177 |
"step": 108
|
| 178 |
},
|
| 179 |
{
|
| 180 |
"epoch": 2.0,
|
| 181 |
-
"
|
|
|
|
| 182 |
"step": 110
|
| 183 |
},
|
| 184 |
{
|
| 185 |
"epoch": 2.0,
|
| 186 |
-
"
|
| 187 |
-
"
|
| 188 |
-
|
| 189 |
-
|
| 190 |
-
"epoch": 2.0,
|
| 191 |
-
"eval_loss": 1.8269920349121094,
|
| 192 |
-
"eval_runtime": 4.9635,
|
| 193 |
-
"eval_samples_per_second": 4.634,
|
| 194 |
-
"eval_steps_per_second": 1.209,
|
| 195 |
"step": 110
|
| 196 |
},
|
| 197 |
{
|
| 198 |
"epoch": 2.04,
|
| 199 |
"learning_rate": 4.838189178074867e-05,
|
| 200 |
-
"loss":
|
| 201 |
"step": 112
|
| 202 |
},
|
| 203 |
{
|
| 204 |
"epoch": 2.11,
|
| 205 |
"learning_rate": 4.8149760984659506e-05,
|
| 206 |
-
"loss": 1.
|
| 207 |
"step": 116
|
| 208 |
},
|
| 209 |
{
|
| 210 |
"epoch": 2.18,
|
| 211 |
"learning_rate": 4.790271143580174e-05,
|
| 212 |
-
"loss": 1.
|
| 213 |
"step": 120
|
| 214 |
},
|
| 215 |
{
|
| 216 |
"epoch": 2.25,
|
| 217 |
"learning_rate": 4.764090234407577e-05,
|
| 218 |
-
"loss": 1.
|
| 219 |
"step": 124
|
| 220 |
},
|
| 221 |
{
|
| 222 |
"epoch": 2.33,
|
| 223 |
"learning_rate": 4.7364502431098844e-05,
|
| 224 |
-
"loss": 1.
|
| 225 |
"step": 128
|
| 226 |
},
|
| 227 |
{
|
| 228 |
"epoch": 2.4,
|
| 229 |
"learning_rate": 4.707368982147318e-05,
|
| 230 |
-
"loss": 1.
|
| 231 |
"step": 132
|
| 232 |
},
|
| 233 |
{
|
| 234 |
"epoch": 2.47,
|
| 235 |
"learning_rate": 4.6768651927994434e-05,
|
| 236 |
-
"loss": 1.
|
| 237 |
"step": 136
|
| 238 |
},
|
| 239 |
{
|
| 240 |
"epoch": 2.55,
|
| 241 |
"learning_rate": 4.644958533087443e-05,
|
| 242 |
-
"loss": 1.
|
| 243 |
"step": 140
|
| 244 |
},
|
| 245 |
{
|
| 246 |
"epoch": 2.62,
|
| 247 |
"learning_rate": 4.611669565105596e-05,
|
| 248 |
-
"loss": 1.
|
| 249 |
"step": 144
|
| 250 |
},
|
| 251 |
{
|
| 252 |
"epoch": 2.69,
|
| 253 |
"learning_rate": 4.5770197417701365e-05,
|
| 254 |
-
"loss": 1.
|
| 255 |
"step": 148
|
| 256 |
},
|
| 257 |
{
|
| 258 |
"epoch": 2.76,
|
| 259 |
"learning_rate": 4.5410313929940244e-05,
|
| 260 |
-
"loss": 1.
|
| 261 |
"step": 152
|
| 262 |
},
|
| 263 |
{
|
| 264 |
"epoch": 2.84,
|
| 265 |
"learning_rate": 4.503727711296538e-05,
|
| 266 |
-
"loss": 1.
|
| 267 |
"step": 156
|
| 268 |
},
|
| 269 |
{
|
| 270 |
"epoch": 2.91,
|
| 271 |
"learning_rate": 4.465132736856969e-05,
|
| 272 |
-
"loss": 1.
|
| 273 |
"step": 160
|
| 274 |
},
|
| 275 |
{
|
| 276 |
"epoch": 2.98,
|
| 277 |
"learning_rate": 4.425271342022039e-05,
|
| 278 |
-
"loss": 1.
|
| 279 |
"step": 164
|
| 280 |
},
|
| 281 |
{
|
| 282 |
"epoch": 3.0,
|
| 283 |
-
"
|
|
|
|
| 284 |
"step": 165
|
| 285 |
},
|
| 286 |
{
|
| 287 |
"epoch": 3.0,
|
| 288 |
-
"
|
| 289 |
-
"
|
| 290 |
-
|
| 291 |
-
|
| 292 |
-
"epoch": 3.0,
|
| 293 |
-
"eval_loss": 1.8256371021270752,
|
| 294 |
-
"eval_runtime": 4.9731,
|
| 295 |
-
"eval_samples_per_second": 4.625,
|
| 296 |
-
"eval_steps_per_second": 1.206,
|
| 297 |
"step": 165
|
| 298 |
},
|
| 299 |
{
|
| 300 |
"epoch": 3.05,
|
| 301 |
"learning_rate": 4.384169215277041e-05,
|
| 302 |
-
"loss": 1.
|
| 303 |
"step": 168
|
| 304 |
},
|
| 305 |
{
|
| 306 |
"epoch": 3.13,
|
| 307 |
"learning_rate": 4.341852844691012e-05,
|
| 308 |
-
"loss": 1.
|
| 309 |
"step": 172
|
| 310 |
},
|
| 311 |
{
|
| 312 |
"epoch": 3.2,
|
| 313 |
"learning_rate": 4.2983495008466276e-05,
|
| 314 |
-
"loss": 1.
|
| 315 |
"step": 176
|
| 316 |
},
|
| 317 |
{
|
| 318 |
"epoch": 3.27,
|
| 319 |
"learning_rate": 4.2536872192658036e-05,
|
| 320 |
-
"loss": 1.
|
| 321 |
"step": 180
|
| 322 |
},
|
| 323 |
{
|
| 324 |
"epoch": 3.35,
|
| 325 |
"learning_rate": 4.2078947823423364e-05,
|
| 326 |
-
"loss": 1.
|
| 327 |
"step": 184
|
| 328 |
},
|
| 329 |
{
|
| 330 |
"epoch": 3.42,
|
| 331 |
"learning_rate": 4.161001700793231e-05,
|
| 332 |
-
"loss": 1.
|
| 333 |
"step": 188
|
| 334 |
},
|
| 335 |
{
|
| 336 |
"epoch": 3.49,
|
| 337 |
"learning_rate": 4.113038194640658e-05,
|
| 338 |
-
"loss": 1.
|
| 339 |
"step": 192
|
| 340 |
},
|
| 341 |
{
|
| 342 |
"epoch": 3.56,
|
| 343 |
"learning_rate": 4.064035173736804e-05,
|
| 344 |
-
"loss": 1.
|
| 345 |
"step": 196
|
| 346 |
},
|
| 347 |
{
|
| 348 |
"epoch": 3.64,
|
| 349 |
"learning_rate": 4.014024217844167e-05,
|
| 350 |
-
"loss": 1.
|
| 351 |
"step": 200
|
| 352 |
},
|
| 353 |
{
|
| 354 |
"epoch": 3.71,
|
| 355 |
"learning_rate": 3.9630375562841295e-05,
|
| 356 |
-
"loss": 1.
|
| 357 |
"step": 204
|
| 358 |
},
|
| 359 |
{
|
| 360 |
"epoch": 3.78,
|
| 361 |
"learning_rate": 3.911108047166924e-05,
|
| 362 |
-
"loss": 1.
|
| 363 |
"step": 208
|
| 364 |
},
|
| 365 |
{
|
| 366 |
"epoch": 3.85,
|
| 367 |
"learning_rate": 3.858269156216383e-05,
|
| 368 |
-
"loss": 1.
|
| 369 |
"step": 212
|
| 370 |
},
|
| 371 |
{
|
| 372 |
"epoch": 3.93,
|
| 373 |
"learning_rate": 3.804554935203115e-05,
|
| 374 |
-
"loss": 1.
|
| 375 |
"step": 216
|
| 376 |
},
|
| 377 |
{
|
| 378 |
"epoch": 4.0,
|
| 379 |
"learning_rate": 3.7500000000000003e-05,
|
| 380 |
-
"loss": 1.
|
| 381 |
-
"step": 220
|
| 382 |
-
},
|
| 383 |
-
{
|
| 384 |
-
"epoch": 4.0,
|
| 385 |
-
"gpt4_scores": 0.6719999999999999,
|
| 386 |
"step": 220
|
| 387 |
},
|
| 388 |
{
|
| 389 |
"epoch": 4.0,
|
| 390 |
-
"
|
|
|
|
| 391 |
"step": 220
|
| 392 |
},
|
| 393 |
{
|
| 394 |
"epoch": 4.0,
|
| 395 |
-
"eval_loss":
|
| 396 |
-
"eval_runtime": 4.
|
| 397 |
-
"eval_samples_per_second": 4.
|
| 398 |
-
"eval_steps_per_second": 1.
|
| 399 |
"step": 220
|
| 400 |
},
|
| 401 |
{
|
| 402 |
"epoch": 4.07,
|
| 403 |
"learning_rate": 3.694639508274158e-05,
|
| 404 |
-
"loss": 1.
|
| 405 |
"step": 224
|
| 406 |
},
|
| 407 |
{
|
| 408 |
"epoch": 4.15,
|
| 409 |
"learning_rate": 3.638509136829758e-05,
|
| 410 |
-
"loss": 1.
|
| 411 |
"step": 228
|
| 412 |
},
|
| 413 |
{
|
| 414 |
"epoch": 4.22,
|
| 415 |
"learning_rate": 3.581645058616271e-05,
|
| 416 |
-
"loss":
|
| 417 |
"step": 232
|
| 418 |
},
|
| 419 |
{
|
| 420 |
"epoch": 4.29,
|
| 421 |
"learning_rate": 3.5240839194169885e-05,
|
| 422 |
-
"loss": 1.
|
| 423 |
"step": 236
|
| 424 |
},
|
| 425 |
{
|
| 426 |
"epoch": 4.36,
|
| 427 |
"learning_rate": 3.465862814232822e-05,
|
| 428 |
-
"loss": 1.
|
| 429 |
"step": 240
|
| 430 |
},
|
| 431 |
{
|
| 432 |
"epoch": 4.44,
|
| 433 |
"learning_rate": 3.4070192633766025e-05,
|
| 434 |
-
"loss":
|
| 435 |
"step": 244
|
| 436 |
},
|
| 437 |
{
|
| 438 |
"epoch": 4.51,
|
| 439 |
"learning_rate": 3.3475911882933015e-05,
|
| 440 |
-
"loss": 1.
|
| 441 |
"step": 248
|
| 442 |
},
|
| 443 |
{
|
| 444 |
"epoch": 4.58,
|
| 445 |
"learning_rate": 3.2876168871217325e-05,
|
| 446 |
-
"loss": 1.
|
| 447 |
"step": 252
|
| 448 |
},
|
| 449 |
{
|
| 450 |
"epoch": 4.65,
|
| 451 |
"learning_rate": 3.2271350100134975e-05,
|
| 452 |
-
"loss": 1.
|
| 453 |
"step": 256
|
| 454 |
},
|
| 455 |
{
|
| 456 |
"epoch": 4.73,
|
| 457 |
"learning_rate": 3.166184534225087e-05,
|
| 458 |
-
"loss":
|
| 459 |
"step": 260
|
| 460 |
},
|
| 461 |
{
|
| 462 |
"epoch": 4.8,
|
| 463 |
"learning_rate": 3.104804738999169e-05,
|
| 464 |
-
"loss": 1.
|
| 465 |
"step": 264
|
| 466 |
},
|
| 467 |
{
|
| 468 |
"epoch": 4.87,
|
| 469 |
"learning_rate": 3.0430351802512698e-05,
|
| 470 |
-
"loss":
|
| 471 |
"step": 268
|
| 472 |
},
|
| 473 |
{
|
| 474 |
"epoch": 4.95,
|
| 475 |
"learning_rate": 2.9809156650781528e-05,
|
| 476 |
-
"loss":
|
| 477 |
"step": 272
|
| 478 |
},
|
| 479 |
{
|
| 480 |
"epoch": 5.0,
|
| 481 |
-
"
|
| 482 |
-
"
|
| 483 |
-
},
|
| 484 |
-
{
|
| 485 |
-
"epoch": 5.0,
|
| 486 |
-
"std": 0.05452045487704591,
|
| 487 |
"step": 275
|
| 488 |
},
|
| 489 |
{
|
| 490 |
"epoch": 5.0,
|
| 491 |
-
"eval_loss": 2.
|
| 492 |
-
"eval_runtime": 4.
|
| 493 |
-
"eval_samples_per_second": 4.
|
| 494 |
-
"eval_steps_per_second": 1.
|
| 495 |
"step": 275
|
| 496 |
},
|
| 497 |
{
|
| 498 |
"epoch": 5.02,
|
| 499 |
"learning_rate": 2.918486226104327e-05,
|
| 500 |
-
"loss":
|
| 501 |
"step": 276
|
| 502 |
},
|
| 503 |
{
|
| 504 |
"epoch": 5.09,
|
| 505 |
"learning_rate": 2.8557870956832132e-05,
|
| 506 |
-
"loss":
|
| 507 |
"step": 280
|
| 508 |
},
|
| 509 |
{
|
| 510 |
"epoch": 5.16,
|
| 511 |
"learning_rate": 2.792858679969596e-05,
|
| 512 |
-
"loss":
|
| 513 |
"step": 284
|
| 514 |
},
|
| 515 |
{
|
| 516 |
"epoch": 5.24,
|
| 517 |
"learning_rate": 2.7297415328800692e-05,
|
| 518 |
-
"loss":
|
| 519 |
"step": 288
|
| 520 |
},
|
| 521 |
{
|
| 522 |
"epoch": 5.31,
|
| 523 |
"learning_rate": 2.6664763299582602e-05,
|
| 524 |
-
"loss": 0.
|
| 525 |
"step": 292
|
| 526 |
},
|
| 527 |
{
|
| 528 |
"epoch": 5.38,
|
| 529 |
"learning_rate": 2.6031038421616683e-05,
|
| 530 |
-
"loss":
|
| 531 |
"step": 296
|
| 532 |
},
|
| 533 |
{
|
| 534 |
"epoch": 5.45,
|
| 535 |
"learning_rate": 2.5396649095870202e-05,
|
| 536 |
-
"loss":
|
| 537 |
"step": 300
|
| 538 |
},
|
| 539 |
{
|
| 540 |
"epoch": 5.53,
|
| 541 |
"learning_rate": 2.4762004151510584e-05,
|
| 542 |
-
"loss": 0.
|
| 543 |
"step": 304
|
| 544 |
},
|
| 545 |
{
|
| 546 |
"epoch": 5.6,
|
| 547 |
"learning_rate": 2.4127512582437485e-05,
|
| 548 |
-
"loss": 0.
|
| 549 |
"step": 308
|
| 550 |
},
|
| 551 |
{
|
| 552 |
"epoch": 5.67,
|
| 553 |
"learning_rate": 2.349358328370854e-05,
|
| 554 |
-
"loss":
|
| 555 |
"step": 312
|
| 556 |
},
|
| 557 |
{
|
| 558 |
"epoch": 5.75,
|
| 559 |
"learning_rate": 2.2860624788029013e-05,
|
| 560 |
-
"loss":
|
| 561 |
"step": 316
|
| 562 |
},
|
| 563 |
{
|
| 564 |
"epoch": 5.82,
|
| 565 |
"learning_rate": 2.222904500247473e-05,
|
| 566 |
-
"loss":
|
| 567 |
"step": 320
|
| 568 |
},
|
| 569 |
{
|
| 570 |
"epoch": 5.89,
|
| 571 |
"learning_rate": 2.1599250945618402e-05,
|
| 572 |
-
"loss": 0.
|
| 573 |
"step": 324
|
| 574 |
},
|
| 575 |
{
|
| 576 |
"epoch": 5.96,
|
| 577 |
"learning_rate": 2.09716484852284e-05,
|
| 578 |
-
"loss":
|
| 579 |
"step": 328
|
| 580 |
},
|
| 581 |
-
{
|
| 582 |
-
"epoch": 6.0,
|
| 583 |
-
"gpt4_scores": 0.54,
|
| 584 |
-
"step": 330
|
| 585 |
-
},
|
| 586 |
-
{
|
| 587 |
-
"epoch": 6.0,
|
| 588 |
-
"std": 0.06151422599691879,
|
| 589 |
-
"step": 330
|
| 590 |
-
},
|
| 591 |
-
{
|
| 592 |
-
"epoch": 6.0,
|
| 593 |
-
"eval_loss": 2.538802146911621,
|
| 594 |
-
"eval_runtime": 4.9512,
|
| 595 |
-
"eval_samples_per_second": 4.645,
|
| 596 |
-
"eval_steps_per_second": 1.212,
|
| 597 |
-
"step": 330
|
| 598 |
-
},
|
| 599 |
{
|
| 600 |
"epoch": 6.04,
|
| 601 |
"learning_rate": 2.034664207670925e-05,
|
| 602 |
-
"loss": 0.
|
| 603 |
"step": 332
|
| 604 |
},
|
| 605 |
{
|
| 606 |
"epoch": 6.11,
|
| 607 |
"learning_rate": 1.972463450245226e-05,
|
| 608 |
-
"loss": 0.
|
| 609 |
"step": 336
|
| 610 |
},
|
| 611 |
{
|
| 612 |
"epoch": 6.18,
|
| 613 |
"learning_rate": 1.9106026612264316e-05,
|
| 614 |
-
"loss": 0.
|
| 615 |
"step": 340
|
| 616 |
},
|
| 617 |
{
|
| 618 |
"epoch": 6.25,
|
| 619 |
"learning_rate": 1.84912170650422e-05,
|
| 620 |
-
"loss": 0.
|
| 621 |
"step": 344
|
| 622 |
},
|
| 623 |
{
|
| 624 |
"epoch": 6.33,
|
| 625 |
"learning_rate": 1.7880602071858692e-05,
|
| 626 |
-
"loss": 0.
|
| 627 |
"step": 348
|
| 628 |
},
|
| 629 |
{
|
| 630 |
"epoch": 6.4,
|
| 631 |
"learning_rate": 1.7274575140626318e-05,
|
| 632 |
-
"loss": 0.
|
| 633 |
"step": 352
|
| 634 |
},
|
| 635 |
{
|
| 636 |
"epoch": 6.47,
|
| 637 |
"learning_rate": 1.667352682250298e-05,
|
| 638 |
-
"loss": 0.
|
| 639 |
"step": 356
|
| 640 |
},
|
| 641 |
{
|
| 642 |
"epoch": 6.55,
|
| 643 |
"learning_rate": 1.6077844460203206e-05,
|
| 644 |
-
"loss": 0.
|
| 645 |
"step": 360
|
| 646 |
},
|
| 647 |
{
|
| 648 |
"epoch": 6.62,
|
| 649 |
"learning_rate": 1.5487911938376924e-05,
|
| 650 |
-
"loss": 0.
|
| 651 |
"step": 364
|
| 652 |
},
|
| 653 |
{
|
| 654 |
"epoch": 6.69,
|
| 655 |
"learning_rate": 1.4904109436216884e-05,
|
| 656 |
-
"loss": 0.
|
| 657 |
"step": 368
|
| 658 |
},
|
| 659 |
{
|
| 660 |
"epoch": 6.76,
|
| 661 |
"learning_rate": 1.4326813182453958e-05,
|
| 662 |
-
"loss": 0.
|
| 663 |
"step": 372
|
| 664 |
},
|
| 665 |
{
|
| 666 |
"epoch": 6.84,
|
| 667 |
"learning_rate": 1.3756395212898359e-05,
|
| 668 |
-
"loss": 0.
|
| 669 |
"step": 376
|
| 670 |
},
|
| 671 |
{
|
| 672 |
"epoch": 6.91,
|
| 673 |
"learning_rate": 1.3193223130682936e-05,
|
| 674 |
-
"loss": 0.
|
| 675 |
"step": 380
|
| 676 |
},
|
| 677 |
{
|
| 678 |
"epoch": 6.98,
|
| 679 |
"learning_rate": 1.2637659869363083e-05,
|
| 680 |
-
"loss": 0.
|
| 681 |
"step": 384
|
| 682 |
},
|
| 683 |
{
|
| 684 |
"epoch": 7.0,
|
| 685 |
-
"
|
|
|
|
| 686 |
"step": 385
|
| 687 |
},
|
| 688 |
{
|
| 689 |
"epoch": 7.0,
|
| 690 |
-
"
|
| 691 |
-
"
|
| 692 |
-
|
| 693 |
-
|
| 694 |
-
"epoch": 7.0,
|
| 695 |
-
"eval_loss": 2.774492025375366,
|
| 696 |
-
"eval_runtime": 4.9875,
|
| 697 |
-
"eval_samples_per_second": 4.612,
|
| 698 |
-
"eval_steps_per_second": 1.203,
|
| 699 |
"step": 385
|
| 700 |
},
|
| 701 |
{
|
| 702 |
"epoch": 7.05,
|
| 703 |
"learning_rate": 1.2090063459025955e-05,
|
| 704 |
-
"loss": 0.
|
| 705 |
"step": 388
|
| 706 |
},
|
| 707 |
{
|
| 708 |
"epoch": 7.13,
|
| 709 |
"learning_rate": 1.155078679555969e-05,
|
| 710 |
-
"loss": 0.
|
| 711 |
"step": 392
|
| 712 |
},
|
| 713 |
{
|
| 714 |
"epoch": 7.2,
|
| 715 |
"learning_rate": 1.1020177413231334e-05,
|
| 716 |
-
"loss": 0.
|
| 717 |
"step": 396
|
| 718 |
},
|
| 719 |
{
|
| 720 |
"epoch": 7.27,
|
| 721 |
"learning_rate": 1.049857726072005e-05,
|
| 722 |
-
"loss": 0.
|
| 723 |
"step": 400
|
| 724 |
},
|
| 725 |
{
|
| 726 |
"epoch": 7.35,
|
| 727 |
"learning_rate": 9.986322480749927e-06,
|
| 728 |
-
"loss": 0.
|
| 729 |
"step": 404
|
| 730 |
},
|
| 731 |
{
|
| 732 |
"epoch": 7.42,
|
| 733 |
"learning_rate": 9.483743193464408e-06,
|
| 734 |
-
"loss": 0.
|
| 735 |
"step": 408
|
| 736 |
},
|
| 737 |
{
|
| 738 |
"epoch": 7.49,
|
| 739 |
"learning_rate": 8.991163283681944e-06,
|
| 740 |
-
"loss": 0.
|
| 741 |
"step": 412
|
| 742 |
},
|
| 743 |
{
|
| 744 |
"epoch": 7.56,
|
| 745 |
"learning_rate": 8.508900192169964e-06,
|
| 746 |
-
"loss": 0.
|
| 747 |
"step": 416
|
| 748 |
},
|
| 749 |
{
|
| 750 |
"epoch": 7.64,
|
| 751 |
"learning_rate": 8.0372647110717e-06,
|
| 752 |
-
"loss": 0.
|
| 753 |
"step": 420
|
| 754 |
},
|
| 755 |
{
|
| 756 |
"epoch": 7.71,
|
| 757 |
"learning_rate": 7.576560783617668e-06,
|
| 758 |
-
"loss": 0.
|
| 759 |
"step": 424
|
| 760 |
},
|
| 761 |
{
|
| 762 |
"epoch": 7.78,
|
| 763 |
"learning_rate": 7.127085308250914e-06,
|
| 764 |
-
"loss": 0.
|
| 765 |
"step": 428
|
| 766 |
},
|
| 767 |
{
|
| 768 |
"epoch": 7.85,
|
| 769 |
"learning_rate": 6.689127947292231e-06,
|
| 770 |
-
"loss": 0.
|
| 771 |
"step": 432
|
| 772 |
},
|
| 773 |
{
|
| 774 |
"epoch": 7.93,
|
| 775 |
"learning_rate": 6.2629709402686535e-06,
|
| 776 |
-
"loss": 0.
|
| 777 |
"step": 436
|
| 778 |
},
|
| 779 |
{
|
| 780 |
"epoch": 8.0,
|
| 781 |
"learning_rate": 5.848888922025553e-06,
|
| 782 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 783 |
"step": 440
|
| 784 |
},
|
| 785 |
{
|
| 786 |
"epoch": 8.07,
|
| 787 |
"learning_rate": 5.4471487457395225e-06,
|
| 788 |
-
"loss":
|
| 789 |
"step": 444
|
| 790 |
},
|
| 791 |
{
|
| 792 |
"epoch": 8.15,
|
| 793 |
"learning_rate": 5.058009310946119e-06,
|
| 794 |
-
"loss":
|
| 795 |
"step": 448
|
| 796 |
},
|
| 797 |
{
|
| 798 |
"epoch": 8.22,
|
| 799 |
"learning_rate": 4.681721396693303e-06,
|
| 800 |
-
"loss":
|
| 801 |
"step": 452
|
| 802 |
},
|
| 803 |
{
|
| 804 |
"epoch": 8.29,
|
| 805 |
"learning_rate": 4.318527499928074e-06,
|
| 806 |
-
"loss": 0.
|
| 807 |
"step": 456
|
| 808 |
},
|
| 809 |
{
|
| 810 |
"epoch": 8.36,
|
| 811 |
"learning_rate": 3.968661679220468e-06,
|
| 812 |
-
"loss": 0.
|
| 813 |
"step": 460
|
| 814 |
},
|
| 815 |
{
|
| 816 |
"epoch": 8.44,
|
| 817 |
"learning_rate": 3.632349403925664e-06,
|
| 818 |
-
"loss": 0.
|
| 819 |
"step": 464
|
| 820 |
},
|
| 821 |
{
|
| 822 |
"epoch": 8.51,
|
| 823 |
"learning_rate": 3.3098074088812686e-06,
|
| 824 |
-
"loss": 0.
|
| 825 |
"step": 468
|
| 826 |
},
|
| 827 |
{
|
| 828 |
"epoch": 8.58,
|
| 829 |
"learning_rate": 3.0012435547336737e-06,
|
| 830 |
-
"loss": 0.
|
| 831 |
"step": 472
|
| 832 |
},
|
| 833 |
{
|
| 834 |
"epoch": 8.65,
|
| 835 |
"learning_rate": 2.7068566939831645e-06,
|
| 836 |
-
"loss": 0.
|
| 837 |
"step": 476
|
| 838 |
},
|
| 839 |
{
|
| 840 |
"epoch": 8.73,
|
| 841 |
"learning_rate": 2.4268365428344736e-06,
|
| 842 |
-
"loss": 0.
|
| 843 |
"step": 480
|
| 844 |
},
|
| 845 |
{
|
| 846 |
"epoch": 8.8,
|
| 847 |
"learning_rate": 2.1613635589349756e-06,
|
| 848 |
-
"loss": 0.
|
| 849 |
"step": 484
|
| 850 |
},
|
| 851 |
{
|
| 852 |
"epoch": 8.87,
|
| 853 |
"learning_rate": 1.9106088250797267e-06,
|
| 854 |
-
"loss": 0.
|
| 855 |
"step": 488
|
| 856 |
},
|
| 857 |
{
|
| 858 |
"epoch": 8.95,
|
| 859 |
"learning_rate": 1.674733938957873e-06,
|
| 860 |
-
"loss": 0.
|
| 861 |
"step": 492
|
| 862 |
},
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 863 |
{
|
| 864 |
"epoch": 9.02,
|
| 865 |
"learning_rate": 1.4538909090118846e-06,
|
| 866 |
-
"loss": 0.
|
| 867 |
"step": 496
|
| 868 |
},
|
| 869 |
{
|
| 870 |
"epoch": 9.09,
|
| 871 |
"learning_rate": 1.248222056476367e-06,
|
| 872 |
-
"loss": 0.
|
| 873 |
"step": 500
|
| 874 |
},
|
| 875 |
{
|
| 876 |
"epoch": 9.16,
|
| 877 |
"learning_rate": 1.0578599236598707e-06,
|
| 878 |
-
"loss": 0.
|
| 879 |
"step": 504
|
| 880 |
},
|
| 881 |
{
|
| 882 |
"epoch": 9.24,
|
| 883 |
"learning_rate": 8.829271885286094e-07,
|
| 884 |
-
"loss": 0.
|
| 885 |
"step": 508
|
| 886 |
},
|
| 887 |
{
|
| 888 |
"epoch": 9.31,
|
| 889 |
"learning_rate": 7.235365856472442e-07,
|
| 890 |
-
"loss": 0.
|
| 891 |
"step": 512
|
| 892 |
},
|
| 893 |
{
|
| 894 |
"epoch": 9.38,
|
| 895 |
"learning_rate": 5.797908335276214e-07,
|
| 896 |
-
"loss": 0.
|
| 897 |
"step": 516
|
| 898 |
},
|
| 899 |
{
|
| 900 |
"epoch": 9.45,
|
| 901 |
"learning_rate": 4.517825684323324e-07,
|
| 902 |
-
"loss": 0.
|
| 903 |
"step": 520
|
| 904 |
},
|
| 905 |
{
|
| 906 |
"epoch": 9.53,
|
| 907 |
"learning_rate": 3.395942846757066e-07,
|
| 908 |
-
"loss": 0.
|
| 909 |
"step": 524
|
| 910 |
},
|
| 911 |
{
|
| 912 |
"epoch": 9.6,
|
| 913 |
"learning_rate": 2.4329828146074095e-07,
|
| 914 |
-
"loss": 0.
|
| 915 |
"step": 528
|
| 916 |
},
|
| 917 |
{
|
| 918 |
"epoch": 9.67,
|
| 919 |
"learning_rate": 1.6295661628624447e-07,
|
| 920 |
-
"loss": 0.
|
| 921 |
"step": 532
|
| 922 |
},
|
| 923 |
{
|
| 924 |
"epoch": 9.75,
|
| 925 |
"learning_rate": 9.862106495415469e-08,
|
| 926 |
-
"loss": 0.
|
| 927 |
"step": 536
|
| 928 |
},
|
| 929 |
{
|
| 930 |
"epoch": 9.82,
|
| 931 |
"learning_rate": 5.033308820289184e-08,
|
| 932 |
-
"loss": 0.
|
| 933 |
"step": 540
|
| 934 |
},
|
| 935 |
{
|
| 936 |
"epoch": 9.89,
|
| 937 |
"learning_rate": 1.812380498815991e-08,
|
| 938 |
-
"loss": 0.
|
| 939 |
"step": 544
|
| 940 |
},
|
| 941 |
{
|
| 942 |
"epoch": 9.96,
|
| 943 |
"learning_rate": 2.0139724285161977e-09,
|
| 944 |
-
"loss": 0.
|
| 945 |
"step": 548
|
| 946 |
},
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 947 |
{
|
| 948 |
"epoch": 10.0,
|
| 949 |
"step": 550,
|
| 950 |
-
"total_flos": 1.
|
| 951 |
-
"train_loss": 0.
|
| 952 |
-
"train_runtime":
|
| 953 |
-
"train_samples_per_second":
|
| 954 |
-
"train_steps_per_second":
|
| 955 |
}
|
| 956 |
],
|
| 957 |
"logging_steps": 4,
|
|
@@ -959,7 +963,7 @@
|
|
| 959 |
"num_input_tokens_seen": 0,
|
| 960 |
"num_train_epochs": 10,
|
| 961 |
"save_steps": 55,
|
| 962 |
-
"total_flos": 1.
|
| 963 |
"train_batch_size": 4,
|
| 964 |
"trial_name": null,
|
| 965 |
"trial_params": null
|
|
|
|
| 11 |
{
|
| 12 |
"epoch": 0.02,
|
| 13 |
"learning_rate": 9.09090909090909e-07,
|
| 14 |
+
"loss": 2.7431,
|
| 15 |
"step": 1
|
| 16 |
},
|
| 17 |
{
|
| 18 |
"epoch": 0.07,
|
| 19 |
"learning_rate": 3.636363636363636e-06,
|
| 20 |
+
"loss": 2.889,
|
| 21 |
"step": 4
|
| 22 |
},
|
| 23 |
{
|
| 24 |
"epoch": 0.15,
|
| 25 |
"learning_rate": 7.272727272727272e-06,
|
| 26 |
+
"loss": 2.7193,
|
| 27 |
"step": 8
|
| 28 |
},
|
| 29 |
{
|
| 30 |
"epoch": 0.22,
|
| 31 |
"learning_rate": 1.0909090909090909e-05,
|
| 32 |
+
"loss": 2.6675,
|
| 33 |
"step": 12
|
| 34 |
},
|
| 35 |
{
|
| 36 |
"epoch": 0.29,
|
| 37 |
"learning_rate": 1.4545454545454545e-05,
|
| 38 |
+
"loss": 2.608,
|
| 39 |
"step": 16
|
| 40 |
},
|
| 41 |
{
|
| 42 |
"epoch": 0.36,
|
| 43 |
"learning_rate": 1.8181818181818182e-05,
|
| 44 |
+
"loss": 2.6623,
|
| 45 |
"step": 20
|
| 46 |
},
|
| 47 |
{
|
| 48 |
"epoch": 0.44,
|
| 49 |
"learning_rate": 2.1818181818181818e-05,
|
| 50 |
+
"loss": 2.7249,
|
| 51 |
"step": 24
|
| 52 |
},
|
| 53 |
{
|
| 54 |
"epoch": 0.51,
|
| 55 |
"learning_rate": 2.5454545454545454e-05,
|
| 56 |
+
"loss": 2.7954,
|
| 57 |
"step": 28
|
| 58 |
},
|
| 59 |
{
|
| 60 |
"epoch": 0.58,
|
| 61 |
"learning_rate": 2.909090909090909e-05,
|
| 62 |
+
"loss": 2.6101,
|
| 63 |
"step": 32
|
| 64 |
},
|
| 65 |
{
|
| 66 |
"epoch": 0.65,
|
| 67 |
"learning_rate": 3.272727272727273e-05,
|
| 68 |
+
"loss": 2.4391,
|
| 69 |
"step": 36
|
| 70 |
},
|
| 71 |
{
|
| 72 |
"epoch": 0.73,
|
| 73 |
"learning_rate": 3.6363636363636364e-05,
|
| 74 |
+
"loss": 2.2331,
|
| 75 |
"step": 40
|
| 76 |
},
|
| 77 |
{
|
| 78 |
"epoch": 0.8,
|
| 79 |
"learning_rate": 4e-05,
|
| 80 |
+
"loss": 2.2489,
|
| 81 |
"step": 44
|
| 82 |
},
|
| 83 |
{
|
| 84 |
"epoch": 0.87,
|
| 85 |
"learning_rate": 4.3636363636363636e-05,
|
| 86 |
+
"loss": 2.1787,
|
| 87 |
"step": 48
|
| 88 |
},
|
| 89 |
{
|
| 90 |
"epoch": 0.95,
|
| 91 |
"learning_rate": 4.7272727272727275e-05,
|
| 92 |
+
"loss": 2.2181,
|
| 93 |
"step": 52
|
| 94 |
},
|
| 95 |
{
|
| 96 |
"epoch": 1.02,
|
| 97 |
"learning_rate": 4.999949650182266e-05,
|
| 98 |
+
"loss": 2.2997,
|
| 99 |
"step": 56
|
| 100 |
},
|
| 101 |
{
|
| 102 |
"epoch": 1.09,
|
| 103 |
"learning_rate": 4.9987413559579636e-05,
|
| 104 |
+
"loss": 2.0619,
|
| 105 |
"step": 60
|
| 106 |
},
|
| 107 |
{
|
| 108 |
"epoch": 1.16,
|
| 109 |
"learning_rate": 4.995922759815339e-05,
|
| 110 |
+
"loss": 2.1645,
|
| 111 |
"step": 64
|
| 112 |
},
|
| 113 |
{
|
| 114 |
"epoch": 1.24,
|
| 115 |
"learning_rate": 4.991495678185202e-05,
|
| 116 |
+
"loss": 1.9819,
|
| 117 |
"step": 68
|
| 118 |
},
|
| 119 |
{
|
| 120 |
"epoch": 1.31,
|
| 121 |
"learning_rate": 4.985462964079137e-05,
|
| 122 |
+
"loss": 2.0363,
|
| 123 |
"step": 72
|
| 124 |
},
|
| 125 |
{
|
| 126 |
"epoch": 1.38,
|
| 127 |
"learning_rate": 4.977828505250903e-05,
|
| 128 |
+
"loss": 1.9152,
|
| 129 |
"step": 76
|
| 130 |
},
|
| 131 |
{
|
| 132 |
"epoch": 1.45,
|
| 133 |
"learning_rate": 4.968597221690986e-05,
|
| 134 |
+
"loss": 1.9794,
|
| 135 |
"step": 80
|
| 136 |
},
|
| 137 |
{
|
| 138 |
"epoch": 1.53,
|
| 139 |
"learning_rate": 4.957775062455933e-05,
|
| 140 |
+
"loss": 1.9805,
|
| 141 |
"step": 84
|
| 142 |
},
|
| 143 |
{
|
| 144 |
"epoch": 1.6,
|
| 145 |
"learning_rate": 4.9453690018345144e-05,
|
| 146 |
+
"loss": 1.9941,
|
| 147 |
"step": 88
|
| 148 |
},
|
| 149 |
{
|
| 150 |
"epoch": 1.67,
|
| 151 |
"learning_rate": 4.931387034853173e-05,
|
| 152 |
+
"loss": 1.9186,
|
| 153 |
"step": 92
|
| 154 |
},
|
| 155 |
{
|
| 156 |
"epoch": 1.75,
|
| 157 |
"learning_rate": 4.915838172123671e-05,
|
| 158 |
+
"loss": 1.9123,
|
| 159 |
"step": 96
|
| 160 |
},
|
| 161 |
{
|
| 162 |
"epoch": 1.82,
|
| 163 |
"learning_rate": 4.898732434036244e-05,
|
| 164 |
+
"loss": 1.965,
|
| 165 |
"step": 100
|
| 166 |
},
|
| 167 |
{
|
| 168 |
"epoch": 1.89,
|
| 169 |
"learning_rate": 4.880080844302004e-05,
|
| 170 |
+
"loss": 1.8973,
|
| 171 |
"step": 104
|
| 172 |
},
|
| 173 |
{
|
| 174 |
"epoch": 1.96,
|
| 175 |
"learning_rate": 4.859895422848767e-05,
|
| 176 |
+
"loss": 1.8473,
|
| 177 |
"step": 108
|
| 178 |
},
|
| 179 |
{
|
| 180 |
"epoch": 2.0,
|
| 181 |
+
"pls_score": 58.0,
|
| 182 |
+
"std": 4.354308211415448,
|
| 183 |
"step": 110
|
| 184 |
},
|
| 185 |
{
|
| 186 |
"epoch": 2.0,
|
| 187 |
+
"eval_loss": 1.9843206405639648,
|
| 188 |
+
"eval_runtime": 4.904,
|
| 189 |
+
"eval_samples_per_second": 4.69,
|
| 190 |
+
"eval_steps_per_second": 1.223,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 191 |
"step": 110
|
| 192 |
},
|
| 193 |
{
|
| 194 |
"epoch": 2.04,
|
| 195 |
"learning_rate": 4.838189178074867e-05,
|
| 196 |
+
"loss": 2.0581,
|
| 197 |
"step": 112
|
| 198 |
},
|
| 199 |
{
|
| 200 |
"epoch": 2.11,
|
| 201 |
"learning_rate": 4.8149760984659506e-05,
|
| 202 |
+
"loss": 1.8467,
|
| 203 |
"step": 116
|
| 204 |
},
|
| 205 |
{
|
| 206 |
"epoch": 2.18,
|
| 207 |
"learning_rate": 4.790271143580174e-05,
|
| 208 |
+
"loss": 1.9691,
|
| 209 |
"step": 120
|
| 210 |
},
|
| 211 |
{
|
| 212 |
"epoch": 2.25,
|
| 213 |
"learning_rate": 4.764090234407577e-05,
|
| 214 |
+
"loss": 1.8183,
|
| 215 |
"step": 124
|
| 216 |
},
|
| 217 |
{
|
| 218 |
"epoch": 2.33,
|
| 219 |
"learning_rate": 4.7364502431098844e-05,
|
| 220 |
+
"loss": 1.7403,
|
| 221 |
"step": 128
|
| 222 |
},
|
| 223 |
{
|
| 224 |
"epoch": 2.4,
|
| 225 |
"learning_rate": 4.707368982147318e-05,
|
| 226 |
+
"loss": 1.7877,
|
| 227 |
"step": 132
|
| 228 |
},
|
| 229 |
{
|
| 230 |
"epoch": 2.47,
|
| 231 |
"learning_rate": 4.6768651927994434e-05,
|
| 232 |
+
"loss": 1.7595,
|
| 233 |
"step": 136
|
| 234 |
},
|
| 235 |
{
|
| 236 |
"epoch": 2.55,
|
| 237 |
"learning_rate": 4.644958533087443e-05,
|
| 238 |
+
"loss": 1.7273,
|
| 239 |
"step": 140
|
| 240 |
},
|
| 241 |
{
|
| 242 |
"epoch": 2.62,
|
| 243 |
"learning_rate": 4.611669565105596e-05,
|
| 244 |
+
"loss": 1.669,
|
| 245 |
"step": 144
|
| 246 |
},
|
| 247 |
{
|
| 248 |
"epoch": 2.69,
|
| 249 |
"learning_rate": 4.5770197417701365e-05,
|
| 250 |
+
"loss": 1.742,
|
| 251 |
"step": 148
|
| 252 |
},
|
| 253 |
{
|
| 254 |
"epoch": 2.76,
|
| 255 |
"learning_rate": 4.5410313929940244e-05,
|
| 256 |
+
"loss": 1.6976,
|
| 257 |
"step": 152
|
| 258 |
},
|
| 259 |
{
|
| 260 |
"epoch": 2.84,
|
| 261 |
"learning_rate": 4.503727711296538e-05,
|
| 262 |
+
"loss": 1.8632,
|
| 263 |
"step": 156
|
| 264 |
},
|
| 265 |
{
|
| 266 |
"epoch": 2.91,
|
| 267 |
"learning_rate": 4.465132736856969e-05,
|
| 268 |
+
"loss": 1.9186,
|
| 269 |
"step": 160
|
| 270 |
},
|
| 271 |
{
|
| 272 |
"epoch": 2.98,
|
| 273 |
"learning_rate": 4.425271342022039e-05,
|
| 274 |
+
"loss": 1.7645,
|
| 275 |
"step": 164
|
| 276 |
},
|
| 277 |
{
|
| 278 |
"epoch": 3.0,
|
| 279 |
+
"pls_score": 64.0,
|
| 280 |
+
"std": 3.7523326078587433,
|
| 281 |
"step": 165
|
| 282 |
},
|
| 283 |
{
|
| 284 |
"epoch": 3.0,
|
| 285 |
+
"eval_loss": 1.968030333518982,
|
| 286 |
+
"eval_runtime": 4.9015,
|
| 287 |
+
"eval_samples_per_second": 4.692,
|
| 288 |
+
"eval_steps_per_second": 1.224,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 289 |
"step": 165
|
| 290 |
},
|
| 291 |
{
|
| 292 |
"epoch": 3.05,
|
| 293 |
"learning_rate": 4.384169215277041e-05,
|
| 294 |
+
"loss": 1.4812,
|
| 295 |
"step": 168
|
| 296 |
},
|
| 297 |
{
|
| 298 |
"epoch": 3.13,
|
| 299 |
"learning_rate": 4.341852844691012e-05,
|
| 300 |
+
"loss": 1.6649,
|
| 301 |
"step": 172
|
| 302 |
},
|
| 303 |
{
|
| 304 |
"epoch": 3.2,
|
| 305 |
"learning_rate": 4.2983495008466276e-05,
|
| 306 |
+
"loss": 1.5835,
|
| 307 |
"step": 176
|
| 308 |
},
|
| 309 |
{
|
| 310 |
"epoch": 3.27,
|
| 311 |
"learning_rate": 4.2536872192658036e-05,
|
| 312 |
+
"loss": 1.6401,
|
| 313 |
"step": 180
|
| 314 |
},
|
| 315 |
{
|
| 316 |
"epoch": 3.35,
|
| 317 |
"learning_rate": 4.2078947823423364e-05,
|
| 318 |
+
"loss": 1.573,
|
| 319 |
"step": 184
|
| 320 |
},
|
| 321 |
{
|
| 322 |
"epoch": 3.42,
|
| 323 |
"learning_rate": 4.161001700793231e-05,
|
| 324 |
+
"loss": 1.4984,
|
| 325 |
"step": 188
|
| 326 |
},
|
| 327 |
{
|
| 328 |
"epoch": 3.49,
|
| 329 |
"learning_rate": 4.113038194640658e-05,
|
| 330 |
+
"loss": 1.5315,
|
| 331 |
"step": 192
|
| 332 |
},
|
| 333 |
{
|
| 334 |
"epoch": 3.56,
|
| 335 |
"learning_rate": 4.064035173736804e-05,
|
| 336 |
+
"loss": 1.541,
|
| 337 |
"step": 196
|
| 338 |
},
|
| 339 |
{
|
| 340 |
"epoch": 3.64,
|
| 341 |
"learning_rate": 4.014024217844167e-05,
|
| 342 |
+
"loss": 1.5492,
|
| 343 |
"step": 200
|
| 344 |
},
|
| 345 |
{
|
| 346 |
"epoch": 3.71,
|
| 347 |
"learning_rate": 3.9630375562841295e-05,
|
| 348 |
+
"loss": 1.6265,
|
| 349 |
"step": 204
|
| 350 |
},
|
| 351 |
{
|
| 352 |
"epoch": 3.78,
|
| 353 |
"learning_rate": 3.911108047166924e-05,
|
| 354 |
+
"loss": 1.5753,
|
| 355 |
"step": 208
|
| 356 |
},
|
| 357 |
{
|
| 358 |
"epoch": 3.85,
|
| 359 |
"learning_rate": 3.858269156216383e-05,
|
| 360 |
+
"loss": 1.4876,
|
| 361 |
"step": 212
|
| 362 |
},
|
| 363 |
{
|
| 364 |
"epoch": 3.93,
|
| 365 |
"learning_rate": 3.804554935203115e-05,
|
| 366 |
+
"loss": 1.5884,
|
| 367 |
"step": 216
|
| 368 |
},
|
| 369 |
{
|
| 370 |
"epoch": 4.0,
|
| 371 |
"learning_rate": 3.7500000000000003e-05,
|
| 372 |
+
"loss": 1.4795,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 373 |
"step": 220
|
| 374 |
},
|
| 375 |
{
|
| 376 |
"epoch": 4.0,
|
| 377 |
+
"pls_score": 60.8,
|
| 378 |
+
"std": 3.750626614313932,
|
| 379 |
"step": 220
|
| 380 |
},
|
| 381 |
{
|
| 382 |
"epoch": 4.0,
|
| 383 |
+
"eval_loss": 2.076031446456909,
|
| 384 |
+
"eval_runtime": 4.9029,
|
| 385 |
+
"eval_samples_per_second": 4.691,
|
| 386 |
+
"eval_steps_per_second": 1.224,
|
| 387 |
"step": 220
|
| 388 |
},
|
| 389 |
{
|
| 390 |
"epoch": 4.07,
|
| 391 |
"learning_rate": 3.694639508274158e-05,
|
| 392 |
+
"loss": 1.3002,
|
| 393 |
"step": 224
|
| 394 |
},
|
| 395 |
{
|
| 396 |
"epoch": 4.15,
|
| 397 |
"learning_rate": 3.638509136829758e-05,
|
| 398 |
+
"loss": 1.344,
|
| 399 |
"step": 228
|
| 400 |
},
|
| 401 |
{
|
| 402 |
"epoch": 4.22,
|
| 403 |
"learning_rate": 3.581645058616271e-05,
|
| 404 |
+
"loss": 1.209,
|
| 405 |
"step": 232
|
| 406 |
},
|
| 407 |
{
|
| 408 |
"epoch": 4.29,
|
| 409 |
"learning_rate": 3.5240839194169885e-05,
|
| 410 |
+
"loss": 1.3386,
|
| 411 |
"step": 236
|
| 412 |
},
|
| 413 |
{
|
| 414 |
"epoch": 4.36,
|
| 415 |
"learning_rate": 3.465862814232822e-05,
|
| 416 |
+
"loss": 1.3247,
|
| 417 |
"step": 240
|
| 418 |
},
|
| 419 |
{
|
| 420 |
"epoch": 4.44,
|
| 421 |
"learning_rate": 3.4070192633766025e-05,
|
| 422 |
+
"loss": 1.2247,
|
| 423 |
"step": 244
|
| 424 |
},
|
| 425 |
{
|
| 426 |
"epoch": 4.51,
|
| 427 |
"learning_rate": 3.3475911882933015e-05,
|
| 428 |
+
"loss": 1.2683,
|
| 429 |
"step": 248
|
| 430 |
},
|
| 431 |
{
|
| 432 |
"epoch": 4.58,
|
| 433 |
"learning_rate": 3.2876168871217325e-05,
|
| 434 |
+
"loss": 1.2856,
|
| 435 |
"step": 252
|
| 436 |
},
|
| 437 |
{
|
| 438 |
"epoch": 4.65,
|
| 439 |
"learning_rate": 3.2271350100134975e-05,
|
| 440 |
+
"loss": 1.271,
|
| 441 |
"step": 256
|
| 442 |
},
|
| 443 |
{
|
| 444 |
"epoch": 4.73,
|
| 445 |
"learning_rate": 3.166184534225087e-05,
|
| 446 |
+
"loss": 1.207,
|
| 447 |
"step": 260
|
| 448 |
},
|
| 449 |
{
|
| 450 |
"epoch": 4.8,
|
| 451 |
"learning_rate": 3.104804738999169e-05,
|
| 452 |
+
"loss": 1.2884,
|
| 453 |
"step": 264
|
| 454 |
},
|
| 455 |
{
|
| 456 |
"epoch": 4.87,
|
| 457 |
"learning_rate": 3.0430351802512698e-05,
|
| 458 |
+
"loss": 1.2375,
|
| 459 |
"step": 268
|
| 460 |
},
|
| 461 |
{
|
| 462 |
"epoch": 4.95,
|
| 463 |
"learning_rate": 2.9809156650781528e-05,
|
| 464 |
+
"loss": 1.2467,
|
| 465 |
"step": 272
|
| 466 |
},
|
| 467 |
{
|
| 468 |
"epoch": 5.0,
|
| 469 |
+
"pls_score": 62.8,
|
| 470 |
+
"std": 4.15730682052696,
|
|
|
|
|
|
|
|
|
|
|
|
|
| 471 |
"step": 275
|
| 472 |
},
|
| 473 |
{
|
| 474 |
"epoch": 5.0,
|
| 475 |
+
"eval_loss": 2.2714545726776123,
|
| 476 |
+
"eval_runtime": 4.904,
|
| 477 |
+
"eval_samples_per_second": 4.69,
|
| 478 |
+
"eval_steps_per_second": 1.224,
|
| 479 |
"step": 275
|
| 480 |
},
|
| 481 |
{
|
| 482 |
"epoch": 5.02,
|
| 483 |
"learning_rate": 2.918486226104327e-05,
|
| 484 |
+
"loss": 1.1868,
|
| 485 |
"step": 276
|
| 486 |
},
|
| 487 |
{
|
| 488 |
"epoch": 5.09,
|
| 489 |
"learning_rate": 2.8557870956832132e-05,
|
| 490 |
+
"loss": 1.099,
|
| 491 |
"step": 280
|
| 492 |
},
|
| 493 |
{
|
| 494 |
"epoch": 5.16,
|
| 495 |
"learning_rate": 2.792858679969596e-05,
|
| 496 |
+
"loss": 1.0657,
|
| 497 |
"step": 284
|
| 498 |
},
|
| 499 |
{
|
| 500 |
"epoch": 5.24,
|
| 501 |
"learning_rate": 2.7297415328800692e-05,
|
| 502 |
+
"loss": 1.0596,
|
| 503 |
"step": 288
|
| 504 |
},
|
| 505 |
{
|
| 506 |
"epoch": 5.31,
|
| 507 |
"learning_rate": 2.6664763299582602e-05,
|
| 508 |
+
"loss": 0.9201,
|
| 509 |
"step": 292
|
| 510 |
},
|
| 511 |
{
|
| 512 |
"epoch": 5.38,
|
| 513 |
"learning_rate": 2.6031038421616683e-05,
|
| 514 |
+
"loss": 1.0683,
|
| 515 |
"step": 296
|
| 516 |
},
|
| 517 |
{
|
| 518 |
"epoch": 5.45,
|
| 519 |
"learning_rate": 2.5396649095870202e-05,
|
| 520 |
+
"loss": 1.0069,
|
| 521 |
"step": 300
|
| 522 |
},
|
| 523 |
{
|
| 524 |
"epoch": 5.53,
|
| 525 |
"learning_rate": 2.4762004151510584e-05,
|
| 526 |
+
"loss": 0.9486,
|
| 527 |
"step": 304
|
| 528 |
},
|
| 529 |
{
|
| 530 |
"epoch": 5.6,
|
| 531 |
"learning_rate": 2.4127512582437485e-05,
|
| 532 |
+
"loss": 0.937,
|
| 533 |
"step": 308
|
| 534 |
},
|
| 535 |
{
|
| 536 |
"epoch": 5.67,
|
| 537 |
"learning_rate": 2.349358328370854e-05,
|
| 538 |
+
"loss": 1.0102,
|
| 539 |
"step": 312
|
| 540 |
},
|
| 541 |
{
|
| 542 |
"epoch": 5.75,
|
| 543 |
"learning_rate": 2.2860624788029013e-05,
|
| 544 |
+
"loss": 1.0058,
|
| 545 |
"step": 316
|
| 546 |
},
|
| 547 |
{
|
| 548 |
"epoch": 5.82,
|
| 549 |
"learning_rate": 2.222904500247473e-05,
|
| 550 |
+
"loss": 1.0564,
|
| 551 |
"step": 320
|
| 552 |
},
|
| 553 |
{
|
| 554 |
"epoch": 5.89,
|
| 555 |
"learning_rate": 2.1599250945618402e-05,
|
| 556 |
+
"loss": 0.9122,
|
| 557 |
"step": 324
|
| 558 |
},
|
| 559 |
{
|
| 560 |
"epoch": 5.96,
|
| 561 |
"learning_rate": 2.09716484852284e-05,
|
| 562 |
+
"loss": 1.0034,
|
| 563 |
"step": 328
|
| 564 |
},
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 565 |
{
|
| 566 |
"epoch": 6.04,
|
| 567 |
"learning_rate": 2.034664207670925e-05,
|
| 568 |
+
"loss": 0.8258,
|
| 569 |
"step": 332
|
| 570 |
},
|
| 571 |
{
|
| 572 |
"epoch": 6.11,
|
| 573 |
"learning_rate": 1.972463450245226e-05,
|
| 574 |
+
"loss": 0.8152,
|
| 575 |
"step": 336
|
| 576 |
},
|
| 577 |
{
|
| 578 |
"epoch": 6.18,
|
| 579 |
"learning_rate": 1.9106026612264316e-05,
|
| 580 |
+
"loss": 0.7462,
|
| 581 |
"step": 340
|
| 582 |
},
|
| 583 |
{
|
| 584 |
"epoch": 6.25,
|
| 585 |
"learning_rate": 1.84912170650422e-05,
|
| 586 |
+
"loss": 0.7995,
|
| 587 |
"step": 344
|
| 588 |
},
|
| 589 |
{
|
| 590 |
"epoch": 6.33,
|
| 591 |
"learning_rate": 1.7880602071858692e-05,
|
| 592 |
+
"loss": 0.7714,
|
| 593 |
"step": 348
|
| 594 |
},
|
| 595 |
{
|
| 596 |
"epoch": 6.4,
|
| 597 |
"learning_rate": 1.7274575140626318e-05,
|
| 598 |
+
"loss": 0.8581,
|
| 599 |
"step": 352
|
| 600 |
},
|
| 601 |
{
|
| 602 |
"epoch": 6.47,
|
| 603 |
"learning_rate": 1.667352682250298e-05,
|
| 604 |
+
"loss": 0.7631,
|
| 605 |
"step": 356
|
| 606 |
},
|
| 607 |
{
|
| 608 |
"epoch": 6.55,
|
| 609 |
"learning_rate": 1.6077844460203206e-05,
|
| 610 |
+
"loss": 0.8117,
|
| 611 |
"step": 360
|
| 612 |
},
|
| 613 |
{
|
| 614 |
"epoch": 6.62,
|
| 615 |
"learning_rate": 1.5487911938376924e-05,
|
| 616 |
+
"loss": 0.7725,
|
| 617 |
"step": 364
|
| 618 |
},
|
| 619 |
{
|
| 620 |
"epoch": 6.69,
|
| 621 |
"learning_rate": 1.4904109436216884e-05,
|
| 622 |
+
"loss": 0.8519,
|
| 623 |
"step": 368
|
| 624 |
},
|
| 625 |
{
|
| 626 |
"epoch": 6.76,
|
| 627 |
"learning_rate": 1.4326813182453958e-05,
|
| 628 |
+
"loss": 0.8037,
|
| 629 |
"step": 372
|
| 630 |
},
|
| 631 |
{
|
| 632 |
"epoch": 6.84,
|
| 633 |
"learning_rate": 1.3756395212898359e-05,
|
| 634 |
+
"loss": 0.848,
|
| 635 |
"step": 376
|
| 636 |
},
|
| 637 |
{
|
| 638 |
"epoch": 6.91,
|
| 639 |
"learning_rate": 1.3193223130682936e-05,
|
| 640 |
+
"loss": 0.797,
|
| 641 |
"step": 380
|
| 642 |
},
|
| 643 |
{
|
| 644 |
"epoch": 6.98,
|
| 645 |
"learning_rate": 1.2637659869363083e-05,
|
| 646 |
+
"loss": 0.7614,
|
| 647 |
"step": 384
|
| 648 |
},
|
| 649 |
{
|
| 650 |
"epoch": 7.0,
|
| 651 |
+
"pls_score": 57.2,
|
| 652 |
+
"std": 4.3085032203771165,
|
| 653 |
"step": 385
|
| 654 |
},
|
| 655 |
{
|
| 656 |
"epoch": 7.0,
|
| 657 |
+
"eval_loss": 2.7865102291107178,
|
| 658 |
+
"eval_runtime": 4.9204,
|
| 659 |
+
"eval_samples_per_second": 4.674,
|
| 660 |
+
"eval_steps_per_second": 1.219,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 661 |
"step": 385
|
| 662 |
},
|
| 663 |
{
|
| 664 |
"epoch": 7.05,
|
| 665 |
"learning_rate": 1.2090063459025955e-05,
|
| 666 |
+
"loss": 0.8261,
|
| 667 |
"step": 388
|
| 668 |
},
|
| 669 |
{
|
| 670 |
"epoch": 7.13,
|
| 671 |
"learning_rate": 1.155078679555969e-05,
|
| 672 |
+
"loss": 0.639,
|
| 673 |
"step": 392
|
| 674 |
},
|
| 675 |
{
|
| 676 |
"epoch": 7.2,
|
| 677 |
"learning_rate": 1.1020177413231334e-05,
|
| 678 |
+
"loss": 0.748,
|
| 679 |
"step": 396
|
| 680 |
},
|
| 681 |
{
|
| 682 |
"epoch": 7.27,
|
| 683 |
"learning_rate": 1.049857726072005e-05,
|
| 684 |
+
"loss": 0.6049,
|
| 685 |
"step": 400
|
| 686 |
},
|
| 687 |
{
|
| 688 |
"epoch": 7.35,
|
| 689 |
"learning_rate": 9.986322480749927e-06,
|
| 690 |
+
"loss": 0.5971,
|
| 691 |
"step": 404
|
| 692 |
},
|
| 693 |
{
|
| 694 |
"epoch": 7.42,
|
| 695 |
"learning_rate": 9.483743193464408e-06,
|
| 696 |
+
"loss": 0.6648,
|
| 697 |
"step": 408
|
| 698 |
},
|
| 699 |
{
|
| 700 |
"epoch": 7.49,
|
| 701 |
"learning_rate": 8.991163283681944e-06,
|
| 702 |
+
"loss": 0.6513,
|
| 703 |
"step": 412
|
| 704 |
},
|
| 705 |
{
|
| 706 |
"epoch": 7.56,
|
| 707 |
"learning_rate": 8.508900192169964e-06,
|
| 708 |
+
"loss": 0.5917,
|
| 709 |
"step": 416
|
| 710 |
},
|
| 711 |
{
|
| 712 |
"epoch": 7.64,
|
| 713 |
"learning_rate": 8.0372647110717e-06,
|
| 714 |
+
"loss": 0.6839,
|
| 715 |
"step": 420
|
| 716 |
},
|
| 717 |
{
|
| 718 |
"epoch": 7.71,
|
| 719 |
"learning_rate": 7.576560783617668e-06,
|
| 720 |
+
"loss": 0.6658,
|
| 721 |
"step": 424
|
| 722 |
},
|
| 723 |
{
|
| 724 |
"epoch": 7.78,
|
| 725 |
"learning_rate": 7.127085308250914e-06,
|
| 726 |
+
"loss": 0.691,
|
| 727 |
"step": 428
|
| 728 |
},
|
| 729 |
{
|
| 730 |
"epoch": 7.85,
|
| 731 |
"learning_rate": 6.689127947292231e-06,
|
| 732 |
+
"loss": 0.7198,
|
| 733 |
"step": 432
|
| 734 |
},
|
| 735 |
{
|
| 736 |
"epoch": 7.93,
|
| 737 |
"learning_rate": 6.2629709402686535e-06,
|
| 738 |
+
"loss": 0.6533,
|
| 739 |
"step": 436
|
| 740 |
},
|
| 741 |
{
|
| 742 |
"epoch": 8.0,
|
| 743 |
"learning_rate": 5.848888922025553e-06,
|
| 744 |
+
"loss": 0.6237,
|
| 745 |
+
"step": 440
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 8.0,
|
| 749 |
+
"pls_score": 60.4,
|
| 750 |
+
"std": 4.507416111254875,
|
| 751 |
+
"step": 440
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 8.0,
|
| 755 |
+
"eval_loss": 2.958909511566162,
|
| 756 |
+
"eval_runtime": 4.9181,
|
| 757 |
+
"eval_samples_per_second": 4.677,
|
| 758 |
+
"eval_steps_per_second": 1.22,
|
| 759 |
"step": 440
|
| 760 |
},
|
| 761 |
{
|
| 762 |
"epoch": 8.07,
|
| 763 |
"learning_rate": 5.4471487457395225e-06,
|
| 764 |
+
"loss": 0.5912,
|
| 765 |
"step": 444
|
| 766 |
},
|
| 767 |
{
|
| 768 |
"epoch": 8.15,
|
| 769 |
"learning_rate": 5.058009310946119e-06,
|
| 770 |
+
"loss": 0.6374,
|
| 771 |
"step": 448
|
| 772 |
},
|
| 773 |
{
|
| 774 |
"epoch": 8.22,
|
| 775 |
"learning_rate": 4.681721396693303e-06,
|
| 776 |
+
"loss": 0.5975,
|
| 777 |
"step": 452
|
| 778 |
},
|
| 779 |
{
|
| 780 |
"epoch": 8.29,
|
| 781 |
"learning_rate": 4.318527499928074e-06,
|
| 782 |
+
"loss": 0.5448,
|
| 783 |
"step": 456
|
| 784 |
},
|
| 785 |
{
|
| 786 |
"epoch": 8.36,
|
| 787 |
"learning_rate": 3.968661679220468e-06,
|
| 788 |
+
"loss": 0.5574,
|
| 789 |
"step": 460
|
| 790 |
},
|
| 791 |
{
|
| 792 |
"epoch": 8.44,
|
| 793 |
"learning_rate": 3.632349403925664e-06,
|
| 794 |
+
"loss": 0.6206,
|
| 795 |
"step": 464
|
| 796 |
},
|
| 797 |
{
|
| 798 |
"epoch": 8.51,
|
| 799 |
"learning_rate": 3.3098074088812686e-06,
|
| 800 |
+
"loss": 0.6207,
|
| 801 |
"step": 468
|
| 802 |
},
|
| 803 |
{
|
| 804 |
"epoch": 8.58,
|
| 805 |
"learning_rate": 3.0012435547336737e-06,
|
| 806 |
+
"loss": 0.5411,
|
| 807 |
"step": 472
|
| 808 |
},
|
| 809 |
{
|
| 810 |
"epoch": 8.65,
|
| 811 |
"learning_rate": 2.7068566939831645e-06,
|
| 812 |
+
"loss": 0.6231,
|
| 813 |
"step": 476
|
| 814 |
},
|
| 815 |
{
|
| 816 |
"epoch": 8.73,
|
| 817 |
"learning_rate": 2.4268365428344736e-06,
|
| 818 |
+
"loss": 0.5944,
|
| 819 |
"step": 480
|
| 820 |
},
|
| 821 |
{
|
| 822 |
"epoch": 8.8,
|
| 823 |
"learning_rate": 2.1613635589349756e-06,
|
| 824 |
+
"loss": 0.6279,
|
| 825 |
"step": 484
|
| 826 |
},
|
| 827 |
{
|
| 828 |
"epoch": 8.87,
|
| 829 |
"learning_rate": 1.9106088250797267e-06,
|
| 830 |
+
"loss": 0.561,
|
| 831 |
"step": 488
|
| 832 |
},
|
| 833 |
{
|
| 834 |
"epoch": 8.95,
|
| 835 |
"learning_rate": 1.674733938957873e-06,
|
| 836 |
+
"loss": 0.5711,
|
| 837 |
"step": 492
|
| 838 |
},
|
| 839 |
+
{
|
| 840 |
+
"epoch": 9.0,
|
| 841 |
+
"pls_score": 55.2,
|
| 842 |
+
"std": 4.365684367885522,
|
| 843 |
+
"step": 495
|
| 844 |
+
},
|
| 845 |
+
{
|
| 846 |
+
"epoch": 9.0,
|
| 847 |
+
"eval_loss": 3.058058977127075,
|
| 848 |
+
"eval_runtime": 4.9251,
|
| 849 |
+
"eval_samples_per_second": 4.67,
|
| 850 |
+
"eval_steps_per_second": 1.218,
|
| 851 |
+
"step": 495
|
| 852 |
+
},
|
| 853 |
{
|
| 854 |
"epoch": 9.02,
|
| 855 |
"learning_rate": 1.4538909090118846e-06,
|
| 856 |
+
"loss": 0.5386,
|
| 857 |
"step": 496
|
| 858 |
},
|
| 859 |
{
|
| 860 |
"epoch": 9.09,
|
| 861 |
"learning_rate": 1.248222056476367e-06,
|
| 862 |
+
"loss": 0.6031,
|
| 863 |
"step": 500
|
| 864 |
},
|
| 865 |
{
|
| 866 |
"epoch": 9.16,
|
| 867 |
"learning_rate": 1.0578599236598707e-06,
|
| 868 |
+
"loss": 0.5676,
|
| 869 |
"step": 504
|
| 870 |
},
|
| 871 |
{
|
| 872 |
"epoch": 9.24,
|
| 873 |
"learning_rate": 8.829271885286094e-07,
|
| 874 |
+
"loss": 0.5298,
|
| 875 |
"step": 508
|
| 876 |
},
|
| 877 |
{
|
| 878 |
"epoch": 9.31,
|
| 879 |
"learning_rate": 7.235365856472442e-07,
|
| 880 |
+
"loss": 0.5428,
|
| 881 |
"step": 512
|
| 882 |
},
|
| 883 |
{
|
| 884 |
"epoch": 9.38,
|
| 885 |
"learning_rate": 5.797908335276214e-07,
|
| 886 |
+
"loss": 0.5811,
|
| 887 |
"step": 516
|
| 888 |
},
|
| 889 |
{
|
| 890 |
"epoch": 9.45,
|
| 891 |
"learning_rate": 4.517825684323324e-07,
|
| 892 |
+
"loss": 0.5703,
|
| 893 |
"step": 520
|
| 894 |
},
|
| 895 |
{
|
| 896 |
"epoch": 9.53,
|
| 897 |
"learning_rate": 3.395942846757066e-07,
|
| 898 |
+
"loss": 0.6147,
|
| 899 |
"step": 524
|
| 900 |
},
|
| 901 |
{
|
| 902 |
"epoch": 9.6,
|
| 903 |
"learning_rate": 2.4329828146074095e-07,
|
| 904 |
+
"loss": 0.5451,
|
| 905 |
"step": 528
|
| 906 |
},
|
| 907 |
{
|
| 908 |
"epoch": 9.67,
|
| 909 |
"learning_rate": 1.6295661628624447e-07,
|
| 910 |
+
"loss": 0.5303,
|
| 911 |
"step": 532
|
| 912 |
},
|
| 913 |
{
|
| 914 |
"epoch": 9.75,
|
| 915 |
"learning_rate": 9.862106495415469e-08,
|
| 916 |
+
"loss": 0.5629,
|
| 917 |
"step": 536
|
| 918 |
},
|
| 919 |
{
|
| 920 |
"epoch": 9.82,
|
| 921 |
"learning_rate": 5.033308820289184e-08,
|
| 922 |
+
"loss": 0.5472,
|
| 923 |
"step": 540
|
| 924 |
},
|
| 925 |
{
|
| 926 |
"epoch": 9.89,
|
| 927 |
"learning_rate": 1.812380498815991e-08,
|
| 928 |
+
"loss": 0.5334,
|
| 929 |
"step": 544
|
| 930 |
},
|
| 931 |
{
|
| 932 |
"epoch": 9.96,
|
| 933 |
"learning_rate": 2.0139724285161977e-09,
|
| 934 |
+
"loss": 0.5884,
|
| 935 |
"step": 548
|
| 936 |
},
|
| 937 |
+
{
|
| 938 |
+
"epoch": 10.0,
|
| 939 |
+
"pls_score": 60.0,
|
| 940 |
+
"std": 4.195235392680606,
|
| 941 |
+
"step": 550
|
| 942 |
+
},
|
| 943 |
+
{
|
| 944 |
+
"epoch": 10.0,
|
| 945 |
+
"eval_loss": 3.072557210922241,
|
| 946 |
+
"eval_runtime": 4.9196,
|
| 947 |
+
"eval_samples_per_second": 4.675,
|
| 948 |
+
"eval_steps_per_second": 1.22,
|
| 949 |
+
"step": 550
|
| 950 |
+
},
|
| 951 |
{
|
| 952 |
"epoch": 10.0,
|
| 953 |
"step": 550,
|
| 954 |
+
"total_flos": 1.870637420231885e+16,
|
| 955 |
+
"train_loss": 0.2617128190127286,
|
| 956 |
+
"train_runtime": 10832.3445,
|
| 957 |
+
"train_samples_per_second": 0.2,
|
| 958 |
+
"train_steps_per_second": 0.051
|
| 959 |
}
|
| 960 |
],
|
| 961 |
"logging_steps": 4,
|
|
|
|
| 963 |
"num_input_tokens_seen": 0,
|
| 964 |
"num_train_epochs": 10,
|
| 965 |
"save_steps": 55,
|
| 966 |
+
"total_flos": 1.870637420231885e+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 sha256:0bc819b0296221e62114b2956d623b867b1651f22681fcdf7256a558db58908a
|
| 3 |
size 4792
|