End of training
Browse files- README.md +3 -3
- all_results.json +11 -11
- eval_results.json +6 -6
- runs/Oct22_18-03-39_549dd11dc1c2/events.out.tfevents.1729635164.549dd11dc1c2.1065.2 +3 -0
- train_results.json +6 -6
- trainer_state.json +562 -1744
README.md
CHANGED
|
@@ -23,7 +23,7 @@ model-index:
|
|
| 23 |
metrics:
|
| 24 |
- name: Accuracy
|
| 25 |
type: accuracy
|
| 26 |
-
value: 0.
|
| 27 |
---
|
| 28 |
|
| 29 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
@@ -33,8 +33,8 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 33 |
|
| 34 |
This model is a fine-tuned version of [google/vit-base-patch16-224-in21k](https://huggingface.co/google/vit-base-patch16-224-in21k) on the imagefolder dataset.
|
| 35 |
It achieves the following results on the evaluation set:
|
| 36 |
-
- Loss: 0.
|
| 37 |
-
- Accuracy: 0.
|
| 38 |
|
| 39 |
## Model description
|
| 40 |
|
|
|
|
| 23 |
metrics:
|
| 24 |
- name: Accuracy
|
| 25 |
type: accuracy
|
| 26 |
+
value: 0.9338235294117647
|
| 27 |
---
|
| 28 |
|
| 29 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
|
|
| 33 |
|
| 34 |
This model is a fine-tuned version of [google/vit-base-patch16-224-in21k](https://huggingface.co/google/vit-base-patch16-224-in21k) on the imagefolder dataset.
|
| 35 |
It achieves the following results on the evaluation set:
|
| 36 |
+
- Loss: 0.2825
|
| 37 |
+
- Accuracy: 0.9338
|
| 38 |
|
| 39 |
## Model description
|
| 40 |
|
all_results.json
CHANGED
|
@@ -1,13 +1,13 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_accuracy": 0.
|
| 4 |
-
"eval_loss": 0.
|
| 5 |
-
"eval_runtime":
|
| 6 |
-
"eval_samples_per_second":
|
| 7 |
-
"eval_steps_per_second": 0.
|
| 8 |
-
"total_flos":
|
| 9 |
-
"train_loss": 0.
|
| 10 |
-
"train_runtime":
|
| 11 |
-
"train_samples_per_second":
|
| 12 |
-
"train_steps_per_second": 0.
|
| 13 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 46.15384615384615,
|
| 3 |
+
"eval_accuracy": 0.9338235294117647,
|
| 4 |
+
"eval_loss": 0.28249993920326233,
|
| 5 |
+
"eval_runtime": 30.22,
|
| 6 |
+
"eval_samples_per_second": 4.5,
|
| 7 |
+
"eval_steps_per_second": 0.165,
|
| 8 |
+
"total_flos": 4.3781443993328026e+18,
|
| 9 |
+
"train_loss": 0.4078153912226359,
|
| 10 |
+
"train_runtime": 14896.2203,
|
| 11 |
+
"train_samples_per_second": 4.108,
|
| 12 |
+
"train_steps_per_second": 0.03
|
| 13 |
}
|
eval_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_accuracy": 0.
|
| 4 |
-
"eval_loss": 0.
|
| 5 |
-
"eval_runtime":
|
| 6 |
-
"eval_samples_per_second":
|
| 7 |
-
"eval_steps_per_second": 0.
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 46.15384615384615,
|
| 3 |
+
"eval_accuracy": 0.9338235294117647,
|
| 4 |
+
"eval_loss": 0.28249993920326233,
|
| 5 |
+
"eval_runtime": 30.22,
|
| 6 |
+
"eval_samples_per_second": 4.5,
|
| 7 |
+
"eval_steps_per_second": 0.165
|
| 8 |
}
|
runs/Oct22_18-03-39_549dd11dc1c2/events.out.tfevents.1729635164.549dd11dc1c2.1065.2
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:89307b63cf5306a373f9652323bdbf974157f375f17284cd6596c0e85eda5fdb
|
| 3 |
+
size 411
|
train_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"total_flos":
|
| 4 |
-
"train_loss": 0.
|
| 5 |
-
"train_runtime":
|
| 6 |
-
"train_samples_per_second":
|
| 7 |
-
"train_steps_per_second": 0.
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 46.15384615384615,
|
| 3 |
+
"total_flos": 4.3781443993328026e+18,
|
| 4 |
+
"train_loss": 0.4078153912226359,
|
| 5 |
+
"train_runtime": 14896.2203,
|
| 6 |
+
"train_samples_per_second": 4.108,
|
| 7 |
+
"train_steps_per_second": 0.03
|
| 8 |
}
|
trainer_state.json
CHANGED
|
@@ -1,1945 +1,763 @@
|
|
| 1 |
{
|
| 2 |
-
"best_metric": 0.
|
| 3 |
-
"best_model_checkpoint": "vit-base-patch16-224-in21k-finetuned-papsmear/checkpoint-
|
| 4 |
-
"epoch":
|
| 5 |
"eval_steps": 500,
|
| 6 |
-
"global_step":
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
-
"epoch": 0.
|
| 13 |
-
"
|
| 14 |
-
"
|
| 15 |
-
"
|
| 16 |
-
"
|
| 17 |
-
|
| 18 |
-
|
| 19 |
-
"epoch": 0.47619047619047616,
|
| 20 |
-
"grad_norm": 1.8307807445526123,
|
| 21 |
-
"learning_rate": 4.7619047619047615e-06,
|
| 22 |
-
"loss": 1.3406,
|
| 23 |
-
"step": 20
|
| 24 |
-
},
|
| 25 |
-
{
|
| 26 |
-
"epoch": 0.7142857142857143,
|
| 27 |
-
"grad_norm": 1.8257180452346802,
|
| 28 |
-
"learning_rate": 7.142857142857143e-06,
|
| 29 |
-
"loss": 1.2533,
|
| 30 |
-
"step": 30
|
| 31 |
-
},
|
| 32 |
-
{
|
| 33 |
-
"epoch": 0.9523809523809523,
|
| 34 |
-
"grad_norm": 1.2919795513153076,
|
| 35 |
-
"learning_rate": 9.523809523809523e-06,
|
| 36 |
-
"loss": 1.1553,
|
| 37 |
-
"step": 40
|
| 38 |
-
},
|
| 39 |
-
{
|
| 40 |
-
"epoch": 1.0,
|
| 41 |
-
"eval_accuracy": 0.5477386934673367,
|
| 42 |
-
"eval_loss": 1.0950442552566528,
|
| 43 |
-
"eval_runtime": 104.5,
|
| 44 |
-
"eval_samples_per_second": 5.713,
|
| 45 |
-
"eval_steps_per_second": 0.182,
|
| 46 |
-
"step": 42
|
| 47 |
-
},
|
| 48 |
-
{
|
| 49 |
-
"epoch": 1.1904761904761905,
|
| 50 |
-
"grad_norm": 0.9992073178291321,
|
| 51 |
-
"learning_rate": 1.1904761904761905e-05,
|
| 52 |
-
"loss": 1.0508,
|
| 53 |
-
"step": 50
|
| 54 |
},
|
| 55 |
{
|
| 56 |
-
"epoch": 1.
|
| 57 |
-
"grad_norm":
|
| 58 |
-
"learning_rate": 1.
|
| 59 |
-
"loss":
|
| 60 |
-
"step":
|
| 61 |
-
},
|
| 62 |
-
{
|
| 63 |
-
"epoch": 1.6666666666666665,
|
| 64 |
-
"grad_norm": 1.0831001996994019,
|
| 65 |
-
"learning_rate": 1.6666666666666667e-05,
|
| 66 |
-
"loss": 0.8539,
|
| 67 |
-
"step": 70
|
| 68 |
-
},
|
| 69 |
-
{
|
| 70 |
-
"epoch": 1.9047619047619047,
|
| 71 |
-
"grad_norm": 1.271881103515625,
|
| 72 |
-
"learning_rate": 1.9047619047619046e-05,
|
| 73 |
-
"loss": 0.7791,
|
| 74 |
-
"step": 80
|
| 75 |
-
},
|
| 76 |
-
{
|
| 77 |
-
"epoch": 2.0,
|
| 78 |
-
"eval_accuracy": 0.8525963149078727,
|
| 79 |
-
"eval_loss": 0.6485655307769775,
|
| 80 |
-
"eval_runtime": 107.1905,
|
| 81 |
-
"eval_samples_per_second": 5.57,
|
| 82 |
-
"eval_steps_per_second": 0.177,
|
| 83 |
-
"step": 84
|
| 84 |
-
},
|
| 85 |
-
{
|
| 86 |
-
"epoch": 2.142857142857143,
|
| 87 |
-
"grad_norm": 1.7645951509475708,
|
| 88 |
-
"learning_rate": 2.1428571428571428e-05,
|
| 89 |
-
"loss": 0.6112,
|
| 90 |
-
"step": 90
|
| 91 |
-
},
|
| 92 |
-
{
|
| 93 |
-
"epoch": 2.380952380952381,
|
| 94 |
-
"grad_norm": 2.271763563156128,
|
| 95 |
-
"learning_rate": 2.380952380952381e-05,
|
| 96 |
-
"loss": 0.5029,
|
| 97 |
-
"step": 100
|
| 98 |
-
},
|
| 99 |
-
{
|
| 100 |
-
"epoch": 2.619047619047619,
|
| 101 |
-
"grad_norm": 2.248612403869629,
|
| 102 |
-
"learning_rate": 2.6190476190476192e-05,
|
| 103 |
-
"loss": 0.4809,
|
| 104 |
-
"step": 110
|
| 105 |
},
|
| 106 |
{
|
| 107 |
-
"epoch":
|
| 108 |
-
"
|
| 109 |
-
"
|
| 110 |
-
"
|
| 111 |
-
"
|
|
|
|
|
|
|
| 112 |
},
|
| 113 |
{
|
| 114 |
-
"epoch":
|
| 115 |
-
"
|
| 116 |
-
"
|
| 117 |
-
"
|
| 118 |
-
"
|
| 119 |
-
"eval_steps_per_second": 0.177,
|
| 120 |
-
"step": 126
|
| 121 |
},
|
| 122 |
{
|
| 123 |
-
"epoch":
|
| 124 |
-
"
|
| 125 |
-
"
|
| 126 |
-
"
|
| 127 |
-
"
|
|
|
|
|
|
|
| 128 |
},
|
| 129 |
{
|
| 130 |
-
"epoch": 3.
|
| 131 |
-
"grad_norm": 1.
|
| 132 |
"learning_rate": 3.3333333333333335e-05,
|
| 133 |
-
"loss":
|
| 134 |
-
"step":
|
| 135 |
-
},
|
| 136 |
-
{
|
| 137 |
-
"epoch": 3.571428571428571,
|
| 138 |
-
"grad_norm": 7.162674903869629,
|
| 139 |
-
"learning_rate": 3.571428571428572e-05,
|
| 140 |
-
"loss": 0.3719,
|
| 141 |
-
"step": 150
|
| 142 |
-
},
|
| 143 |
-
{
|
| 144 |
-
"epoch": 3.8095238095238093,
|
| 145 |
-
"grad_norm": 2.170807123184204,
|
| 146 |
-
"learning_rate": 3.809523809523809e-05,
|
| 147 |
-
"loss": 0.3495,
|
| 148 |
-
"step": 160
|
| 149 |
},
|
| 150 |
{
|
| 151 |
"epoch": 4.0,
|
| 152 |
-
"eval_accuracy": 0.
|
| 153 |
-
"eval_loss":
|
| 154 |
-
"eval_runtime":
|
| 155 |
-
"eval_samples_per_second":
|
| 156 |
-
"eval_steps_per_second": 0.
|
| 157 |
-
"step":
|
| 158 |
-
},
|
| 159 |
-
{
|
| 160 |
-
"epoch": 4.0476190476190474,
|
| 161 |
-
"grad_norm": 1.2484022378921509,
|
| 162 |
-
"learning_rate": 4.047619047619048e-05,
|
| 163 |
-
"loss": 0.2892,
|
| 164 |
-
"step": 170
|
| 165 |
-
},
|
| 166 |
-
{
|
| 167 |
-
"epoch": 4.285714285714286,
|
| 168 |
-
"grad_norm": 7.085641860961914,
|
| 169 |
-
"learning_rate": 4.2857142857142856e-05,
|
| 170 |
-
"loss": 0.3386,
|
| 171 |
-
"step": 180
|
| 172 |
-
},
|
| 173 |
-
{
|
| 174 |
-
"epoch": 4.523809523809524,
|
| 175 |
-
"grad_norm": 1.4148013591766357,
|
| 176 |
-
"learning_rate": 4.523809523809524e-05,
|
| 177 |
-
"loss": 0.3225,
|
| 178 |
-
"step": 190
|
| 179 |
-
},
|
| 180 |
-
{
|
| 181 |
-
"epoch": 4.761904761904762,
|
| 182 |
-
"grad_norm": 2.7628612518310547,
|
| 183 |
-
"learning_rate": 4.761904761904762e-05,
|
| 184 |
-
"loss": 0.2979,
|
| 185 |
-
"step": 200
|
| 186 |
-
},
|
| 187 |
-
{
|
| 188 |
-
"epoch": 5.0,
|
| 189 |
-
"grad_norm": 1.5106964111328125,
|
| 190 |
-
"learning_rate": 5e-05,
|
| 191 |
-
"loss": 0.2556,
|
| 192 |
-
"step": 210
|
| 193 |
-
},
|
| 194 |
-
{
|
| 195 |
-
"epoch": 5.0,
|
| 196 |
-
"eval_accuracy": 0.9279731993299832,
|
| 197 |
-
"eval_loss": 0.27220866084098816,
|
| 198 |
-
"eval_runtime": 106.2256,
|
| 199 |
-
"eval_samples_per_second": 5.62,
|
| 200 |
-
"eval_steps_per_second": 0.179,
|
| 201 |
-
"step": 210
|
| 202 |
-
},
|
| 203 |
-
{
|
| 204 |
-
"epoch": 5.238095238095238,
|
| 205 |
-
"grad_norm": 4.107568740844727,
|
| 206 |
-
"learning_rate": 4.973544973544973e-05,
|
| 207 |
-
"loss": 0.2635,
|
| 208 |
-
"step": 220
|
| 209 |
},
|
| 210 |
{
|
| 211 |
-
"epoch":
|
| 212 |
-
"grad_norm":
|
| 213 |
-
"learning_rate": 4.
|
| 214 |
-
"loss":
|
| 215 |
-
"step":
|
| 216 |
-
},
|
| 217 |
-
{
|
| 218 |
-
"epoch": 5.714285714285714,
|
| 219 |
-
"grad_norm": 2.4218671321868896,
|
| 220 |
-
"learning_rate": 4.9206349206349204e-05,
|
| 221 |
-
"loss": 0.2303,
|
| 222 |
-
"step": 240
|
| 223 |
-
},
|
| 224 |
-
{
|
| 225 |
-
"epoch": 5.9523809523809526,
|
| 226 |
-
"grad_norm": 2.0036144256591797,
|
| 227 |
-
"learning_rate": 4.894179894179895e-05,
|
| 228 |
-
"loss": 0.2791,
|
| 229 |
-
"step": 250
|
| 230 |
},
|
| 231 |
{
|
| 232 |
-
"epoch":
|
| 233 |
-
"eval_accuracy": 0.
|
| 234 |
-
"eval_loss":
|
| 235 |
-
"eval_runtime":
|
| 236 |
-
"eval_samples_per_second":
|
| 237 |
-
"eval_steps_per_second": 0.
|
| 238 |
-
"step":
|
| 239 |
},
|
| 240 |
{
|
| 241 |
-
"epoch":
|
| 242 |
-
"grad_norm":
|
| 243 |
-
"learning_rate": 4.
|
| 244 |
-
"loss":
|
| 245 |
-
"step":
|
| 246 |
},
|
| 247 |
{
|
| 248 |
-
"epoch":
|
| 249 |
-
"
|
| 250 |
-
"
|
| 251 |
-
"
|
| 252 |
-
"
|
|
|
|
|
|
|
| 253 |
},
|
| 254 |
{
|
| 255 |
-
"epoch": 6.
|
| 256 |
-
"grad_norm": 1.
|
| 257 |
"learning_rate": 4.814814814814815e-05,
|
| 258 |
-
"loss": 0.
|
| 259 |
-
"step":
|
| 260 |
-
},
|
| 261 |
-
{
|
| 262 |
-
"epoch": 6.904761904761905,
|
| 263 |
-
"grad_norm": 2.22580885887146,
|
| 264 |
-
"learning_rate": 4.7883597883597884e-05,
|
| 265 |
-
"loss": 0.2343,
|
| 266 |
-
"step": 290
|
| 267 |
-
},
|
| 268 |
-
{
|
| 269 |
-
"epoch": 7.0,
|
| 270 |
-
"eval_accuracy": 0.9380234505862647,
|
| 271 |
-
"eval_loss": 0.2376711070537567,
|
| 272 |
-
"eval_runtime": 104.2609,
|
| 273 |
-
"eval_samples_per_second": 5.726,
|
| 274 |
-
"eval_steps_per_second": 0.182,
|
| 275 |
-
"step": 294
|
| 276 |
-
},
|
| 277 |
-
{
|
| 278 |
-
"epoch": 7.142857142857143,
|
| 279 |
-
"grad_norm": 1.8247556686401367,
|
| 280 |
-
"learning_rate": 4.761904761904762e-05,
|
| 281 |
-
"loss": 0.2295,
|
| 282 |
-
"step": 300
|
| 283 |
-
},
|
| 284 |
-
{
|
| 285 |
-
"epoch": 7.380952380952381,
|
| 286 |
-
"grad_norm": 2.389512538909912,
|
| 287 |
-
"learning_rate": 4.7354497354497356e-05,
|
| 288 |
-
"loss": 0.2059,
|
| 289 |
-
"step": 310
|
| 290 |
},
|
| 291 |
{
|
| 292 |
-
"epoch":
|
| 293 |
-
"
|
| 294 |
-
"
|
| 295 |
-
"
|
| 296 |
-
"
|
|
|
|
|
|
|
| 297 |
},
|
| 298 |
{
|
| 299 |
-
"epoch": 7.
|
| 300 |
-
"grad_norm": 1.
|
| 301 |
-
"learning_rate": 4.
|
| 302 |
-
"loss": 0.
|
| 303 |
-
"step":
|
| 304 |
},
|
| 305 |
{
|
| 306 |
"epoch": 8.0,
|
| 307 |
-
"eval_accuracy": 0.
|
| 308 |
-
"eval_loss": 0.
|
| 309 |
-
"eval_runtime":
|
| 310 |
-
"eval_samples_per_second":
|
| 311 |
-
"eval_steps_per_second": 0.
|
| 312 |
-
"step":
|
| 313 |
-
},
|
| 314 |
-
{
|
| 315 |
-
"epoch": 8.
|
| 316 |
-
"grad_norm": 1.
|
| 317 |
-
"learning_rate": 4.
|
| 318 |
-
"loss": 0.
|
| 319 |
-
"step":
|
| 320 |
-
},
|
| 321 |
-
{
|
| 322 |
-
"epoch": 8.333333333333334,
|
| 323 |
-
"grad_norm": 1.4709361791610718,
|
| 324 |
-
"learning_rate": 4.62962962962963e-05,
|
| 325 |
-
"loss": 0.1809,
|
| 326 |
-
"step": 350
|
| 327 |
-
},
|
| 328 |
-
{
|
| 329 |
-
"epoch": 8.571428571428571,
|
| 330 |
-
"grad_norm": 1.731719732284546,
|
| 331 |
-
"learning_rate": 4.603174603174603e-05,
|
| 332 |
-
"loss": 0.2163,
|
| 333 |
-
"step": 360
|
| 334 |
-
},
|
| 335 |
-
{
|
| 336 |
-
"epoch": 8.80952380952381,
|
| 337 |
-
"grad_norm": 1.8033312559127808,
|
| 338 |
-
"learning_rate": 4.576719576719577e-05,
|
| 339 |
-
"loss": 0.1984,
|
| 340 |
-
"step": 370
|
| 341 |
-
},
|
| 342 |
-
{
|
| 343 |
-
"epoch": 9.0,
|
| 344 |
-
"eval_accuracy": 0.9346733668341709,
|
| 345 |
-
"eval_loss": 0.22224725782871246,
|
| 346 |
-
"eval_runtime": 104.8934,
|
| 347 |
-
"eval_samples_per_second": 5.691,
|
| 348 |
-
"eval_steps_per_second": 0.181,
|
| 349 |
-
"step": 378
|
| 350 |
-
},
|
| 351 |
-
{
|
| 352 |
-
"epoch": 9.047619047619047,
|
| 353 |
-
"grad_norm": 1.9741649627685547,
|
| 354 |
-
"learning_rate": 4.55026455026455e-05,
|
| 355 |
-
"loss": 0.2093,
|
| 356 |
-
"step": 380
|
| 357 |
-
},
|
| 358 |
-
{
|
| 359 |
-
"epoch": 9.285714285714286,
|
| 360 |
-
"grad_norm": 1.6360487937927246,
|
| 361 |
-
"learning_rate": 4.523809523809524e-05,
|
| 362 |
-
"loss": 0.1586,
|
| 363 |
-
"step": 390
|
| 364 |
-
},
|
| 365 |
-
{
|
| 366 |
-
"epoch": 9.523809523809524,
|
| 367 |
-
"grad_norm": 2.772472858428955,
|
| 368 |
-
"learning_rate": 4.4973544973544974e-05,
|
| 369 |
-
"loss": 0.1423,
|
| 370 |
-
"step": 400
|
| 371 |
},
|
| 372 |
{
|
| 373 |
-
"epoch":
|
| 374 |
-
"
|
| 375 |
-
"
|
| 376 |
-
"
|
| 377 |
-
"
|
|
|
|
|
|
|
| 378 |
},
|
| 379 |
{
|
| 380 |
-
"epoch":
|
| 381 |
-
"grad_norm":
|
| 382 |
"learning_rate": 4.4444444444444447e-05,
|
| 383 |
-
"loss": 0.
|
| 384 |
-
"step":
|
| 385 |
-
},
|
| 386 |
-
{
|
| 387 |
-
"epoch": 10.0,
|
| 388 |
-
"eval_accuracy": 0.9514237855946399,
|
| 389 |
-
"eval_loss": 0.19929908215999603,
|
| 390 |
-
"eval_runtime": 106.1844,
|
| 391 |
-
"eval_samples_per_second": 5.622,
|
| 392 |
-
"eval_steps_per_second": 0.179,
|
| 393 |
-
"step": 420
|
| 394 |
-
},
|
| 395 |
-
{
|
| 396 |
-
"epoch": 10.238095238095237,
|
| 397 |
-
"grad_norm": 2.9462811946868896,
|
| 398 |
-
"learning_rate": 4.417989417989418e-05,
|
| 399 |
-
"loss": 0.1305,
|
| 400 |
-
"step": 430
|
| 401 |
-
},
|
| 402 |
-
{
|
| 403 |
-
"epoch": 10.476190476190476,
|
| 404 |
-
"grad_norm": 2.5522372722625732,
|
| 405 |
-
"learning_rate": 4.391534391534391e-05,
|
| 406 |
-
"loss": 0.1555,
|
| 407 |
-
"step": 440
|
| 408 |
-
},
|
| 409 |
-
{
|
| 410 |
-
"epoch": 10.714285714285714,
|
| 411 |
-
"grad_norm": 1.2257236242294312,
|
| 412 |
-
"learning_rate": 4.3650793650793655e-05,
|
| 413 |
-
"loss": 0.1575,
|
| 414 |
-
"step": 450
|
| 415 |
-
},
|
| 416 |
-
{
|
| 417 |
-
"epoch": 10.952380952380953,
|
| 418 |
-
"grad_norm": 0.9242783188819885,
|
| 419 |
-
"learning_rate": 4.3386243386243384e-05,
|
| 420 |
-
"loss": 0.1529,
|
| 421 |
-
"step": 460
|
| 422 |
-
},
|
| 423 |
-
{
|
| 424 |
-
"epoch": 11.0,
|
| 425 |
-
"eval_accuracy": 0.9430485762144054,
|
| 426 |
-
"eval_loss": 0.2100822627544403,
|
| 427 |
-
"eval_runtime": 102.6058,
|
| 428 |
-
"eval_samples_per_second": 5.818,
|
| 429 |
-
"eval_steps_per_second": 0.185,
|
| 430 |
-
"step": 462
|
| 431 |
},
|
| 432 |
{
|
| 433 |
-
"epoch":
|
| 434 |
-
"
|
| 435 |
-
"
|
| 436 |
-
"
|
| 437 |
-
"
|
|
|
|
|
|
|
| 438 |
},
|
| 439 |
{
|
| 440 |
-
"epoch":
|
| 441 |
-
"grad_norm": 1.
|
| 442 |
-
"learning_rate": 4.
|
| 443 |
-
"loss": 0.
|
| 444 |
-
"step":
|
| 445 |
},
|
| 446 |
{
|
| 447 |
-
"epoch":
|
| 448 |
-
"
|
| 449 |
-
"
|
| 450 |
-
"
|
| 451 |
-
"
|
|
|
|
|
|
|
| 452 |
},
|
| 453 |
{
|
| 454 |
-
"epoch": 11.
|
| 455 |
-
"grad_norm":
|
| 456 |
-
"learning_rate": 4.
|
| 457 |
-
"loss": 0.
|
| 458 |
-
"step":
|
| 459 |
},
|
| 460 |
{
|
| 461 |
"epoch": 12.0,
|
| 462 |
-
"eval_accuracy": 0.
|
| 463 |
-
"eval_loss": 0.
|
| 464 |
-
"eval_runtime":
|
| 465 |
-
"eval_samples_per_second":
|
| 466 |
-
"eval_steps_per_second": 0.
|
| 467 |
-
"step":
|
| 468 |
-
},
|
| 469 |
-
{
|
| 470 |
-
"epoch": 12.142857142857142,
|
| 471 |
-
"grad_norm": 1.015551209449768,
|
| 472 |
-
"learning_rate": 4.2063492063492065e-05,
|
| 473 |
-
"loss": 0.1374,
|
| 474 |
-
"step": 510
|
| 475 |
-
},
|
| 476 |
-
{
|
| 477 |
-
"epoch": 12.380952380952381,
|
| 478 |
-
"grad_norm": 1.5744130611419678,
|
| 479 |
-
"learning_rate": 4.17989417989418e-05,
|
| 480 |
-
"loss": 0.1219,
|
| 481 |
-
"step": 520
|
| 482 |
-
},
|
| 483 |
-
{
|
| 484 |
-
"epoch": 12.619047619047619,
|
| 485 |
-
"grad_norm": 2.7706940174102783,
|
| 486 |
-
"learning_rate": 4.153439153439154e-05,
|
| 487 |
-
"loss": 0.1159,
|
| 488 |
-
"step": 530
|
| 489 |
-
},
|
| 490 |
-
{
|
| 491 |
-
"epoch": 12.857142857142858,
|
| 492 |
-
"grad_norm": 0.9542893767356873,
|
| 493 |
-
"learning_rate": 4.126984126984127e-05,
|
| 494 |
-
"loss": 0.1404,
|
| 495 |
-
"step": 540
|
| 496 |
-
},
|
| 497 |
-
{
|
| 498 |
-
"epoch": 13.0,
|
| 499 |
-
"eval_accuracy": 0.9396984924623115,
|
| 500 |
-
"eval_loss": 0.2028820812702179,
|
| 501 |
-
"eval_runtime": 106.2276,
|
| 502 |
-
"eval_samples_per_second": 5.62,
|
| 503 |
-
"eval_steps_per_second": 0.179,
|
| 504 |
-
"step": 546
|
| 505 |
},
|
| 506 |
{
|
| 507 |
-
"epoch":
|
| 508 |
-
"grad_norm":
|
| 509 |
-
"learning_rate": 4.100529100529101e-05,
|
| 510 |
-
"loss": 0.1128,
|
| 511 |
-
"step": 550
|
| 512 |
-
},
|
| 513 |
-
{
|
| 514 |
-
"epoch": 13.333333333333334,
|
| 515 |
-
"grad_norm": 0.4872543513774872,
|
| 516 |
"learning_rate": 4.074074074074074e-05,
|
| 517 |
-
"loss": 0.
|
| 518 |
-
"step":
|
| 519 |
-
},
|
| 520 |
-
{
|
| 521 |
-
"epoch": 13.571428571428571,
|
| 522 |
-
"grad_norm": 1.1375516653060913,
|
| 523 |
-
"learning_rate": 4.047619047619048e-05,
|
| 524 |
-
"loss": 0.1195,
|
| 525 |
-
"step": 570
|
| 526 |
-
},
|
| 527 |
-
{
|
| 528 |
-
"epoch": 13.80952380952381,
|
| 529 |
-
"grad_norm": 2.1001851558685303,
|
| 530 |
-
"learning_rate": 4.021164021164021e-05,
|
| 531 |
-
"loss": 0.1078,
|
| 532 |
-
"step": 580
|
| 533 |
-
},
|
| 534 |
-
{
|
| 535 |
-
"epoch": 14.0,
|
| 536 |
-
"eval_accuracy": 0.9413735343383585,
|
| 537 |
-
"eval_loss": 0.20870448648929596,
|
| 538 |
-
"eval_runtime": 103.1036,
|
| 539 |
-
"eval_samples_per_second": 5.79,
|
| 540 |
-
"eval_steps_per_second": 0.184,
|
| 541 |
-
"step": 588
|
| 542 |
-
},
|
| 543 |
-
{
|
| 544 |
-
"epoch": 14.047619047619047,
|
| 545 |
-
"grad_norm": 1.273701786994934,
|
| 546 |
-
"learning_rate": 3.9947089947089946e-05,
|
| 547 |
-
"loss": 0.1078,
|
| 548 |
-
"step": 590
|
| 549 |
-
},
|
| 550 |
-
{
|
| 551 |
-
"epoch": 14.285714285714286,
|
| 552 |
-
"grad_norm": 2.2141387462615967,
|
| 553 |
-
"learning_rate": 3.968253968253968e-05,
|
| 554 |
-
"loss": 0.1005,
|
| 555 |
-
"step": 600
|
| 556 |
-
},
|
| 557 |
-
{
|
| 558 |
-
"epoch": 14.523809523809524,
|
| 559 |
-
"grad_norm": 2.147643566131592,
|
| 560 |
-
"learning_rate": 3.941798941798942e-05,
|
| 561 |
-
"loss": 0.1195,
|
| 562 |
-
"step": 610
|
| 563 |
-
},
|
| 564 |
-
{
|
| 565 |
-
"epoch": 14.761904761904763,
|
| 566 |
-
"grad_norm": 1.8408890962600708,
|
| 567 |
-
"learning_rate": 3.9153439153439155e-05,
|
| 568 |
-
"loss": 0.0981,
|
| 569 |
-
"step": 620
|
| 570 |
-
},
|
| 571 |
-
{
|
| 572 |
-
"epoch": 15.0,
|
| 573 |
-
"grad_norm": 2.7218682765960693,
|
| 574 |
-
"learning_rate": 3.888888888888889e-05,
|
| 575 |
-
"loss": 0.1109,
|
| 576 |
-
"step": 630
|
| 577 |
-
},
|
| 578 |
-
{
|
| 579 |
-
"epoch": 15.0,
|
| 580 |
-
"eval_accuracy": 0.9614740368509213,
|
| 581 |
-
"eval_loss": 0.1381397545337677,
|
| 582 |
-
"eval_runtime": 102.743,
|
| 583 |
-
"eval_samples_per_second": 5.811,
|
| 584 |
-
"eval_steps_per_second": 0.185,
|
| 585 |
-
"step": 630
|
| 586 |
-
},
|
| 587 |
-
{
|
| 588 |
-
"epoch": 15.238095238095237,
|
| 589 |
-
"grad_norm": 2.087874412536621,
|
| 590 |
-
"learning_rate": 3.862433862433863e-05,
|
| 591 |
-
"loss": 0.0932,
|
| 592 |
-
"step": 640
|
| 593 |
-
},
|
| 594 |
-
{
|
| 595 |
-
"epoch": 15.476190476190476,
|
| 596 |
-
"grad_norm": 1.6582282781600952,
|
| 597 |
-
"learning_rate": 3.835978835978836e-05,
|
| 598 |
-
"loss": 0.12,
|
| 599 |
-
"step": 650
|
| 600 |
},
|
| 601 |
{
|
| 602 |
-
"epoch":
|
| 603 |
-
"
|
| 604 |
-
"
|
| 605 |
-
"
|
| 606 |
-
"
|
|
|
|
|
|
|
| 607 |
},
|
| 608 |
{
|
| 609 |
-
"epoch":
|
| 610 |
-
"grad_norm": 1.
|
| 611 |
-
"learning_rate": 3.
|
| 612 |
-
"loss": 0.
|
| 613 |
-
"step":
|
| 614 |
},
|
| 615 |
{
|
| 616 |
-
"epoch":
|
| 617 |
-
"eval_accuracy": 0.
|
| 618 |
-
"eval_loss": 0.
|
| 619 |
-
"eval_runtime":
|
| 620 |
-
"eval_samples_per_second":
|
| 621 |
-
"eval_steps_per_second": 0.
|
| 622 |
-
"step":
|
| 623 |
},
|
| 624 |
{
|
| 625 |
-
"epoch":
|
| 626 |
-
"grad_norm":
|
| 627 |
-
"learning_rate": 3.
|
| 628 |
-
"loss": 0.
|
| 629 |
-
"step":
|
| 630 |
},
|
| 631 |
{
|
| 632 |
-
"epoch":
|
| 633 |
-
"
|
| 634 |
-
"
|
| 635 |
-
"
|
| 636 |
-
"
|
|
|
|
|
|
|
| 637 |
},
|
| 638 |
{
|
| 639 |
-
"epoch":
|
| 640 |
-
"grad_norm":
|
| 641 |
"learning_rate": 3.7037037037037037e-05,
|
| 642 |
-
"loss": 0.
|
| 643 |
-
"step":
|
| 644 |
-
},
|
| 645 |
-
{
|
| 646 |
-
"epoch": 16.904761904761905,
|
| 647 |
-
"grad_norm": 0.7912140488624573,
|
| 648 |
-
"learning_rate": 3.677248677248677e-05,
|
| 649 |
-
"loss": 0.0949,
|
| 650 |
-
"step": 710
|
| 651 |
-
},
|
| 652 |
-
{
|
| 653 |
-
"epoch": 17.0,
|
| 654 |
-
"eval_accuracy": 0.9396984924623115,
|
| 655 |
-
"eval_loss": 0.19812369346618652,
|
| 656 |
-
"eval_runtime": 103.2395,
|
| 657 |
-
"eval_samples_per_second": 5.783,
|
| 658 |
-
"eval_steps_per_second": 0.184,
|
| 659 |
-
"step": 714
|
| 660 |
-
},
|
| 661 |
-
{
|
| 662 |
-
"epoch": 17.142857142857142,
|
| 663 |
-
"grad_norm": 2.5102455615997314,
|
| 664 |
-
"learning_rate": 3.650793650793651e-05,
|
| 665 |
-
"loss": 0.0942,
|
| 666 |
-
"step": 720
|
| 667 |
-
},
|
| 668 |
-
{
|
| 669 |
-
"epoch": 17.38095238095238,
|
| 670 |
-
"grad_norm": 0.907739520072937,
|
| 671 |
-
"learning_rate": 3.6243386243386245e-05,
|
| 672 |
-
"loss": 0.081,
|
| 673 |
-
"step": 730
|
| 674 |
-
},
|
| 675 |
-
{
|
| 676 |
-
"epoch": 17.61904761904762,
|
| 677 |
-
"grad_norm": 0.9580628275871277,
|
| 678 |
-
"learning_rate": 3.597883597883598e-05,
|
| 679 |
-
"loss": 0.0939,
|
| 680 |
-
"step": 740
|
| 681 |
-
},
|
| 682 |
-
{
|
| 683 |
-
"epoch": 17.857142857142858,
|
| 684 |
-
"grad_norm": 0.4892265498638153,
|
| 685 |
-
"learning_rate": 3.571428571428572e-05,
|
| 686 |
-
"loss": 0.0908,
|
| 687 |
-
"step": 750
|
| 688 |
-
},
|
| 689 |
-
{
|
| 690 |
-
"epoch": 18.0,
|
| 691 |
-
"eval_accuracy": 0.9581239530988275,
|
| 692 |
-
"eval_loss": 0.16083765029907227,
|
| 693 |
-
"eval_runtime": 103.0208,
|
| 694 |
-
"eval_samples_per_second": 5.795,
|
| 695 |
-
"eval_steps_per_second": 0.184,
|
| 696 |
-
"step": 756
|
| 697 |
-
},
|
| 698 |
-
{
|
| 699 |
-
"epoch": 18.095238095238095,
|
| 700 |
-
"grad_norm": 0.6354324221611023,
|
| 701 |
-
"learning_rate": 3.5449735449735446e-05,
|
| 702 |
-
"loss": 0.0674,
|
| 703 |
-
"step": 760
|
| 704 |
-
},
|
| 705 |
-
{
|
| 706 |
-
"epoch": 18.333333333333332,
|
| 707 |
-
"grad_norm": 1.7310459613800049,
|
| 708 |
-
"learning_rate": 3.518518518518519e-05,
|
| 709 |
-
"loss": 0.0963,
|
| 710 |
-
"step": 770
|
| 711 |
-
},
|
| 712 |
-
{
|
| 713 |
-
"epoch": 18.571428571428573,
|
| 714 |
-
"grad_norm": 1.0339725017547607,
|
| 715 |
-
"learning_rate": 3.492063492063492e-05,
|
| 716 |
-
"loss": 0.0846,
|
| 717 |
-
"step": 780
|
| 718 |
-
},
|
| 719 |
-
{
|
| 720 |
-
"epoch": 18.80952380952381,
|
| 721 |
-
"grad_norm": 2.478813648223877,
|
| 722 |
-
"learning_rate": 3.465608465608466e-05,
|
| 723 |
-
"loss": 0.0809,
|
| 724 |
-
"step": 790
|
| 725 |
-
},
|
| 726 |
-
{
|
| 727 |
-
"epoch": 19.0,
|
| 728 |
-
"eval_accuracy": 0.9581239530988275,
|
| 729 |
-
"eval_loss": 0.1764398217201233,
|
| 730 |
-
"eval_runtime": 104.4696,
|
| 731 |
-
"eval_samples_per_second": 5.715,
|
| 732 |
-
"eval_steps_per_second": 0.182,
|
| 733 |
-
"step": 798
|
| 734 |
},
|
| 735 |
{
|
| 736 |
-
"epoch":
|
| 737 |
-
"
|
| 738 |
-
"
|
| 739 |
-
"
|
| 740 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 741 |
},
|
| 742 |
{
|
| 743 |
-
"epoch":
|
| 744 |
-
"
|
| 745 |
-
"
|
| 746 |
-
"
|
| 747 |
-
"
|
|
|
|
|
|
|
| 748 |
},
|
| 749 |
{
|
| 750 |
-
"epoch":
|
| 751 |
-
"grad_norm":
|
| 752 |
-
"learning_rate": 3.
|
| 753 |
-
"loss": 0.
|
| 754 |
-
"step":
|
| 755 |
},
|
| 756 |
{
|
| 757 |
-
"epoch":
|
| 758 |
-
"
|
| 759 |
-
"
|
| 760 |
-
"
|
| 761 |
-
"
|
|
|
|
|
|
|
| 762 |
},
|
| 763 |
{
|
| 764 |
-
"epoch":
|
| 765 |
-
"grad_norm":
|
| 766 |
"learning_rate": 3.3333333333333335e-05,
|
| 767 |
-
"loss": 0.
|
| 768 |
-
"step":
|
| 769 |
-
},
|
| 770 |
-
{
|
| 771 |
-
"epoch": 20.0,
|
| 772 |
-
"eval_accuracy": 0.9530988274706867,
|
| 773 |
-
"eval_loss": 0.15123647451400757,
|
| 774 |
-
"eval_runtime": 102.2975,
|
| 775 |
-
"eval_samples_per_second": 5.836,
|
| 776 |
-
"eval_steps_per_second": 0.186,
|
| 777 |
-
"step": 840
|
| 778 |
-
},
|
| 779 |
-
{
|
| 780 |
-
"epoch": 20.238095238095237,
|
| 781 |
-
"grad_norm": 4.499787330627441,
|
| 782 |
-
"learning_rate": 3.306878306878307e-05,
|
| 783 |
-
"loss": 0.0936,
|
| 784 |
-
"step": 850
|
| 785 |
-
},
|
| 786 |
-
{
|
| 787 |
-
"epoch": 20.476190476190474,
|
| 788 |
-
"grad_norm": 5.423216819763184,
|
| 789 |
-
"learning_rate": 3.280423280423281e-05,
|
| 790 |
-
"loss": 0.0712,
|
| 791 |
-
"step": 860
|
| 792 |
-
},
|
| 793 |
-
{
|
| 794 |
-
"epoch": 20.714285714285715,
|
| 795 |
-
"grad_norm": 1.0831531286239624,
|
| 796 |
-
"learning_rate": 3.253968253968254e-05,
|
| 797 |
-
"loss": 0.0817,
|
| 798 |
-
"step": 870
|
| 799 |
-
},
|
| 800 |
-
{
|
| 801 |
-
"epoch": 20.952380952380953,
|
| 802 |
-
"grad_norm": 1.6317639350891113,
|
| 803 |
-
"learning_rate": 3.227513227513227e-05,
|
| 804 |
-
"loss": 0.0757,
|
| 805 |
-
"step": 880
|
| 806 |
-
},
|
| 807 |
-
{
|
| 808 |
-
"epoch": 21.0,
|
| 809 |
-
"eval_accuracy": 0.948073701842546,
|
| 810 |
-
"eval_loss": 0.20271137356758118,
|
| 811 |
-
"eval_runtime": 102.4721,
|
| 812 |
-
"eval_samples_per_second": 5.826,
|
| 813 |
-
"eval_steps_per_second": 0.185,
|
| 814 |
-
"step": 882
|
| 815 |
-
},
|
| 816 |
-
{
|
| 817 |
-
"epoch": 21.19047619047619,
|
| 818 |
-
"grad_norm": 2.1182172298431396,
|
| 819 |
-
"learning_rate": 3.2010582010582015e-05,
|
| 820 |
-
"loss": 0.0882,
|
| 821 |
-
"step": 890
|
| 822 |
-
},
|
| 823 |
-
{
|
| 824 |
-
"epoch": 21.428571428571427,
|
| 825 |
-
"grad_norm": 0.5835541486740112,
|
| 826 |
-
"learning_rate": 3.1746031746031745e-05,
|
| 827 |
-
"loss": 0.0607,
|
| 828 |
-
"step": 900
|
| 829 |
-
},
|
| 830 |
-
{
|
| 831 |
-
"epoch": 21.666666666666668,
|
| 832 |
-
"grad_norm": 1.441300392150879,
|
| 833 |
-
"learning_rate": 3.148148148148148e-05,
|
| 834 |
-
"loss": 0.0859,
|
| 835 |
-
"step": 910
|
| 836 |
-
},
|
| 837 |
-
{
|
| 838 |
-
"epoch": 21.904761904761905,
|
| 839 |
-
"grad_norm": 0.6337174773216248,
|
| 840 |
-
"learning_rate": 3.121693121693122e-05,
|
| 841 |
-
"loss": 0.0919,
|
| 842 |
-
"step": 920
|
| 843 |
-
},
|
| 844 |
-
{
|
| 845 |
-
"epoch": 22.0,
|
| 846 |
-
"eval_accuracy": 0.9614740368509213,
|
| 847 |
-
"eval_loss": 0.14867298305034637,
|
| 848 |
-
"eval_runtime": 103.3409,
|
| 849 |
-
"eval_samples_per_second": 5.777,
|
| 850 |
-
"eval_steps_per_second": 0.184,
|
| 851 |
-
"step": 924
|
| 852 |
-
},
|
| 853 |
-
{
|
| 854 |
-
"epoch": 22.142857142857142,
|
| 855 |
-
"grad_norm": 1.7783087491989136,
|
| 856 |
-
"learning_rate": 3.095238095238095e-05,
|
| 857 |
-
"loss": 0.0591,
|
| 858 |
-
"step": 930
|
| 859 |
-
},
|
| 860 |
-
{
|
| 861 |
-
"epoch": 22.38095238095238,
|
| 862 |
-
"grad_norm": 0.4258907735347748,
|
| 863 |
-
"learning_rate": 3.068783068783069e-05,
|
| 864 |
-
"loss": 0.0722,
|
| 865 |
-
"step": 940
|
| 866 |
},
|
| 867 |
{
|
| 868 |
-
"epoch":
|
| 869 |
-
"
|
| 870 |
-
"
|
| 871 |
-
"
|
| 872 |
-
"
|
|
|
|
|
|
|
| 873 |
},
|
| 874 |
{
|
| 875 |
-
"epoch":
|
| 876 |
-
"grad_norm":
|
| 877 |
-
"learning_rate": 3.
|
| 878 |
-
"loss": 0.
|
| 879 |
-
"step":
|
| 880 |
},
|
| 881 |
{
|
| 882 |
-
"epoch":
|
| 883 |
-
"eval_accuracy": 0.
|
| 884 |
-
"eval_loss": 0.
|
| 885 |
-
"eval_runtime":
|
| 886 |
-
"eval_samples_per_second":
|
| 887 |
-
"eval_steps_per_second": 0.
|
| 888 |
-
"step":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 889 |
},
|
| 890 |
{
|
| 891 |
-
"epoch":
|
| 892 |
-
"
|
| 893 |
-
"
|
| 894 |
-
"
|
| 895 |
-
"
|
|
|
|
|
|
|
| 896 |
},
|
| 897 |
{
|
| 898 |
-
"epoch":
|
| 899 |
-
"grad_norm":
|
| 900 |
"learning_rate": 2.962962962962963e-05,
|
| 901 |
-
"loss": 0.
|
| 902 |
-
"step":
|
| 903 |
-
},
|
| 904 |
-
{
|
| 905 |
-
"epoch": 23.571428571428573,
|
| 906 |
-
"grad_norm": 2.2438175678253174,
|
| 907 |
-
"learning_rate": 2.9365079365079366e-05,
|
| 908 |
-
"loss": 0.0644,
|
| 909 |
-
"step": 990
|
| 910 |
-
},
|
| 911 |
-
{
|
| 912 |
-
"epoch": 23.80952380952381,
|
| 913 |
-
"grad_norm": 0.8747345805168152,
|
| 914 |
-
"learning_rate": 2.91005291005291e-05,
|
| 915 |
-
"loss": 0.0629,
|
| 916 |
-
"step": 1000
|
| 917 |
-
},
|
| 918 |
-
{
|
| 919 |
-
"epoch": 24.0,
|
| 920 |
-
"eval_accuracy": 0.9530988274706867,
|
| 921 |
-
"eval_loss": 0.19044645130634308,
|
| 922 |
-
"eval_runtime": 103.4446,
|
| 923 |
-
"eval_samples_per_second": 5.771,
|
| 924 |
-
"eval_steps_per_second": 0.184,
|
| 925 |
-
"step": 1008
|
| 926 |
-
},
|
| 927 |
-
{
|
| 928 |
-
"epoch": 24.047619047619047,
|
| 929 |
-
"grad_norm": 1.7004871368408203,
|
| 930 |
-
"learning_rate": 2.8835978835978838e-05,
|
| 931 |
-
"loss": 0.0496,
|
| 932 |
-
"step": 1010
|
| 933 |
-
},
|
| 934 |
-
{
|
| 935 |
-
"epoch": 24.285714285714285,
|
| 936 |
-
"grad_norm": 1.3886641263961792,
|
| 937 |
-
"learning_rate": 2.857142857142857e-05,
|
| 938 |
-
"loss": 0.0468,
|
| 939 |
-
"step": 1020
|
| 940 |
-
},
|
| 941 |
-
{
|
| 942 |
-
"epoch": 24.523809523809526,
|
| 943 |
-
"grad_norm": 2.0595200061798096,
|
| 944 |
-
"learning_rate": 2.830687830687831e-05,
|
| 945 |
-
"loss": 0.0657,
|
| 946 |
-
"step": 1030
|
| 947 |
-
},
|
| 948 |
-
{
|
| 949 |
-
"epoch": 24.761904761904763,
|
| 950 |
-
"grad_norm": 0.9633322954177856,
|
| 951 |
-
"learning_rate": 2.8042328042328043e-05,
|
| 952 |
-
"loss": 0.0626,
|
| 953 |
-
"step": 1040
|
| 954 |
-
},
|
| 955 |
-
{
|
| 956 |
-
"epoch": 25.0,
|
| 957 |
-
"grad_norm": 0.629084050655365,
|
| 958 |
-
"learning_rate": 2.777777777777778e-05,
|
| 959 |
-
"loss": 0.0584,
|
| 960 |
-
"step": 1050
|
| 961 |
-
},
|
| 962 |
-
{
|
| 963 |
-
"epoch": 25.0,
|
| 964 |
-
"eval_accuracy": 0.9631490787269682,
|
| 965 |
-
"eval_loss": 0.15212486684322357,
|
| 966 |
-
"eval_runtime": 104.2052,
|
| 967 |
-
"eval_samples_per_second": 5.729,
|
| 968 |
-
"eval_steps_per_second": 0.182,
|
| 969 |
-
"step": 1050
|
| 970 |
-
},
|
| 971 |
-
{
|
| 972 |
-
"epoch": 25.238095238095237,
|
| 973 |
-
"grad_norm": 1.9510831832885742,
|
| 974 |
-
"learning_rate": 2.7513227513227512e-05,
|
| 975 |
-
"loss": 0.0514,
|
| 976 |
-
"step": 1060
|
| 977 |
-
},
|
| 978 |
-
{
|
| 979 |
-
"epoch": 25.476190476190474,
|
| 980 |
-
"grad_norm": 0.6461337208747864,
|
| 981 |
-
"learning_rate": 2.724867724867725e-05,
|
| 982 |
-
"loss": 0.0619,
|
| 983 |
-
"step": 1070
|
| 984 |
},
|
| 985 |
{
|
| 986 |
-
"epoch":
|
| 987 |
-
"
|
| 988 |
-
"
|
| 989 |
-
"
|
| 990 |
-
"
|
|
|
|
|
|
|
| 991 |
},
|
| 992 |
{
|
| 993 |
-
"epoch":
|
| 994 |
-
"grad_norm": 1.
|
| 995 |
-
"learning_rate": 2.
|
| 996 |
-
"loss": 0.
|
| 997 |
-
"step":
|
| 998 |
},
|
| 999 |
{
|
| 1000 |
-
"epoch":
|
| 1001 |
-
"eval_accuracy": 0.
|
| 1002 |
-
"eval_loss": 0.
|
| 1003 |
-
"eval_runtime":
|
| 1004 |
-
"eval_samples_per_second":
|
| 1005 |
-
"eval_steps_per_second": 0.
|
| 1006 |
-
"step":
|
| 1007 |
},
|
| 1008 |
{
|
| 1009 |
-
"epoch":
|
| 1010 |
-
"grad_norm":
|
| 1011 |
-
"learning_rate": 2.
|
| 1012 |
-
"loss": 0.
|
| 1013 |
-
"step":
|
| 1014 |
},
|
| 1015 |
{
|
| 1016 |
-
"epoch":
|
| 1017 |
-
"
|
| 1018 |
-
"
|
| 1019 |
-
"
|
| 1020 |
-
"
|
|
|
|
|
|
|
| 1021 |
},
|
| 1022 |
{
|
| 1023 |
-
"epoch":
|
| 1024 |
-
"grad_norm":
|
| 1025 |
"learning_rate": 2.5925925925925925e-05,
|
| 1026 |
-
"loss": 0.
|
| 1027 |
-
"step":
|
| 1028 |
-
},
|
| 1029 |
-
{
|
| 1030 |
-
"epoch": 26.904761904761905,
|
| 1031 |
-
"grad_norm": 0.8411057591438293,
|
| 1032 |
-
"learning_rate": 2.5661375661375664e-05,
|
| 1033 |
-
"loss": 0.062,
|
| 1034 |
-
"step": 1130
|
| 1035 |
-
},
|
| 1036 |
-
{
|
| 1037 |
-
"epoch": 27.0,
|
| 1038 |
-
"eval_accuracy": 0.9564489112227805,
|
| 1039 |
-
"eval_loss": 0.17715045809745789,
|
| 1040 |
-
"eval_runtime": 102.5942,
|
| 1041 |
-
"eval_samples_per_second": 5.819,
|
| 1042 |
-
"eval_steps_per_second": 0.185,
|
| 1043 |
-
"step": 1134
|
| 1044 |
-
},
|
| 1045 |
-
{
|
| 1046 |
-
"epoch": 27.142857142857142,
|
| 1047 |
-
"grad_norm": 1.1352622509002686,
|
| 1048 |
-
"learning_rate": 2.5396825396825397e-05,
|
| 1049 |
-
"loss": 0.0396,
|
| 1050 |
-
"step": 1140
|
| 1051 |
-
},
|
| 1052 |
-
{
|
| 1053 |
-
"epoch": 27.38095238095238,
|
| 1054 |
-
"grad_norm": 1.9047770500183105,
|
| 1055 |
-
"learning_rate": 2.5132275132275137e-05,
|
| 1056 |
-
"loss": 0.0383,
|
| 1057 |
-
"step": 1150
|
| 1058 |
-
},
|
| 1059 |
-
{
|
| 1060 |
-
"epoch": 27.61904761904762,
|
| 1061 |
-
"grad_norm": 2.154599666595459,
|
| 1062 |
-
"learning_rate": 2.4867724867724866e-05,
|
| 1063 |
-
"loss": 0.0728,
|
| 1064 |
-
"step": 1160
|
| 1065 |
-
},
|
| 1066 |
-
{
|
| 1067 |
-
"epoch": 27.857142857142858,
|
| 1068 |
-
"grad_norm": 1.5056850910186768,
|
| 1069 |
-
"learning_rate": 2.4603174603174602e-05,
|
| 1070 |
-
"loss": 0.0568,
|
| 1071 |
-
"step": 1170
|
| 1072 |
-
},
|
| 1073 |
-
{
|
| 1074 |
-
"epoch": 28.0,
|
| 1075 |
-
"eval_accuracy": 0.9564489112227805,
|
| 1076 |
-
"eval_loss": 0.14654366672039032,
|
| 1077 |
-
"eval_runtime": 103.1379,
|
| 1078 |
-
"eval_samples_per_second": 5.788,
|
| 1079 |
-
"eval_steps_per_second": 0.184,
|
| 1080 |
-
"step": 1176
|
| 1081 |
-
},
|
| 1082 |
-
{
|
| 1083 |
-
"epoch": 28.095238095238095,
|
| 1084 |
-
"grad_norm": 1.634865641593933,
|
| 1085 |
-
"learning_rate": 2.4338624338624338e-05,
|
| 1086 |
-
"loss": 0.0663,
|
| 1087 |
-
"step": 1180
|
| 1088 |
-
},
|
| 1089 |
-
{
|
| 1090 |
-
"epoch": 28.333333333333332,
|
| 1091 |
-
"grad_norm": 1.265386939048767,
|
| 1092 |
-
"learning_rate": 2.4074074074074074e-05,
|
| 1093 |
-
"loss": 0.0487,
|
| 1094 |
-
"step": 1190
|
| 1095 |
-
},
|
| 1096 |
-
{
|
| 1097 |
-
"epoch": 28.571428571428573,
|
| 1098 |
-
"grad_norm": 0.6159355044364929,
|
| 1099 |
-
"learning_rate": 2.380952380952381e-05,
|
| 1100 |
-
"loss": 0.0596,
|
| 1101 |
-
"step": 1200
|
| 1102 |
-
},
|
| 1103 |
-
{
|
| 1104 |
-
"epoch": 28.80952380952381,
|
| 1105 |
-
"grad_norm": 1.0339206457138062,
|
| 1106 |
-
"learning_rate": 2.3544973544973546e-05,
|
| 1107 |
-
"loss": 0.0453,
|
| 1108 |
-
"step": 1210
|
| 1109 |
},
|
| 1110 |
{
|
| 1111 |
-
"epoch":
|
| 1112 |
-
"eval_accuracy": 0.
|
| 1113 |
-
"eval_loss": 0.
|
| 1114 |
-
"eval_runtime":
|
| 1115 |
-
"eval_samples_per_second":
|
| 1116 |
-
"eval_steps_per_second": 0.
|
| 1117 |
-
"step":
|
| 1118 |
},
|
| 1119 |
{
|
| 1120 |
-
"epoch":
|
| 1121 |
-
"grad_norm":
|
| 1122 |
-
"learning_rate": 2.
|
| 1123 |
-
"loss": 0.
|
| 1124 |
-
"step":
|
| 1125 |
},
|
| 1126 |
{
|
| 1127 |
-
"epoch":
|
| 1128 |
-
"
|
| 1129 |
-
"
|
| 1130 |
-
"
|
| 1131 |
-
"
|
|
|
|
|
|
|
| 1132 |
},
|
| 1133 |
{
|
| 1134 |
-
"epoch":
|
| 1135 |
-
"grad_norm":
|
| 1136 |
-
"learning_rate": 2.
|
| 1137 |
-
"loss": 0.
|
| 1138 |
-
"step":
|
| 1139 |
},
|
| 1140 |
{
|
| 1141 |
-
"epoch":
|
| 1142 |
-
"
|
| 1143 |
-
"
|
| 1144 |
-
"
|
| 1145 |
-
"
|
|
|
|
|
|
|
| 1146 |
},
|
| 1147 |
{
|
| 1148 |
-
"epoch":
|
| 1149 |
-
"grad_norm":
|
| 1150 |
"learning_rate": 2.2222222222222223e-05,
|
| 1151 |
-
"loss": 0.
|
| 1152 |
-
"step":
|
| 1153 |
-
},
|
| 1154 |
-
{
|
| 1155 |
-
"epoch": 30.0,
|
| 1156 |
-
"eval_accuracy": 0.9631490787269682,
|
| 1157 |
-
"eval_loss": 0.16871798038482666,
|
| 1158 |
-
"eval_runtime": 105.6517,
|
| 1159 |
-
"eval_samples_per_second": 5.651,
|
| 1160 |
-
"eval_steps_per_second": 0.18,
|
| 1161 |
-
"step": 1260
|
| 1162 |
-
},
|
| 1163 |
-
{
|
| 1164 |
-
"epoch": 30.238095238095237,
|
| 1165 |
-
"grad_norm": 0.8756251931190491,
|
| 1166 |
-
"learning_rate": 2.1957671957671956e-05,
|
| 1167 |
-
"loss": 0.0536,
|
| 1168 |
-
"step": 1270
|
| 1169 |
-
},
|
| 1170 |
-
{
|
| 1171 |
-
"epoch": 30.476190476190474,
|
| 1172 |
-
"grad_norm": 0.7314756512641907,
|
| 1173 |
-
"learning_rate": 2.1693121693121692e-05,
|
| 1174 |
-
"loss": 0.0394,
|
| 1175 |
-
"step": 1280
|
| 1176 |
-
},
|
| 1177 |
-
{
|
| 1178 |
-
"epoch": 30.714285714285715,
|
| 1179 |
-
"grad_norm": 1.9777828454971313,
|
| 1180 |
-
"learning_rate": 2.1428571428571428e-05,
|
| 1181 |
-
"loss": 0.0346,
|
| 1182 |
-
"step": 1290
|
| 1183 |
-
},
|
| 1184 |
-
{
|
| 1185 |
-
"epoch": 30.952380952380953,
|
| 1186 |
-
"grad_norm": 1.4753316640853882,
|
| 1187 |
-
"learning_rate": 2.1164021164021164e-05,
|
| 1188 |
-
"loss": 0.0541,
|
| 1189 |
-
"step": 1300
|
| 1190 |
-
},
|
| 1191 |
-
{
|
| 1192 |
-
"epoch": 31.0,
|
| 1193 |
-
"eval_accuracy": 0.9715242881072027,
|
| 1194 |
-
"eval_loss": 0.13902144134044647,
|
| 1195 |
-
"eval_runtime": 104.4849,
|
| 1196 |
-
"eval_samples_per_second": 5.714,
|
| 1197 |
-
"eval_steps_per_second": 0.182,
|
| 1198 |
-
"step": 1302
|
| 1199 |
-
},
|
| 1200 |
-
{
|
| 1201 |
-
"epoch": 31.19047619047619,
|
| 1202 |
-
"grad_norm": 1.706081748008728,
|
| 1203 |
-
"learning_rate": 2.08994708994709e-05,
|
| 1204 |
-
"loss": 0.0613,
|
| 1205 |
-
"step": 1310
|
| 1206 |
-
},
|
| 1207 |
-
{
|
| 1208 |
-
"epoch": 31.428571428571427,
|
| 1209 |
-
"grad_norm": 0.12419818341732025,
|
| 1210 |
-
"learning_rate": 2.0634920634920636e-05,
|
| 1211 |
-
"loss": 0.075,
|
| 1212 |
-
"step": 1320
|
| 1213 |
-
},
|
| 1214 |
-
{
|
| 1215 |
-
"epoch": 31.666666666666668,
|
| 1216 |
-
"grad_norm": 0.7071540951728821,
|
| 1217 |
-
"learning_rate": 2.037037037037037e-05,
|
| 1218 |
-
"loss": 0.0468,
|
| 1219 |
-
"step": 1330
|
| 1220 |
-
},
|
| 1221 |
-
{
|
| 1222 |
-
"epoch": 31.904761904761905,
|
| 1223 |
-
"grad_norm": 0.10454891622066498,
|
| 1224 |
-
"learning_rate": 2.0105820105820105e-05,
|
| 1225 |
-
"loss": 0.0602,
|
| 1226 |
-
"step": 1340
|
| 1227 |
-
},
|
| 1228 |
-
{
|
| 1229 |
-
"epoch": 32.0,
|
| 1230 |
-
"eval_accuracy": 0.9614740368509213,
|
| 1231 |
-
"eval_loss": 0.16181902587413788,
|
| 1232 |
-
"eval_runtime": 102.8803,
|
| 1233 |
-
"eval_samples_per_second": 5.803,
|
| 1234 |
-
"eval_steps_per_second": 0.185,
|
| 1235 |
-
"step": 1344
|
| 1236 |
-
},
|
| 1237 |
-
{
|
| 1238 |
-
"epoch": 32.142857142857146,
|
| 1239 |
-
"grad_norm": 0.773094654083252,
|
| 1240 |
-
"learning_rate": 1.984126984126984e-05,
|
| 1241 |
-
"loss": 0.0555,
|
| 1242 |
-
"step": 1350
|
| 1243 |
-
},
|
| 1244 |
-
{
|
| 1245 |
-
"epoch": 32.38095238095238,
|
| 1246 |
-
"grad_norm": 1.865349292755127,
|
| 1247 |
-
"learning_rate": 1.9576719576719577e-05,
|
| 1248 |
-
"loss": 0.0518,
|
| 1249 |
-
"step": 1360
|
| 1250 |
},
|
| 1251 |
{
|
| 1252 |
-
"epoch":
|
| 1253 |
-
"
|
| 1254 |
-
"
|
| 1255 |
-
"
|
| 1256 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1257 |
},
|
| 1258 |
{
|
| 1259 |
-
"epoch":
|
| 1260 |
-
"
|
| 1261 |
-
"
|
| 1262 |
-
"
|
| 1263 |
-
"
|
|
|
|
|
|
|
| 1264 |
},
|
| 1265 |
{
|
| 1266 |
-
"epoch":
|
| 1267 |
-
"
|
| 1268 |
-
"
|
| 1269 |
-
"
|
| 1270 |
-
"
|
| 1271 |
-
"eval_steps_per_second": 0.184,
|
| 1272 |
-
"step": 1386
|
| 1273 |
},
|
| 1274 |
{
|
| 1275 |
-
"epoch":
|
| 1276 |
-
"
|
| 1277 |
-
"
|
| 1278 |
-
"
|
| 1279 |
-
"
|
|
|
|
|
|
|
| 1280 |
},
|
| 1281 |
{
|
| 1282 |
-
"epoch":
|
| 1283 |
-
"grad_norm": 1.
|
| 1284 |
"learning_rate": 1.8518518518518518e-05,
|
| 1285 |
-
"loss": 0.
|
| 1286 |
-
"step":
|
| 1287 |
-
},
|
| 1288 |
-
{
|
| 1289 |
-
"epoch": 33.57142857142857,
|
| 1290 |
-
"grad_norm": 0.6880443096160889,
|
| 1291 |
-
"learning_rate": 1.8253968253968254e-05,
|
| 1292 |
-
"loss": 0.0454,
|
| 1293 |
-
"step": 1410
|
| 1294 |
-
},
|
| 1295 |
-
{
|
| 1296 |
-
"epoch": 33.80952380952381,
|
| 1297 |
-
"grad_norm": 1.0119032859802246,
|
| 1298 |
-
"learning_rate": 1.798941798941799e-05,
|
| 1299 |
-
"loss": 0.0493,
|
| 1300 |
-
"step": 1420
|
| 1301 |
-
},
|
| 1302 |
-
{
|
| 1303 |
-
"epoch": 34.0,
|
| 1304 |
-
"eval_accuracy": 0.9631490787269682,
|
| 1305 |
-
"eval_loss": 0.1520875245332718,
|
| 1306 |
-
"eval_runtime": 104.0599,
|
| 1307 |
-
"eval_samples_per_second": 5.737,
|
| 1308 |
-
"eval_steps_per_second": 0.183,
|
| 1309 |
-
"step": 1428
|
| 1310 |
-
},
|
| 1311 |
-
{
|
| 1312 |
-
"epoch": 34.04761904761905,
|
| 1313 |
-
"grad_norm": 1.220492959022522,
|
| 1314 |
-
"learning_rate": 1.7724867724867723e-05,
|
| 1315 |
-
"loss": 0.0361,
|
| 1316 |
-
"step": 1430
|
| 1317 |
-
},
|
| 1318 |
-
{
|
| 1319 |
-
"epoch": 34.285714285714285,
|
| 1320 |
-
"grad_norm": 0.5547053217887878,
|
| 1321 |
-
"learning_rate": 1.746031746031746e-05,
|
| 1322 |
-
"loss": 0.0412,
|
| 1323 |
-
"step": 1440
|
| 1324 |
-
},
|
| 1325 |
-
{
|
| 1326 |
-
"epoch": 34.523809523809526,
|
| 1327 |
-
"grad_norm": 0.7015855312347412,
|
| 1328 |
-
"learning_rate": 1.7195767195767195e-05,
|
| 1329 |
-
"loss": 0.0425,
|
| 1330 |
-
"step": 1450
|
| 1331 |
-
},
|
| 1332 |
-
{
|
| 1333 |
-
"epoch": 34.76190476190476,
|
| 1334 |
-
"grad_norm": 1.388316035270691,
|
| 1335 |
-
"learning_rate": 1.693121693121693e-05,
|
| 1336 |
-
"loss": 0.0342,
|
| 1337 |
-
"step": 1460
|
| 1338 |
-
},
|
| 1339 |
-
{
|
| 1340 |
-
"epoch": 35.0,
|
| 1341 |
-
"grad_norm": 1.6578996181488037,
|
| 1342 |
-
"learning_rate": 1.6666666666666667e-05,
|
| 1343 |
-
"loss": 0.0606,
|
| 1344 |
-
"step": 1470
|
| 1345 |
-
},
|
| 1346 |
-
{
|
| 1347 |
-
"epoch": 35.0,
|
| 1348 |
-
"eval_accuracy": 0.9698492462311558,
|
| 1349 |
-
"eval_loss": 0.14287406206130981,
|
| 1350 |
-
"eval_runtime": 103.938,
|
| 1351 |
-
"eval_samples_per_second": 5.744,
|
| 1352 |
-
"eval_steps_per_second": 0.183,
|
| 1353 |
-
"step": 1470
|
| 1354 |
-
},
|
| 1355 |
-
{
|
| 1356 |
-
"epoch": 35.23809523809524,
|
| 1357 |
-
"grad_norm": 1.2321009635925293,
|
| 1358 |
-
"learning_rate": 1.6402116402116404e-05,
|
| 1359 |
-
"loss": 0.036,
|
| 1360 |
-
"step": 1480
|
| 1361 |
-
},
|
| 1362 |
-
{
|
| 1363 |
-
"epoch": 35.476190476190474,
|
| 1364 |
-
"grad_norm": 1.9735788106918335,
|
| 1365 |
-
"learning_rate": 1.6137566137566136e-05,
|
| 1366 |
-
"loss": 0.0485,
|
| 1367 |
-
"step": 1490
|
| 1368 |
-
},
|
| 1369 |
-
{
|
| 1370 |
-
"epoch": 35.714285714285715,
|
| 1371 |
-
"grad_norm": 0.7221766114234924,
|
| 1372 |
-
"learning_rate": 1.5873015873015872e-05,
|
| 1373 |
-
"loss": 0.0518,
|
| 1374 |
-
"step": 1500
|
| 1375 |
},
|
| 1376 |
{
|
| 1377 |
-
"epoch":
|
| 1378 |
-
"
|
| 1379 |
-
"
|
| 1380 |
-
"
|
| 1381 |
-
"
|
|
|
|
|
|
|
| 1382 |
},
|
| 1383 |
{
|
| 1384 |
-
"epoch":
|
| 1385 |
-
"
|
| 1386 |
-
"
|
| 1387 |
-
"
|
| 1388 |
-
"
|
| 1389 |
-
"eval_steps_per_second": 0.184,
|
| 1390 |
-
"step": 1512
|
| 1391 |
},
|
| 1392 |
{
|
| 1393 |
-
"epoch":
|
| 1394 |
-
"
|
| 1395 |
-
"
|
| 1396 |
-
"
|
| 1397 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1398 |
},
|
| 1399 |
{
|
| 1400 |
-
"epoch":
|
| 1401 |
-
"
|
| 1402 |
-
"
|
| 1403 |
-
"
|
| 1404 |
-
"
|
|
|
|
|
|
|
| 1405 |
},
|
| 1406 |
{
|
| 1407 |
-
"epoch":
|
| 1408 |
-
"grad_norm":
|
| 1409 |
"learning_rate": 1.4814814814814815e-05,
|
| 1410 |
-
"loss": 0.
|
| 1411 |
-
"step":
|
| 1412 |
-
},
|
| 1413 |
-
{
|
| 1414 |
-
"epoch": 36.904761904761905,
|
| 1415 |
-
"grad_norm": 0.9731617569923401,
|
| 1416 |
-
"learning_rate": 1.455026455026455e-05,
|
| 1417 |
-
"loss": 0.0432,
|
| 1418 |
-
"step": 1550
|
| 1419 |
-
},
|
| 1420 |
-
{
|
| 1421 |
-
"epoch": 37.0,
|
| 1422 |
-
"eval_accuracy": 0.966499162479062,
|
| 1423 |
-
"eval_loss": 0.14411257207393646,
|
| 1424 |
-
"eval_runtime": 103.3621,
|
| 1425 |
-
"eval_samples_per_second": 5.776,
|
| 1426 |
-
"eval_steps_per_second": 0.184,
|
| 1427 |
-
"step": 1554
|
| 1428 |
-
},
|
| 1429 |
-
{
|
| 1430 |
-
"epoch": 37.142857142857146,
|
| 1431 |
-
"grad_norm": 1.3034260272979736,
|
| 1432 |
-
"learning_rate": 1.4285714285714285e-05,
|
| 1433 |
-
"loss": 0.0376,
|
| 1434 |
-
"step": 1560
|
| 1435 |
-
},
|
| 1436 |
-
{
|
| 1437 |
-
"epoch": 37.38095238095238,
|
| 1438 |
-
"grad_norm": 0.8573827743530273,
|
| 1439 |
-
"learning_rate": 1.4021164021164022e-05,
|
| 1440 |
-
"loss": 0.0475,
|
| 1441 |
-
"step": 1570
|
| 1442 |
-
},
|
| 1443 |
-
{
|
| 1444 |
-
"epoch": 37.61904761904762,
|
| 1445 |
-
"grad_norm": 0.7534766793251038,
|
| 1446 |
-
"learning_rate": 1.3756613756613756e-05,
|
| 1447 |
-
"loss": 0.0519,
|
| 1448 |
-
"step": 1580
|
| 1449 |
-
},
|
| 1450 |
-
{
|
| 1451 |
-
"epoch": 37.857142857142854,
|
| 1452 |
-
"grad_norm": 0.7222949266433716,
|
| 1453 |
-
"learning_rate": 1.3492063492063492e-05,
|
| 1454 |
-
"loss": 0.0354,
|
| 1455 |
-
"step": 1590
|
| 1456 |
-
},
|
| 1457 |
-
{
|
| 1458 |
-
"epoch": 38.0,
|
| 1459 |
-
"eval_accuracy": 0.9681742043551089,
|
| 1460 |
-
"eval_loss": 0.15929608047008514,
|
| 1461 |
-
"eval_runtime": 107.453,
|
| 1462 |
-
"eval_samples_per_second": 5.556,
|
| 1463 |
-
"eval_steps_per_second": 0.177,
|
| 1464 |
-
"step": 1596
|
| 1465 |
-
},
|
| 1466 |
-
{
|
| 1467 |
-
"epoch": 38.095238095238095,
|
| 1468 |
-
"grad_norm": 1.8834507465362549,
|
| 1469 |
-
"learning_rate": 1.3227513227513228e-05,
|
| 1470 |
-
"loss": 0.0338,
|
| 1471 |
-
"step": 1600
|
| 1472 |
-
},
|
| 1473 |
-
{
|
| 1474 |
-
"epoch": 38.333333333333336,
|
| 1475 |
-
"grad_norm": 0.6104734539985657,
|
| 1476 |
-
"learning_rate": 1.2962962962962962e-05,
|
| 1477 |
-
"loss": 0.0462,
|
| 1478 |
-
"step": 1610
|
| 1479 |
-
},
|
| 1480 |
-
{
|
| 1481 |
-
"epoch": 38.57142857142857,
|
| 1482 |
-
"grad_norm": 0.7713958024978638,
|
| 1483 |
-
"learning_rate": 1.2698412698412699e-05,
|
| 1484 |
-
"loss": 0.0279,
|
| 1485 |
-
"step": 1620
|
| 1486 |
-
},
|
| 1487 |
-
{
|
| 1488 |
-
"epoch": 38.80952380952381,
|
| 1489 |
-
"grad_norm": 0.2185550034046173,
|
| 1490 |
-
"learning_rate": 1.2433862433862433e-05,
|
| 1491 |
-
"loss": 0.0432,
|
| 1492 |
-
"step": 1630
|
| 1493 |
},
|
| 1494 |
{
|
| 1495 |
-
"epoch":
|
| 1496 |
-
"eval_accuracy": 0.
|
| 1497 |
-
"eval_loss": 0.
|
| 1498 |
-
"eval_runtime":
|
| 1499 |
-
"eval_samples_per_second":
|
| 1500 |
-
"eval_steps_per_second": 0.
|
| 1501 |
-
"step":
|
| 1502 |
},
|
| 1503 |
{
|
| 1504 |
-
"epoch":
|
| 1505 |
-
"grad_norm":
|
| 1506 |
-
"learning_rate": 1.
|
| 1507 |
-
"loss": 0.
|
| 1508 |
-
"step":
|
| 1509 |
},
|
| 1510 |
{
|
| 1511 |
-
"epoch":
|
| 1512 |
-
"
|
| 1513 |
-
"
|
| 1514 |
-
"
|
| 1515 |
-
"
|
|
|
|
|
|
|
| 1516 |
},
|
| 1517 |
{
|
| 1518 |
-
"epoch":
|
| 1519 |
-
"grad_norm": 0.
|
| 1520 |
-
"learning_rate": 1.
|
| 1521 |
-
"loss": 0.
|
| 1522 |
-
"step":
|
| 1523 |
},
|
| 1524 |
{
|
| 1525 |
-
"epoch":
|
| 1526 |
-
"
|
| 1527 |
-
"
|
| 1528 |
-
"
|
| 1529 |
-
"
|
|
|
|
|
|
|
| 1530 |
},
|
| 1531 |
{
|
| 1532 |
-
"epoch":
|
| 1533 |
-
"grad_norm": 0.
|
| 1534 |
"learning_rate": 1.1111111111111112e-05,
|
| 1535 |
-
"loss": 0.
|
| 1536 |
-
"step":
|
| 1537 |
-
},
|
| 1538 |
-
{
|
| 1539 |
-
"epoch": 40.0,
|
| 1540 |
-
"eval_accuracy": 0.9731993299832495,
|
| 1541 |
-
"eval_loss": 0.10922118276357651,
|
| 1542 |
-
"eval_runtime": 107.5538,
|
| 1543 |
-
"eval_samples_per_second": 5.551,
|
| 1544 |
-
"eval_steps_per_second": 0.177,
|
| 1545 |
-
"step": 1680
|
| 1546 |
-
},
|
| 1547 |
-
{
|
| 1548 |
-
"epoch": 40.23809523809524,
|
| 1549 |
-
"grad_norm": 1.7765388488769531,
|
| 1550 |
-
"learning_rate": 1.0846560846560846e-05,
|
| 1551 |
-
"loss": 0.0337,
|
| 1552 |
-
"step": 1690
|
| 1553 |
-
},
|
| 1554 |
-
{
|
| 1555 |
-
"epoch": 40.476190476190474,
|
| 1556 |
-
"grad_norm": 1.7078924179077148,
|
| 1557 |
-
"learning_rate": 1.0582010582010582e-05,
|
| 1558 |
-
"loss": 0.0272,
|
| 1559 |
-
"step": 1700
|
| 1560 |
-
},
|
| 1561 |
-
{
|
| 1562 |
-
"epoch": 40.714285714285715,
|
| 1563 |
-
"grad_norm": 1.3066785335540771,
|
| 1564 |
-
"learning_rate": 1.0317460317460318e-05,
|
| 1565 |
-
"loss": 0.0408,
|
| 1566 |
-
"step": 1710
|
| 1567 |
-
},
|
| 1568 |
-
{
|
| 1569 |
-
"epoch": 40.95238095238095,
|
| 1570 |
-
"grad_norm": 0.261738121509552,
|
| 1571 |
-
"learning_rate": 1.0052910052910053e-05,
|
| 1572 |
-
"loss": 0.0288,
|
| 1573 |
-
"step": 1720
|
| 1574 |
-
},
|
| 1575 |
-
{
|
| 1576 |
-
"epoch": 41.0,
|
| 1577 |
-
"eval_accuracy": 0.966499162479062,
|
| 1578 |
-
"eval_loss": 0.15500447154045105,
|
| 1579 |
-
"eval_runtime": 107.8053,
|
| 1580 |
-
"eval_samples_per_second": 5.538,
|
| 1581 |
-
"eval_steps_per_second": 0.176,
|
| 1582 |
-
"step": 1722
|
| 1583 |
-
},
|
| 1584 |
-
{
|
| 1585 |
-
"epoch": 41.19047619047619,
|
| 1586 |
-
"grad_norm": 0.7850339412689209,
|
| 1587 |
-
"learning_rate": 9.788359788359789e-06,
|
| 1588 |
-
"loss": 0.031,
|
| 1589 |
-
"step": 1730
|
| 1590 |
-
},
|
| 1591 |
-
{
|
| 1592 |
-
"epoch": 41.42857142857143,
|
| 1593 |
-
"grad_norm": 1.2644939422607422,
|
| 1594 |
-
"learning_rate": 9.523809523809523e-06,
|
| 1595 |
-
"loss": 0.0435,
|
| 1596 |
-
"step": 1740
|
| 1597 |
-
},
|
| 1598 |
-
{
|
| 1599 |
-
"epoch": 41.666666666666664,
|
| 1600 |
-
"grad_norm": 1.832721471786499,
|
| 1601 |
-
"learning_rate": 9.259259259259259e-06,
|
| 1602 |
-
"loss": 0.0275,
|
| 1603 |
-
"step": 1750
|
| 1604 |
-
},
|
| 1605 |
-
{
|
| 1606 |
-
"epoch": 41.904761904761905,
|
| 1607 |
-
"grad_norm": 0.4885420501232147,
|
| 1608 |
-
"learning_rate": 8.994708994708995e-06,
|
| 1609 |
-
"loss": 0.0305,
|
| 1610 |
-
"step": 1760
|
| 1611 |
-
},
|
| 1612 |
-
{
|
| 1613 |
-
"epoch": 42.0,
|
| 1614 |
-
"eval_accuracy": 0.9681742043551089,
|
| 1615 |
-
"eval_loss": 0.14619530737400055,
|
| 1616 |
-
"eval_runtime": 107.0057,
|
| 1617 |
-
"eval_samples_per_second": 5.579,
|
| 1618 |
-
"eval_steps_per_second": 0.178,
|
| 1619 |
-
"step": 1764
|
| 1620 |
-
},
|
| 1621 |
-
{
|
| 1622 |
-
"epoch": 42.142857142857146,
|
| 1623 |
-
"grad_norm": 1.3336294889450073,
|
| 1624 |
-
"learning_rate": 8.73015873015873e-06,
|
| 1625 |
-
"loss": 0.0442,
|
| 1626 |
-
"step": 1770
|
| 1627 |
},
|
| 1628 |
{
|
| 1629 |
-
"epoch":
|
| 1630 |
-
"
|
| 1631 |
-
"
|
| 1632 |
-
"
|
| 1633 |
-
"
|
|
|
|
|
|
|
| 1634 |
},
|
| 1635 |
{
|
| 1636 |
-
"epoch":
|
| 1637 |
-
"grad_norm": 0.
|
| 1638 |
-
"learning_rate":
|
| 1639 |
-
"loss": 0.
|
| 1640 |
-
"step":
|
| 1641 |
},
|
| 1642 |
{
|
| 1643 |
-
"epoch":
|
| 1644 |
-
"
|
| 1645 |
-
"
|
| 1646 |
-
"
|
| 1647 |
-
"
|
|
|
|
|
|
|
| 1648 |
},
|
| 1649 |
{
|
| 1650 |
-
"epoch":
|
| 1651 |
-
"
|
| 1652 |
-
"
|
| 1653 |
-
"
|
| 1654 |
-
"
|
| 1655 |
-
"eval_steps_per_second": 0.177,
|
| 1656 |
-
"step": 1806
|
| 1657 |
},
|
| 1658 |
{
|
| 1659 |
-
"epoch":
|
| 1660 |
-
"
|
| 1661 |
-
"
|
| 1662 |
-
"
|
| 1663 |
-
"
|
|
|
|
|
|
|
| 1664 |
},
|
| 1665 |
{
|
| 1666 |
-
"epoch":
|
| 1667 |
-
"grad_norm":
|
| 1668 |
"learning_rate": 7.4074074074074075e-06,
|
| 1669 |
-
"loss": 0.
|
| 1670 |
-
"step":
|
| 1671 |
-
},
|
| 1672 |
-
{
|
| 1673 |
-
"epoch": 43.57142857142857,
|
| 1674 |
-
"grad_norm": 1.6939359903335571,
|
| 1675 |
-
"learning_rate": 7.142857142857143e-06,
|
| 1676 |
-
"loss": 0.0274,
|
| 1677 |
-
"step": 1830
|
| 1678 |
-
},
|
| 1679 |
-
{
|
| 1680 |
-
"epoch": 43.80952380952381,
|
| 1681 |
-
"grad_norm": 0.285452663898468,
|
| 1682 |
-
"learning_rate": 6.878306878306878e-06,
|
| 1683 |
-
"loss": 0.027,
|
| 1684 |
-
"step": 1840
|
| 1685 |
-
},
|
| 1686 |
-
{
|
| 1687 |
-
"epoch": 44.0,
|
| 1688 |
-
"eval_accuracy": 0.9731993299832495,
|
| 1689 |
-
"eval_loss": 0.11093433946371078,
|
| 1690 |
-
"eval_runtime": 107.916,
|
| 1691 |
-
"eval_samples_per_second": 5.532,
|
| 1692 |
-
"eval_steps_per_second": 0.176,
|
| 1693 |
-
"step": 1848
|
| 1694 |
-
},
|
| 1695 |
-
{
|
| 1696 |
-
"epoch": 44.04761904761905,
|
| 1697 |
-
"grad_norm": 1.4939181804656982,
|
| 1698 |
-
"learning_rate": 6.613756613756614e-06,
|
| 1699 |
-
"loss": 0.035,
|
| 1700 |
-
"step": 1850
|
| 1701 |
-
},
|
| 1702 |
-
{
|
| 1703 |
-
"epoch": 44.285714285714285,
|
| 1704 |
-
"grad_norm": 0.328795850276947,
|
| 1705 |
-
"learning_rate": 6.349206349206349e-06,
|
| 1706 |
-
"loss": 0.0328,
|
| 1707 |
-
"step": 1860
|
| 1708 |
-
},
|
| 1709 |
-
{
|
| 1710 |
-
"epoch": 44.523809523809526,
|
| 1711 |
-
"grad_norm": 0.7609361410140991,
|
| 1712 |
-
"learning_rate": 6.0846560846560845e-06,
|
| 1713 |
-
"loss": 0.0383,
|
| 1714 |
-
"step": 1870
|
| 1715 |
-
},
|
| 1716 |
-
{
|
| 1717 |
-
"epoch": 44.76190476190476,
|
| 1718 |
-
"grad_norm": 0.4592248201370239,
|
| 1719 |
-
"learning_rate": 5.82010582010582e-06,
|
| 1720 |
-
"loss": 0.0248,
|
| 1721 |
-
"step": 1880
|
| 1722 |
-
},
|
| 1723 |
-
{
|
| 1724 |
-
"epoch": 45.0,
|
| 1725 |
-
"grad_norm": 0.208053857088089,
|
| 1726 |
-
"learning_rate": 5.555555555555556e-06,
|
| 1727 |
-
"loss": 0.0233,
|
| 1728 |
-
"step": 1890
|
| 1729 |
-
},
|
| 1730 |
-
{
|
| 1731 |
-
"epoch": 45.0,
|
| 1732 |
-
"eval_accuracy": 0.9731993299832495,
|
| 1733 |
-
"eval_loss": 0.1315459907054901,
|
| 1734 |
-
"eval_runtime": 105.7748,
|
| 1735 |
-
"eval_samples_per_second": 5.644,
|
| 1736 |
-
"eval_steps_per_second": 0.18,
|
| 1737 |
-
"step": 1890
|
| 1738 |
-
},
|
| 1739 |
-
{
|
| 1740 |
-
"epoch": 45.23809523809524,
|
| 1741 |
-
"grad_norm": 0.4980098605155945,
|
| 1742 |
-
"learning_rate": 5.291005291005291e-06,
|
| 1743 |
-
"loss": 0.0209,
|
| 1744 |
-
"step": 1900
|
| 1745 |
},
|
| 1746 |
{
|
| 1747 |
-
"epoch":
|
| 1748 |
-
"
|
| 1749 |
-
"
|
| 1750 |
-
"
|
| 1751 |
-
"
|
|
|
|
|
|
|
| 1752 |
},
|
| 1753 |
{
|
| 1754 |
-
"epoch":
|
| 1755 |
-
"
|
| 1756 |
-
"
|
| 1757 |
-
"
|
| 1758 |
-
"
|
|
|
|
|
|
|
| 1759 |
},
|
| 1760 |
{
|
| 1761 |
-
"epoch":
|
| 1762 |
-
"grad_norm": 0.
|
| 1763 |
-
"learning_rate":
|
| 1764 |
-
"loss": 0.
|
| 1765 |
-
"step":
|
| 1766 |
},
|
| 1767 |
{
|
| 1768 |
-
"epoch":
|
| 1769 |
-
"eval_accuracy": 0.
|
| 1770 |
-
"eval_loss": 0.
|
| 1771 |
-
"eval_runtime":
|
| 1772 |
-
"eval_samples_per_second":
|
| 1773 |
-
"eval_steps_per_second": 0.
|
| 1774 |
-
"step":
|
| 1775 |
},
|
| 1776 |
{
|
| 1777 |
-
"epoch":
|
| 1778 |
-
"grad_norm":
|
| 1779 |
-
"learning_rate": 4.
|
| 1780 |
-
"loss": 0.
|
| 1781 |
-
"step":
|
| 1782 |
},
|
| 1783 |
{
|
| 1784 |
-
"epoch":
|
| 1785 |
-
"
|
| 1786 |
-
"
|
| 1787 |
-
"
|
| 1788 |
-
"
|
|
|
|
|
|
|
| 1789 |
},
|
| 1790 |
{
|
| 1791 |
-
"epoch":
|
| 1792 |
-
"grad_norm":
|
| 1793 |
"learning_rate": 3.7037037037037037e-06,
|
| 1794 |
-
"loss": 0.
|
| 1795 |
-
"step":
|
| 1796 |
-
},
|
| 1797 |
-
{
|
| 1798 |
-
"epoch": 46.904761904761905,
|
| 1799 |
-
"grad_norm": 0.760875403881073,
|
| 1800 |
-
"learning_rate": 3.439153439153439e-06,
|
| 1801 |
-
"loss": 0.0251,
|
| 1802 |
-
"step": 1970
|
| 1803 |
-
},
|
| 1804 |
-
{
|
| 1805 |
-
"epoch": 47.0,
|
| 1806 |
-
"eval_accuracy": 0.9731993299832495,
|
| 1807 |
-
"eval_loss": 0.13198845088481903,
|
| 1808 |
-
"eval_runtime": 107.949,
|
| 1809 |
-
"eval_samples_per_second": 5.53,
|
| 1810 |
-
"eval_steps_per_second": 0.176,
|
| 1811 |
-
"step": 1974
|
| 1812 |
-
},
|
| 1813 |
-
{
|
| 1814 |
-
"epoch": 47.142857142857146,
|
| 1815 |
-
"grad_norm": 0.19793380796909332,
|
| 1816 |
-
"learning_rate": 3.1746031746031746e-06,
|
| 1817 |
-
"loss": 0.0194,
|
| 1818 |
-
"step": 1980
|
| 1819 |
-
},
|
| 1820 |
-
{
|
| 1821 |
-
"epoch": 47.38095238095238,
|
| 1822 |
-
"grad_norm": 1.171229362487793,
|
| 1823 |
-
"learning_rate": 2.91005291005291e-06,
|
| 1824 |
-
"loss": 0.0207,
|
| 1825 |
-
"step": 1990
|
| 1826 |
-
},
|
| 1827 |
-
{
|
| 1828 |
-
"epoch": 47.61904761904762,
|
| 1829 |
-
"grad_norm": 1.4025137424468994,
|
| 1830 |
-
"learning_rate": 2.6455026455026455e-06,
|
| 1831 |
-
"loss": 0.0264,
|
| 1832 |
-
"step": 2000
|
| 1833 |
-
},
|
| 1834 |
-
{
|
| 1835 |
-
"epoch": 47.857142857142854,
|
| 1836 |
-
"grad_norm": 0.5469716191291809,
|
| 1837 |
-
"learning_rate": 2.3809523809523808e-06,
|
| 1838 |
-
"loss": 0.041,
|
| 1839 |
-
"step": 2010
|
| 1840 |
-
},
|
| 1841 |
-
{
|
| 1842 |
-
"epoch": 48.0,
|
| 1843 |
-
"eval_accuracy": 0.9731993299832495,
|
| 1844 |
-
"eval_loss": 0.12820355594158173,
|
| 1845 |
-
"eval_runtime": 107.9105,
|
| 1846 |
-
"eval_samples_per_second": 5.532,
|
| 1847 |
-
"eval_steps_per_second": 0.176,
|
| 1848 |
-
"step": 2016
|
| 1849 |
-
},
|
| 1850 |
-
{
|
| 1851 |
-
"epoch": 48.095238095238095,
|
| 1852 |
-
"grad_norm": 0.5962417125701904,
|
| 1853 |
-
"learning_rate": 2.1164021164021164e-06,
|
| 1854 |
-
"loss": 0.0257,
|
| 1855 |
-
"step": 2020
|
| 1856 |
-
},
|
| 1857 |
-
{
|
| 1858 |
-
"epoch": 48.333333333333336,
|
| 1859 |
-
"grad_norm": 1.9686793088912964,
|
| 1860 |
-
"learning_rate": 1.8518518518518519e-06,
|
| 1861 |
-
"loss": 0.0297,
|
| 1862 |
-
"step": 2030
|
| 1863 |
-
},
|
| 1864 |
-
{
|
| 1865 |
-
"epoch": 48.57142857142857,
|
| 1866 |
-
"grad_norm": 1.7944121360778809,
|
| 1867 |
-
"learning_rate": 1.5873015873015873e-06,
|
| 1868 |
-
"loss": 0.0272,
|
| 1869 |
-
"step": 2040
|
| 1870 |
},
|
| 1871 |
{
|
| 1872 |
-
"epoch":
|
| 1873 |
-
"
|
| 1874 |
-
"
|
| 1875 |
-
"
|
| 1876 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1877 |
},
|
| 1878 |
{
|
| 1879 |
-
"epoch":
|
| 1880 |
-
"eval_accuracy": 0.
|
| 1881 |
-
"eval_loss": 0.
|
| 1882 |
-
"eval_runtime":
|
| 1883 |
-
"eval_samples_per_second":
|
| 1884 |
-
"eval_steps_per_second": 0.
|
| 1885 |
-
"step":
|
| 1886 |
},
|
| 1887 |
{
|
| 1888 |
-
"epoch":
|
| 1889 |
-
"grad_norm": 0.
|
| 1890 |
-
"learning_rate": 1.
|
| 1891 |
-
"loss": 0.
|
| 1892 |
-
"step":
|
| 1893 |
},
|
| 1894 |
{
|
| 1895 |
-
"epoch":
|
| 1896 |
-
"
|
| 1897 |
-
"
|
| 1898 |
-
"
|
| 1899 |
-
"
|
|
|
|
|
|
|
| 1900 |
},
|
| 1901 |
{
|
| 1902 |
-
"epoch":
|
| 1903 |
-
"grad_norm":
|
| 1904 |
-
"learning_rate":
|
| 1905 |
-
"loss": 0.
|
| 1906 |
-
"step":
|
| 1907 |
},
|
| 1908 |
{
|
| 1909 |
-
"epoch":
|
| 1910 |
-
"
|
| 1911 |
-
"
|
| 1912 |
-
"
|
| 1913 |
-
"
|
|
|
|
|
|
|
| 1914 |
},
|
| 1915 |
{
|
| 1916 |
-
"epoch":
|
| 1917 |
-
"
|
| 1918 |
-
"
|
| 1919 |
-
"
|
| 1920 |
-
"
|
| 1921 |
-
|
| 1922 |
-
|
| 1923 |
-
"epoch": 50.0,
|
| 1924 |
-
"eval_accuracy": 0.9731993299832495,
|
| 1925 |
-
"eval_loss": 0.13253676891326904,
|
| 1926 |
-
"eval_runtime": 110.4625,
|
| 1927 |
-
"eval_samples_per_second": 5.405,
|
| 1928 |
-
"eval_steps_per_second": 0.172,
|
| 1929 |
-
"step": 2100
|
| 1930 |
-
},
|
| 1931 |
-
{
|
| 1932 |
-
"epoch": 50.0,
|
| 1933 |
-
"step": 2100,
|
| 1934 |
-
"total_flos": 2.0803097508518707e+19,
|
| 1935 |
-
"train_loss": 0.1354110169055916,
|
| 1936 |
-
"train_runtime": 52600.7464,
|
| 1937 |
-
"train_samples_per_second": 5.104,
|
| 1938 |
-
"train_steps_per_second": 0.04
|
| 1939 |
}
|
| 1940 |
],
|
| 1941 |
"logging_steps": 10,
|
| 1942 |
-
"max_steps":
|
| 1943 |
"num_input_tokens_seen": 0,
|
| 1944 |
"num_train_epochs": 50,
|
| 1945 |
"save_steps": 500,
|
|
@@ -1955,7 +773,7 @@
|
|
| 1955 |
"attributes": {}
|
| 1956 |
}
|
| 1957 |
},
|
| 1958 |
-
"total_flos":
|
| 1959 |
"train_batch_size": 32,
|
| 1960 |
"trial_name": null,
|
| 1961 |
"trial_params": null
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_metric": 0.9338235294117647,
|
| 3 |
+
"best_model_checkpoint": "vit-base-patch16-224-in21k-finetuned-papsmear/checkpoint-419",
|
| 4 |
+
"epoch": 46.15384615384615,
|
| 5 |
"eval_steps": 500,
|
| 6 |
+
"global_step": 450,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
+
"epoch": 0.9230769230769231,
|
| 13 |
+
"eval_accuracy": 0.2647058823529412,
|
| 14 |
+
"eval_loss": 1.7345659732818604,
|
| 15 |
+
"eval_runtime": 30.9498,
|
| 16 |
+
"eval_samples_per_second": 4.394,
|
| 17 |
+
"eval_steps_per_second": 0.162,
|
| 18 |
+
"step": 9
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 19 |
},
|
| 20 |
{
|
| 21 |
+
"epoch": 1.0256410256410255,
|
| 22 |
+
"grad_norm": 1.5590156316757202,
|
| 23 |
+
"learning_rate": 1.1111111111111112e-05,
|
| 24 |
+
"loss": 1.7645,
|
| 25 |
+
"step": 10
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 26 |
},
|
| 27 |
{
|
| 28 |
+
"epoch": 1.9487179487179487,
|
| 29 |
+
"eval_accuracy": 0.3088235294117647,
|
| 30 |
+
"eval_loss": 1.6151821613311768,
|
| 31 |
+
"eval_runtime": 30.7387,
|
| 32 |
+
"eval_samples_per_second": 4.424,
|
| 33 |
+
"eval_steps_per_second": 0.163,
|
| 34 |
+
"step": 19
|
| 35 |
},
|
| 36 |
{
|
| 37 |
+
"epoch": 2.051282051282051,
|
| 38 |
+
"grad_norm": 1.000002384185791,
|
| 39 |
+
"learning_rate": 2.2222222222222223e-05,
|
| 40 |
+
"loss": 1.661,
|
| 41 |
+
"step": 20
|
|
|
|
|
|
|
| 42 |
},
|
| 43 |
{
|
| 44 |
+
"epoch": 2.9743589743589745,
|
| 45 |
+
"eval_accuracy": 0.4117647058823529,
|
| 46 |
+
"eval_loss": 1.4663141965866089,
|
| 47 |
+
"eval_runtime": 30.7848,
|
| 48 |
+
"eval_samples_per_second": 4.418,
|
| 49 |
+
"eval_steps_per_second": 0.162,
|
| 50 |
+
"step": 29
|
| 51 |
},
|
| 52 |
{
|
| 53 |
+
"epoch": 3.076923076923077,
|
| 54 |
+
"grad_norm": 1.144618272781372,
|
| 55 |
"learning_rate": 3.3333333333333335e-05,
|
| 56 |
+
"loss": 1.496,
|
| 57 |
+
"step": 30
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 58 |
},
|
| 59 |
{
|
| 60 |
"epoch": 4.0,
|
| 61 |
+
"eval_accuracy": 0.4852941176470588,
|
| 62 |
+
"eval_loss": 1.2988938093185425,
|
| 63 |
+
"eval_runtime": 29.3106,
|
| 64 |
+
"eval_samples_per_second": 4.64,
|
| 65 |
+
"eval_steps_per_second": 0.171,
|
| 66 |
+
"step": 39
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 67 |
},
|
| 68 |
{
|
| 69 |
+
"epoch": 4.102564102564102,
|
| 70 |
+
"grad_norm": 0.9698243141174316,
|
| 71 |
+
"learning_rate": 4.4444444444444447e-05,
|
| 72 |
+
"loss": 1.3097,
|
| 73 |
+
"step": 40
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 74 |
},
|
| 75 |
{
|
| 76 |
+
"epoch": 4.923076923076923,
|
| 77 |
+
"eval_accuracy": 0.5588235294117647,
|
| 78 |
+
"eval_loss": 1.1490617990493774,
|
| 79 |
+
"eval_runtime": 29.781,
|
| 80 |
+
"eval_samples_per_second": 4.567,
|
| 81 |
+
"eval_steps_per_second": 0.168,
|
| 82 |
+
"step": 48
|
| 83 |
},
|
| 84 |
{
|
| 85 |
+
"epoch": 5.128205128205128,
|
| 86 |
+
"grad_norm": 1.00479257106781,
|
| 87 |
+
"learning_rate": 4.938271604938271e-05,
|
| 88 |
+
"loss": 1.091,
|
| 89 |
+
"step": 50
|
| 90 |
},
|
| 91 |
{
|
| 92 |
+
"epoch": 5.948717948717949,
|
| 93 |
+
"eval_accuracy": 0.7205882352941176,
|
| 94 |
+
"eval_loss": 0.9932733178138733,
|
| 95 |
+
"eval_runtime": 30.4494,
|
| 96 |
+
"eval_samples_per_second": 4.466,
|
| 97 |
+
"eval_steps_per_second": 0.164,
|
| 98 |
+
"step": 58
|
| 99 |
},
|
| 100 |
{
|
| 101 |
+
"epoch": 6.153846153846154,
|
| 102 |
+
"grad_norm": 1.4680265188217163,
|
| 103 |
"learning_rate": 4.814814814814815e-05,
|
| 104 |
+
"loss": 0.9088,
|
| 105 |
+
"step": 60
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 106 |
},
|
| 107 |
{
|
| 108 |
+
"epoch": 6.9743589743589745,
|
| 109 |
+
"eval_accuracy": 0.6985294117647058,
|
| 110 |
+
"eval_loss": 0.9170573353767395,
|
| 111 |
+
"eval_runtime": 29.5689,
|
| 112 |
+
"eval_samples_per_second": 4.599,
|
| 113 |
+
"eval_steps_per_second": 0.169,
|
| 114 |
+
"step": 68
|
| 115 |
},
|
| 116 |
{
|
| 117 |
+
"epoch": 7.17948717948718,
|
| 118 |
+
"grad_norm": 1.2828285694122314,
|
| 119 |
+
"learning_rate": 4.691358024691358e-05,
|
| 120 |
+
"loss": 0.7858,
|
| 121 |
+
"step": 70
|
| 122 |
},
|
| 123 |
{
|
| 124 |
"epoch": 8.0,
|
| 125 |
+
"eval_accuracy": 0.7720588235294118,
|
| 126 |
+
"eval_loss": 0.8300582766532898,
|
| 127 |
+
"eval_runtime": 30.9278,
|
| 128 |
+
"eval_samples_per_second": 4.397,
|
| 129 |
+
"eval_steps_per_second": 0.162,
|
| 130 |
+
"step": 78
|
| 131 |
+
},
|
| 132 |
+
{
|
| 133 |
+
"epoch": 8.205128205128204,
|
| 134 |
+
"grad_norm": 1.7692698240280151,
|
| 135 |
+
"learning_rate": 4.567901234567901e-05,
|
| 136 |
+
"loss": 0.7016,
|
| 137 |
+
"step": 80
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 138 |
},
|
| 139 |
{
|
| 140 |
+
"epoch": 8.923076923076923,
|
| 141 |
+
"eval_accuracy": 0.7352941176470589,
|
| 142 |
+
"eval_loss": 0.7925190925598145,
|
| 143 |
+
"eval_runtime": 32.3375,
|
| 144 |
+
"eval_samples_per_second": 4.206,
|
| 145 |
+
"eval_steps_per_second": 0.155,
|
| 146 |
+
"step": 87
|
| 147 |
},
|
| 148 |
{
|
| 149 |
+
"epoch": 9.23076923076923,
|
| 150 |
+
"grad_norm": 1.7534127235412598,
|
| 151 |
"learning_rate": 4.4444444444444447e-05,
|
| 152 |
+
"loss": 0.6136,
|
| 153 |
+
"step": 90
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 154 |
},
|
| 155 |
{
|
| 156 |
+
"epoch": 9.948717948717949,
|
| 157 |
+
"eval_accuracy": 0.7647058823529411,
|
| 158 |
+
"eval_loss": 0.6992273330688477,
|
| 159 |
+
"eval_runtime": 30.502,
|
| 160 |
+
"eval_samples_per_second": 4.459,
|
| 161 |
+
"eval_steps_per_second": 0.164,
|
| 162 |
+
"step": 97
|
| 163 |
},
|
| 164 |
{
|
| 165 |
+
"epoch": 10.256410256410255,
|
| 166 |
+
"grad_norm": 1.2740432024002075,
|
| 167 |
+
"learning_rate": 4.3209876543209875e-05,
|
| 168 |
+
"loss": 0.532,
|
| 169 |
+
"step": 100
|
| 170 |
},
|
| 171 |
{
|
| 172 |
+
"epoch": 10.974358974358974,
|
| 173 |
+
"eval_accuracy": 0.8308823529411765,
|
| 174 |
+
"eval_loss": 0.6400743126869202,
|
| 175 |
+
"eval_runtime": 31.9094,
|
| 176 |
+
"eval_samples_per_second": 4.262,
|
| 177 |
+
"eval_steps_per_second": 0.157,
|
| 178 |
+
"step": 107
|
| 179 |
},
|
| 180 |
{
|
| 181 |
+
"epoch": 11.282051282051283,
|
| 182 |
+
"grad_norm": 1.563349723815918,
|
| 183 |
+
"learning_rate": 4.197530864197531e-05,
|
| 184 |
+
"loss": 0.5018,
|
| 185 |
+
"step": 110
|
| 186 |
},
|
| 187 |
{
|
| 188 |
"epoch": 12.0,
|
| 189 |
+
"eval_accuracy": 0.8382352941176471,
|
| 190 |
+
"eval_loss": 0.5786880254745483,
|
| 191 |
+
"eval_runtime": 29.8171,
|
| 192 |
+
"eval_samples_per_second": 4.561,
|
| 193 |
+
"eval_steps_per_second": 0.168,
|
| 194 |
+
"step": 117
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 195 |
},
|
| 196 |
{
|
| 197 |
+
"epoch": 12.307692307692308,
|
| 198 |
+
"grad_norm": 2.5129995346069336,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 199 |
"learning_rate": 4.074074074074074e-05,
|
| 200 |
+
"loss": 0.4279,
|
| 201 |
+
"step": 120
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 202 |
},
|
| 203 |
{
|
| 204 |
+
"epoch": 12.923076923076923,
|
| 205 |
+
"eval_accuracy": 0.8088235294117647,
|
| 206 |
+
"eval_loss": 0.6129594445228577,
|
| 207 |
+
"eval_runtime": 29.826,
|
| 208 |
+
"eval_samples_per_second": 4.56,
|
| 209 |
+
"eval_steps_per_second": 0.168,
|
| 210 |
+
"step": 126
|
| 211 |
},
|
| 212 |
{
|
| 213 |
+
"epoch": 13.333333333333334,
|
| 214 |
+
"grad_norm": 1.6143475770950317,
|
| 215 |
+
"learning_rate": 3.950617283950617e-05,
|
| 216 |
+
"loss": 0.4116,
|
| 217 |
+
"step": 130
|
| 218 |
},
|
| 219 |
{
|
| 220 |
+
"epoch": 13.948717948717949,
|
| 221 |
+
"eval_accuracy": 0.8382352941176471,
|
| 222 |
+
"eval_loss": 0.5089898109436035,
|
| 223 |
+
"eval_runtime": 29.9983,
|
| 224 |
+
"eval_samples_per_second": 4.534,
|
| 225 |
+
"eval_steps_per_second": 0.167,
|
| 226 |
+
"step": 136
|
| 227 |
},
|
| 228 |
{
|
| 229 |
+
"epoch": 14.35897435897436,
|
| 230 |
+
"grad_norm": 1.6241114139556885,
|
| 231 |
+
"learning_rate": 3.82716049382716e-05,
|
| 232 |
+
"loss": 0.3848,
|
| 233 |
+
"step": 140
|
| 234 |
},
|
| 235 |
{
|
| 236 |
+
"epoch": 14.974358974358974,
|
| 237 |
+
"eval_accuracy": 0.8676470588235294,
|
| 238 |
+
"eval_loss": 0.5165212154388428,
|
| 239 |
+
"eval_runtime": 30.2024,
|
| 240 |
+
"eval_samples_per_second": 4.503,
|
| 241 |
+
"eval_steps_per_second": 0.166,
|
| 242 |
+
"step": 146
|
| 243 |
},
|
| 244 |
{
|
| 245 |
+
"epoch": 15.384615384615385,
|
| 246 |
+
"grad_norm": 2.7321174144744873,
|
| 247 |
"learning_rate": 3.7037037037037037e-05,
|
| 248 |
+
"loss": 0.3449,
|
| 249 |
+
"step": 150
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 250 |
},
|
| 251 |
{
|
| 252 |
+
"epoch": 16.0,
|
| 253 |
+
"eval_accuracy": 0.8382352941176471,
|
| 254 |
+
"eval_loss": 0.4842991530895233,
|
| 255 |
+
"eval_runtime": 29.5578,
|
| 256 |
+
"eval_samples_per_second": 4.601,
|
| 257 |
+
"eval_steps_per_second": 0.169,
|
| 258 |
+
"step": 156
|
| 259 |
+
},
|
| 260 |
+
{
|
| 261 |
+
"epoch": 16.41025641025641,
|
| 262 |
+
"grad_norm": 1.7041429281234741,
|
| 263 |
+
"learning_rate": 3.580246913580247e-05,
|
| 264 |
+
"loss": 0.3008,
|
| 265 |
+
"step": 160
|
| 266 |
},
|
| 267 |
{
|
| 268 |
+
"epoch": 16.923076923076923,
|
| 269 |
+
"eval_accuracy": 0.8455882352941176,
|
| 270 |
+
"eval_loss": 0.5460208058357239,
|
| 271 |
+
"eval_runtime": 29.7055,
|
| 272 |
+
"eval_samples_per_second": 4.578,
|
| 273 |
+
"eval_steps_per_second": 0.168,
|
| 274 |
+
"step": 165
|
| 275 |
},
|
| 276 |
{
|
| 277 |
+
"epoch": 17.435897435897434,
|
| 278 |
+
"grad_norm": 2.1159939765930176,
|
| 279 |
+
"learning_rate": 3.45679012345679e-05,
|
| 280 |
+
"loss": 0.2797,
|
| 281 |
+
"step": 170
|
| 282 |
},
|
| 283 |
{
|
| 284 |
+
"epoch": 17.94871794871795,
|
| 285 |
+
"eval_accuracy": 0.8308823529411765,
|
| 286 |
+
"eval_loss": 0.4984971880912781,
|
| 287 |
+
"eval_runtime": 29.7303,
|
| 288 |
+
"eval_samples_per_second": 4.574,
|
| 289 |
+
"eval_steps_per_second": 0.168,
|
| 290 |
+
"step": 175
|
| 291 |
},
|
| 292 |
{
|
| 293 |
+
"epoch": 18.46153846153846,
|
| 294 |
+
"grad_norm": 1.050848126411438,
|
| 295 |
"learning_rate": 3.3333333333333335e-05,
|
| 296 |
+
"loss": 0.2696,
|
| 297 |
+
"step": 180
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 298 |
},
|
| 299 |
{
|
| 300 |
+
"epoch": 18.974358974358974,
|
| 301 |
+
"eval_accuracy": 0.8455882352941176,
|
| 302 |
+
"eval_loss": 0.5585992336273193,
|
| 303 |
+
"eval_runtime": 29.5612,
|
| 304 |
+
"eval_samples_per_second": 4.601,
|
| 305 |
+
"eval_steps_per_second": 0.169,
|
| 306 |
+
"step": 185
|
| 307 |
},
|
| 308 |
{
|
| 309 |
+
"epoch": 19.487179487179485,
|
| 310 |
+
"grad_norm": 2.385378360748291,
|
| 311 |
+
"learning_rate": 3.209876543209876e-05,
|
| 312 |
+
"loss": 0.2633,
|
| 313 |
+
"step": 190
|
| 314 |
},
|
| 315 |
{
|
| 316 |
+
"epoch": 20.0,
|
| 317 |
+
"eval_accuracy": 0.9044117647058824,
|
| 318 |
+
"eval_loss": 0.43493831157684326,
|
| 319 |
+
"eval_runtime": 29.6728,
|
| 320 |
+
"eval_samples_per_second": 4.583,
|
| 321 |
+
"eval_steps_per_second": 0.169,
|
| 322 |
+
"step": 195
|
| 323 |
+
},
|
| 324 |
+
{
|
| 325 |
+
"epoch": 20.51282051282051,
|
| 326 |
+
"grad_norm": 1.972583293914795,
|
| 327 |
+
"learning_rate": 3.08641975308642e-05,
|
| 328 |
+
"loss": 0.2569,
|
| 329 |
+
"step": 200
|
| 330 |
},
|
| 331 |
{
|
| 332 |
+
"epoch": 20.923076923076923,
|
| 333 |
+
"eval_accuracy": 0.8897058823529411,
|
| 334 |
+
"eval_loss": 0.4017449617385864,
|
| 335 |
+
"eval_runtime": 31.5372,
|
| 336 |
+
"eval_samples_per_second": 4.312,
|
| 337 |
+
"eval_steps_per_second": 0.159,
|
| 338 |
+
"step": 204
|
| 339 |
},
|
| 340 |
{
|
| 341 |
+
"epoch": 21.53846153846154,
|
| 342 |
+
"grad_norm": 2.712113380432129,
|
| 343 |
"learning_rate": 2.962962962962963e-05,
|
| 344 |
+
"loss": 0.27,
|
| 345 |
+
"step": 210
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 346 |
},
|
| 347 |
{
|
| 348 |
+
"epoch": 21.94871794871795,
|
| 349 |
+
"eval_accuracy": 0.8602941176470589,
|
| 350 |
+
"eval_loss": 0.4758412837982178,
|
| 351 |
+
"eval_runtime": 30.0407,
|
| 352 |
+
"eval_samples_per_second": 4.527,
|
| 353 |
+
"eval_steps_per_second": 0.166,
|
| 354 |
+
"step": 214
|
| 355 |
},
|
| 356 |
{
|
| 357 |
+
"epoch": 22.564102564102566,
|
| 358 |
+
"grad_norm": 1.7184677124023438,
|
| 359 |
+
"learning_rate": 2.839506172839506e-05,
|
| 360 |
+
"loss": 0.2706,
|
| 361 |
+
"step": 220
|
| 362 |
},
|
| 363 |
{
|
| 364 |
+
"epoch": 22.974358974358974,
|
| 365 |
+
"eval_accuracy": 0.8897058823529411,
|
| 366 |
+
"eval_loss": 0.41326794028282166,
|
| 367 |
+
"eval_runtime": 29.7459,
|
| 368 |
+
"eval_samples_per_second": 4.572,
|
| 369 |
+
"eval_steps_per_second": 0.168,
|
| 370 |
+
"step": 224
|
| 371 |
},
|
| 372 |
{
|
| 373 |
+
"epoch": 23.58974358974359,
|
| 374 |
+
"grad_norm": 1.8792766332626343,
|
| 375 |
+
"learning_rate": 2.7160493827160493e-05,
|
| 376 |
+
"loss": 0.2211,
|
| 377 |
+
"step": 230
|
| 378 |
},
|
| 379 |
{
|
| 380 |
+
"epoch": 24.0,
|
| 381 |
+
"eval_accuracy": 0.9117647058823529,
|
| 382 |
+
"eval_loss": 0.3844151794910431,
|
| 383 |
+
"eval_runtime": 29.8072,
|
| 384 |
+
"eval_samples_per_second": 4.563,
|
| 385 |
+
"eval_steps_per_second": 0.168,
|
| 386 |
+
"step": 234
|
| 387 |
},
|
| 388 |
{
|
| 389 |
+
"epoch": 24.615384615384617,
|
| 390 |
+
"grad_norm": 2.2763214111328125,
|
| 391 |
"learning_rate": 2.5925925925925925e-05,
|
| 392 |
+
"loss": 0.1977,
|
| 393 |
+
"step": 240
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 394 |
},
|
| 395 |
{
|
| 396 |
+
"epoch": 24.923076923076923,
|
| 397 |
+
"eval_accuracy": 0.9264705882352942,
|
| 398 |
+
"eval_loss": 0.34974199533462524,
|
| 399 |
+
"eval_runtime": 29.6847,
|
| 400 |
+
"eval_samples_per_second": 4.581,
|
| 401 |
+
"eval_steps_per_second": 0.168,
|
| 402 |
+
"step": 243
|
| 403 |
},
|
| 404 |
{
|
| 405 |
+
"epoch": 25.641025641025642,
|
| 406 |
+
"grad_norm": 4.190616130828857,
|
| 407 |
+
"learning_rate": 2.4691358024691357e-05,
|
| 408 |
+
"loss": 0.1969,
|
| 409 |
+
"step": 250
|
| 410 |
},
|
| 411 |
{
|
| 412 |
+
"epoch": 25.94871794871795,
|
| 413 |
+
"eval_accuracy": 0.9044117647058824,
|
| 414 |
+
"eval_loss": 0.37360796332359314,
|
| 415 |
+
"eval_runtime": 31.1254,
|
| 416 |
+
"eval_samples_per_second": 4.369,
|
| 417 |
+
"eval_steps_per_second": 0.161,
|
| 418 |
+
"step": 253
|
| 419 |
},
|
| 420 |
{
|
| 421 |
+
"epoch": 26.666666666666668,
|
| 422 |
+
"grad_norm": 0.6893692016601562,
|
| 423 |
+
"learning_rate": 2.345679012345679e-05,
|
| 424 |
+
"loss": 0.1776,
|
| 425 |
+
"step": 260
|
| 426 |
},
|
| 427 |
{
|
| 428 |
+
"epoch": 26.974358974358974,
|
| 429 |
+
"eval_accuracy": 0.9044117647058824,
|
| 430 |
+
"eval_loss": 0.3796656131744385,
|
| 431 |
+
"eval_runtime": 29.8455,
|
| 432 |
+
"eval_samples_per_second": 4.557,
|
| 433 |
+
"eval_steps_per_second": 0.168,
|
| 434 |
+
"step": 263
|
| 435 |
},
|
| 436 |
{
|
| 437 |
+
"epoch": 27.692307692307693,
|
| 438 |
+
"grad_norm": 2.18731427192688,
|
| 439 |
"learning_rate": 2.2222222222222223e-05,
|
| 440 |
+
"loss": 0.1787,
|
| 441 |
+
"step": 270
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 442 |
},
|
| 443 |
{
|
| 444 |
+
"epoch": 28.0,
|
| 445 |
+
"eval_accuracy": 0.8897058823529411,
|
| 446 |
+
"eval_loss": 0.39490777254104614,
|
| 447 |
+
"eval_runtime": 29.2795,
|
| 448 |
+
"eval_samples_per_second": 4.645,
|
| 449 |
+
"eval_steps_per_second": 0.171,
|
| 450 |
+
"step": 273
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 28.71794871794872,
|
| 454 |
+
"grad_norm": 1.656586766242981,
|
| 455 |
+
"learning_rate": 2.0987654320987655e-05,
|
| 456 |
+
"loss": 0.18,
|
| 457 |
+
"step": 280
|
| 458 |
},
|
| 459 |
{
|
| 460 |
+
"epoch": 28.923076923076923,
|
| 461 |
+
"eval_accuracy": 0.9264705882352942,
|
| 462 |
+
"eval_loss": 0.32775887846946716,
|
| 463 |
+
"eval_runtime": 29.3934,
|
| 464 |
+
"eval_samples_per_second": 4.627,
|
| 465 |
+
"eval_steps_per_second": 0.17,
|
| 466 |
+
"step": 282
|
| 467 |
},
|
| 468 |
{
|
| 469 |
+
"epoch": 29.743589743589745,
|
| 470 |
+
"grad_norm": 1.3221951723098755,
|
| 471 |
+
"learning_rate": 1.9753086419753087e-05,
|
| 472 |
+
"loss": 0.1797,
|
| 473 |
+
"step": 290
|
|
|
|
|
|
|
| 474 |
},
|
| 475 |
{
|
| 476 |
+
"epoch": 29.94871794871795,
|
| 477 |
+
"eval_accuracy": 0.9044117647058824,
|
| 478 |
+
"eval_loss": 0.36148715019226074,
|
| 479 |
+
"eval_runtime": 29.46,
|
| 480 |
+
"eval_samples_per_second": 4.616,
|
| 481 |
+
"eval_steps_per_second": 0.17,
|
| 482 |
+
"step": 292
|
| 483 |
},
|
| 484 |
{
|
| 485 |
+
"epoch": 30.76923076923077,
|
| 486 |
+
"grad_norm": 1.227844476699829,
|
| 487 |
"learning_rate": 1.8518518518518518e-05,
|
| 488 |
+
"loss": 0.1665,
|
| 489 |
+
"step": 300
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 490 |
},
|
| 491 |
{
|
| 492 |
+
"epoch": 30.974358974358974,
|
| 493 |
+
"eval_accuracy": 0.8602941176470589,
|
| 494 |
+
"eval_loss": 0.4174344539642334,
|
| 495 |
+
"eval_runtime": 29.6798,
|
| 496 |
+
"eval_samples_per_second": 4.582,
|
| 497 |
+
"eval_steps_per_second": 0.168,
|
| 498 |
+
"step": 302
|
| 499 |
},
|
| 500 |
{
|
| 501 |
+
"epoch": 31.794871794871796,
|
| 502 |
+
"grad_norm": 2.3732173442840576,
|
| 503 |
+
"learning_rate": 1.728395061728395e-05,
|
| 504 |
+
"loss": 0.163,
|
| 505 |
+
"step": 310
|
|
|
|
|
|
|
| 506 |
},
|
| 507 |
{
|
| 508 |
+
"epoch": 32.0,
|
| 509 |
+
"eval_accuracy": 0.8970588235294118,
|
| 510 |
+
"eval_loss": 0.3574081063270569,
|
| 511 |
+
"eval_runtime": 29.8114,
|
| 512 |
+
"eval_samples_per_second": 4.562,
|
| 513 |
+
"eval_steps_per_second": 0.168,
|
| 514 |
+
"step": 312
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 32.82051282051282,
|
| 518 |
+
"grad_norm": 2.783026933670044,
|
| 519 |
+
"learning_rate": 1.604938271604938e-05,
|
| 520 |
+
"loss": 0.1498,
|
| 521 |
+
"step": 320
|
| 522 |
},
|
| 523 |
{
|
| 524 |
+
"epoch": 32.92307692307692,
|
| 525 |
+
"eval_accuracy": 0.9044117647058824,
|
| 526 |
+
"eval_loss": 0.35905760526657104,
|
| 527 |
+
"eval_runtime": 30.5584,
|
| 528 |
+
"eval_samples_per_second": 4.45,
|
| 529 |
+
"eval_steps_per_second": 0.164,
|
| 530 |
+
"step": 321
|
| 531 |
},
|
| 532 |
{
|
| 533 |
+
"epoch": 33.84615384615385,
|
| 534 |
+
"grad_norm": 2.496835231781006,
|
| 535 |
"learning_rate": 1.4814814814814815e-05,
|
| 536 |
+
"loss": 0.1405,
|
| 537 |
+
"step": 330
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 538 |
},
|
| 539 |
{
|
| 540 |
+
"epoch": 33.94871794871795,
|
| 541 |
+
"eval_accuracy": 0.9191176470588235,
|
| 542 |
+
"eval_loss": 0.30173459649086,
|
| 543 |
+
"eval_runtime": 30.2786,
|
| 544 |
+
"eval_samples_per_second": 4.492,
|
| 545 |
+
"eval_steps_per_second": 0.165,
|
| 546 |
+
"step": 331
|
| 547 |
},
|
| 548 |
{
|
| 549 |
+
"epoch": 34.87179487179487,
|
| 550 |
+
"grad_norm": 2.3252382278442383,
|
| 551 |
+
"learning_rate": 1.3580246913580247e-05,
|
| 552 |
+
"loss": 0.155,
|
| 553 |
+
"step": 340
|
| 554 |
},
|
| 555 |
{
|
| 556 |
+
"epoch": 34.97435897435897,
|
| 557 |
+
"eval_accuracy": 0.9264705882352942,
|
| 558 |
+
"eval_loss": 0.3303259611129761,
|
| 559 |
+
"eval_runtime": 29.5808,
|
| 560 |
+
"eval_samples_per_second": 4.598,
|
| 561 |
+
"eval_steps_per_second": 0.169,
|
| 562 |
+
"step": 341
|
| 563 |
},
|
| 564 |
{
|
| 565 |
+
"epoch": 35.8974358974359,
|
| 566 |
+
"grad_norm": 0.8514438271522522,
|
| 567 |
+
"learning_rate": 1.2345679012345678e-05,
|
| 568 |
+
"loss": 0.1519,
|
| 569 |
+
"step": 350
|
| 570 |
},
|
| 571 |
{
|
| 572 |
+
"epoch": 36.0,
|
| 573 |
+
"eval_accuracy": 0.8970588235294118,
|
| 574 |
+
"eval_loss": 0.3559113144874573,
|
| 575 |
+
"eval_runtime": 30.264,
|
| 576 |
+
"eval_samples_per_second": 4.494,
|
| 577 |
+
"eval_steps_per_second": 0.165,
|
| 578 |
+
"step": 351
|
| 579 |
},
|
| 580 |
{
|
| 581 |
+
"epoch": 36.92307692307692,
|
| 582 |
+
"grad_norm": 0.9986769556999207,
|
| 583 |
"learning_rate": 1.1111111111111112e-05,
|
| 584 |
+
"loss": 0.1415,
|
| 585 |
+
"step": 360
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 586 |
},
|
| 587 |
{
|
| 588 |
+
"epoch": 36.92307692307692,
|
| 589 |
+
"eval_accuracy": 0.9191176470588235,
|
| 590 |
+
"eval_loss": 0.2890462279319763,
|
| 591 |
+
"eval_runtime": 30.7844,
|
| 592 |
+
"eval_samples_per_second": 4.418,
|
| 593 |
+
"eval_steps_per_second": 0.162,
|
| 594 |
+
"step": 360
|
| 595 |
},
|
| 596 |
{
|
| 597 |
+
"epoch": 37.94871794871795,
|
| 598 |
+
"grad_norm": 0.9639208912849426,
|
| 599 |
+
"learning_rate": 9.876543209876543e-06,
|
| 600 |
+
"loss": 0.1256,
|
| 601 |
+
"step": 370
|
| 602 |
},
|
| 603 |
{
|
| 604 |
+
"epoch": 37.94871794871795,
|
| 605 |
+
"eval_accuracy": 0.8897058823529411,
|
| 606 |
+
"eval_loss": 0.3445207476615906,
|
| 607 |
+
"eval_runtime": 29.9807,
|
| 608 |
+
"eval_samples_per_second": 4.536,
|
| 609 |
+
"eval_steps_per_second": 0.167,
|
| 610 |
+
"step": 370
|
| 611 |
},
|
| 612 |
{
|
| 613 |
+
"epoch": 38.97435897435897,
|
| 614 |
+
"grad_norm": 0.9389523267745972,
|
| 615 |
+
"learning_rate": 8.641975308641975e-06,
|
| 616 |
+
"loss": 0.1217,
|
| 617 |
+
"step": 380
|
|
|
|
|
|
|
| 618 |
},
|
| 619 |
{
|
| 620 |
+
"epoch": 38.97435897435897,
|
| 621 |
+
"eval_accuracy": 0.9117647058823529,
|
| 622 |
+
"eval_loss": 0.3434993028640747,
|
| 623 |
+
"eval_runtime": 30.2833,
|
| 624 |
+
"eval_samples_per_second": 4.491,
|
| 625 |
+
"eval_steps_per_second": 0.165,
|
| 626 |
+
"step": 380
|
| 627 |
},
|
| 628 |
{
|
| 629 |
+
"epoch": 40.0,
|
| 630 |
+
"grad_norm": 7.135587215423584,
|
| 631 |
"learning_rate": 7.4074074074074075e-06,
|
| 632 |
+
"loss": 0.1285,
|
| 633 |
+
"step": 390
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 634 |
},
|
| 635 |
{
|
| 636 |
+
"epoch": 40.0,
|
| 637 |
+
"eval_accuracy": 0.9191176470588235,
|
| 638 |
+
"eval_loss": 0.30249181389808655,
|
| 639 |
+
"eval_runtime": 29.8301,
|
| 640 |
+
"eval_samples_per_second": 4.559,
|
| 641 |
+
"eval_steps_per_second": 0.168,
|
| 642 |
+
"step": 390
|
| 643 |
},
|
| 644 |
{
|
| 645 |
+
"epoch": 40.92307692307692,
|
| 646 |
+
"eval_accuracy": 0.8823529411764706,
|
| 647 |
+
"eval_loss": 0.36019423604011536,
|
| 648 |
+
"eval_runtime": 29.9271,
|
| 649 |
+
"eval_samples_per_second": 4.544,
|
| 650 |
+
"eval_steps_per_second": 0.167,
|
| 651 |
+
"step": 399
|
| 652 |
},
|
| 653 |
{
|
| 654 |
+
"epoch": 41.02564102564103,
|
| 655 |
+
"grad_norm": 0.5793786644935608,
|
| 656 |
+
"learning_rate": 6.172839506172839e-06,
|
| 657 |
+
"loss": 0.1301,
|
| 658 |
+
"step": 400
|
| 659 |
},
|
| 660 |
{
|
| 661 |
+
"epoch": 41.94871794871795,
|
| 662 |
+
"eval_accuracy": 0.8897058823529411,
|
| 663 |
+
"eval_loss": 0.3336350917816162,
|
| 664 |
+
"eval_runtime": 32.0768,
|
| 665 |
+
"eval_samples_per_second": 4.24,
|
| 666 |
+
"eval_steps_per_second": 0.156,
|
| 667 |
+
"step": 409
|
| 668 |
},
|
| 669 |
{
|
| 670 |
+
"epoch": 42.05128205128205,
|
| 671 |
+
"grad_norm": 2.5887136459350586,
|
| 672 |
+
"learning_rate": 4.938271604938272e-06,
|
| 673 |
+
"loss": 0.1243,
|
| 674 |
+
"step": 410
|
| 675 |
},
|
| 676 |
{
|
| 677 |
+
"epoch": 42.97435897435897,
|
| 678 |
+
"eval_accuracy": 0.9338235294117647,
|
| 679 |
+
"eval_loss": 0.28249993920326233,
|
| 680 |
+
"eval_runtime": 29.95,
|
| 681 |
+
"eval_samples_per_second": 4.541,
|
| 682 |
+
"eval_steps_per_second": 0.167,
|
| 683 |
+
"step": 419
|
| 684 |
},
|
| 685 |
{
|
| 686 |
+
"epoch": 43.07692307692308,
|
| 687 |
+
"grad_norm": 1.4639347791671753,
|
| 688 |
"learning_rate": 3.7037037037037037e-06,
|
| 689 |
+
"loss": 0.1191,
|
| 690 |
+
"step": 420
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 691 |
},
|
| 692 |
{
|
| 693 |
+
"epoch": 44.0,
|
| 694 |
+
"eval_accuracy": 0.9264705882352942,
|
| 695 |
+
"eval_loss": 0.28346800804138184,
|
| 696 |
+
"eval_runtime": 32.4599,
|
| 697 |
+
"eval_samples_per_second": 4.19,
|
| 698 |
+
"eval_steps_per_second": 0.154,
|
| 699 |
+
"step": 429
|
| 700 |
+
},
|
| 701 |
+
{
|
| 702 |
+
"epoch": 44.1025641025641,
|
| 703 |
+
"grad_norm": 1.6088590621948242,
|
| 704 |
+
"learning_rate": 2.469135802469136e-06,
|
| 705 |
+
"loss": 0.1221,
|
| 706 |
+
"step": 430
|
| 707 |
},
|
| 708 |
{
|
| 709 |
+
"epoch": 44.92307692307692,
|
| 710 |
+
"eval_accuracy": 0.9191176470588235,
|
| 711 |
+
"eval_loss": 0.2723533809185028,
|
| 712 |
+
"eval_runtime": 30.5746,
|
| 713 |
+
"eval_samples_per_second": 4.448,
|
| 714 |
+
"eval_steps_per_second": 0.164,
|
| 715 |
+
"step": 438
|
| 716 |
},
|
| 717 |
{
|
| 718 |
+
"epoch": 45.12820512820513,
|
| 719 |
+
"grad_norm": 0.848240077495575,
|
| 720 |
+
"learning_rate": 1.234567901234568e-06,
|
| 721 |
+
"loss": 0.1151,
|
| 722 |
+
"step": 440
|
| 723 |
},
|
| 724 |
{
|
| 725 |
+
"epoch": 45.94871794871795,
|
| 726 |
+
"eval_accuracy": 0.9191176470588235,
|
| 727 |
+
"eval_loss": 0.27075088024139404,
|
| 728 |
+
"eval_runtime": 30.0665,
|
| 729 |
+
"eval_samples_per_second": 4.523,
|
| 730 |
+
"eval_steps_per_second": 0.166,
|
| 731 |
+
"step": 448
|
| 732 |
},
|
| 733 |
{
|
| 734 |
+
"epoch": 46.15384615384615,
|
| 735 |
+
"grad_norm": 3.371528148651123,
|
| 736 |
+
"learning_rate": 0.0,
|
| 737 |
+
"loss": 0.1195,
|
| 738 |
+
"step": 450
|
| 739 |
},
|
| 740 |
{
|
| 741 |
+
"epoch": 46.15384615384615,
|
| 742 |
+
"eval_accuracy": 0.9191176470588235,
|
| 743 |
+
"eval_loss": 0.2707464396953583,
|
| 744 |
+
"eval_runtime": 33.0815,
|
| 745 |
+
"eval_samples_per_second": 4.111,
|
| 746 |
+
"eval_steps_per_second": 0.151,
|
| 747 |
+
"step": 450
|
| 748 |
},
|
| 749 |
{
|
| 750 |
+
"epoch": 46.15384615384615,
|
| 751 |
+
"step": 450,
|
| 752 |
+
"total_flos": 4.3781443993328026e+18,
|
| 753 |
+
"train_loss": 0.4078153912226359,
|
| 754 |
+
"train_runtime": 14896.2203,
|
| 755 |
+
"train_samples_per_second": 4.108,
|
| 756 |
+
"train_steps_per_second": 0.03
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 757 |
}
|
| 758 |
],
|
| 759 |
"logging_steps": 10,
|
| 760 |
+
"max_steps": 450,
|
| 761 |
"num_input_tokens_seen": 0,
|
| 762 |
"num_train_epochs": 50,
|
| 763 |
"save_steps": 500,
|
|
|
|
| 773 |
"attributes": {}
|
| 774 |
}
|
| 775 |
},
|
| 776 |
+
"total_flos": 4.3781443993328026e+18,
|
| 777 |
"train_batch_size": 32,
|
| 778 |
"trial_name": null,
|
| 779 |
"trial_params": null
|