zireael08/swin-msldv2
Browse files- README.md +14 -14
- all_results.json +9 -9
- eval_results.json +5 -5
- model.safetensors +1 -1
- train_results.json +4 -4
- trainer_state.json +304 -304
- training_args.bin +1 -1
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 [microsoft/swinv2-tiny-patch4-window8-256](https://huggingface.co/microsoft/swinv2-tiny-patch4-window8-256) 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 |
|
|
@@ -53,7 +53,7 @@ More information needed
|
|
| 53 |
### Training hyperparameters
|
| 54 |
|
| 55 |
The following hyperparameters were used during training:
|
| 56 |
-
- learning_rate: 0.
|
| 57 |
- train_batch_size: 32
|
| 58 |
- eval_batch_size: 64
|
| 59 |
- seed: 42
|
|
@@ -66,16 +66,16 @@ The following hyperparameters were used during training:
|
|
| 66 |
|
| 67 |
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|
| 68 |
|:-------------:|:-----:|:----:|:---------------:|:--------:|
|
| 69 |
-
| 0.
|
| 70 |
-
| 0.
|
| 71 |
-
| 0.
|
| 72 |
-
| 0.
|
| 73 |
-
| 0.
|
| 74 |
-
| 0.
|
| 75 |
-
| 0.
|
| 76 |
-
| 0.
|
| 77 |
-
| 0.
|
| 78 |
-
| 0.
|
| 79 |
|
| 80 |
|
| 81 |
### Framework versions
|
|
|
|
| 23 |
metrics:
|
| 24 |
- name: Accuracy
|
| 25 |
type: accuracy
|
| 26 |
+
value: 0.9982425307557118
|
| 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 [microsoft/swinv2-tiny-patch4-window8-256](https://huggingface.co/microsoft/swinv2-tiny-patch4-window8-256) on the imagefolder dataset.
|
| 35 |
It achieves the following results on the evaluation set:
|
| 36 |
+
- Loss: 0.0098
|
| 37 |
+
- Accuracy: 0.9982
|
| 38 |
|
| 39 |
## Model description
|
| 40 |
|
|
|
|
| 53 |
### Training hyperparameters
|
| 54 |
|
| 55 |
The following hyperparameters were used during training:
|
| 56 |
+
- learning_rate: 0.0001
|
| 57 |
- train_batch_size: 32
|
| 58 |
- eval_batch_size: 64
|
| 59 |
- seed: 42
|
|
|
|
| 66 |
|
| 67 |
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|
| 68 |
|:-------------:|:-----:|:----:|:---------------:|:--------:|
|
| 69 |
+
| 0.6367 | 1.0 | 83 | 0.4612 | 0.8248 |
|
| 70 |
+
| 0.4656 | 2.0 | 166 | 0.3608 | 0.8496 |
|
| 71 |
+
| 0.4911 | 3.0 | 249 | 0.1344 | 0.9646 |
|
| 72 |
+
| 0.1630 | 4.0 | 332 | 0.1347 | 0.9575 |
|
| 73 |
+
| 0.1872 | 5.0 | 415 | 0.1106 | 0.9628 |
|
| 74 |
+
| 0.1801 | 6.0 | 498 | 0.0968 | 0.9823 |
|
| 75 |
+
| 0.1453 | 7.0 | 581 | 0.1196 | 0.9717 |
|
| 76 |
+
| 0.0787 | 8.0 | 664 | 0.0838 | 0.9894 |
|
| 77 |
+
| 0.0353 | 9.0 | 747 | 0.0801 | 0.9912 |
|
| 78 |
+
| 0.0878 | 10.0 | 830 | 0.0818 | 0.9912 |
|
| 79 |
|
| 80 |
|
| 81 |
### Framework versions
|
all_results.json
CHANGED
|
@@ -1,13 +1,13 @@
|
|
| 1 |
{
|
| 2 |
"epoch": 10.0,
|
| 3 |
-
"eval_accuracy": 0.
|
| 4 |
-
"eval_loss": 0.
|
| 5 |
-
"eval_runtime":
|
| 6 |
-
"eval_samples_per_second":
|
| 7 |
-
"eval_steps_per_second": 1.
|
| 8 |
"total_flos": 8.593274471605862e+17,
|
| 9 |
-
"train_loss": 0.
|
| 10 |
-
"train_runtime":
|
| 11 |
-
"train_samples_per_second":
|
| 12 |
-
"train_steps_per_second": 0.
|
| 13 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"epoch": 10.0,
|
| 3 |
+
"eval_accuracy": 0.9982425307557118,
|
| 4 |
+
"eval_loss": 0.009822744876146317,
|
| 5 |
+
"eval_runtime": 6.0857,
|
| 6 |
+
"eval_samples_per_second": 93.498,
|
| 7 |
+
"eval_steps_per_second": 1.479,
|
| 8 |
"total_flos": 8.593274471605862e+17,
|
| 9 |
+
"train_loss": 0.29410572172288435,
|
| 10 |
+
"train_runtime": 896.4382,
|
| 11 |
+
"train_samples_per_second": 29.461,
|
| 12 |
+
"train_steps_per_second": 0.926
|
| 13 |
}
|
eval_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
"epoch": 10.0,
|
| 3 |
-
"eval_accuracy": 0.
|
| 4 |
-
"eval_loss": 0.
|
| 5 |
-
"eval_runtime":
|
| 6 |
-
"eval_samples_per_second":
|
| 7 |
-
"eval_steps_per_second": 1.
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"epoch": 10.0,
|
| 3 |
+
"eval_accuracy": 0.9982425307557118,
|
| 4 |
+
"eval_loss": 0.009822744876146317,
|
| 5 |
+
"eval_runtime": 6.0857,
|
| 6 |
+
"eval_samples_per_second": 93.498,
|
| 7 |
+
"eval_steps_per_second": 1.479
|
| 8 |
}
|
model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 110362448
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:fe2399d36d2c712a58c79aa8eada79a504907fecd03646c12c6865d1884ee997
|
| 3 |
size 110362448
|
train_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
"epoch": 10.0,
|
| 3 |
"total_flos": 8.593274471605862e+17,
|
| 4 |
-
"train_loss": 0.
|
| 5 |
-
"train_runtime":
|
| 6 |
-
"train_samples_per_second":
|
| 7 |
-
"train_steps_per_second": 0.
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"epoch": 10.0,
|
| 3 |
"total_flos": 8.593274471605862e+17,
|
| 4 |
+
"train_loss": 0.29410572172288435,
|
| 5 |
+
"train_runtime": 896.4382,
|
| 6 |
+
"train_samples_per_second": 29.461,
|
| 7 |
+
"train_steps_per_second": 0.926
|
| 8 |
}
|
trainer_state.json
CHANGED
|
@@ -1,6 +1,6 @@
|
|
| 1 |
{
|
| 2 |
"best_global_step": 747,
|
| 3 |
-
"best_metric": 0.
|
| 4 |
"best_model_checkpoint": "./logs/checkpoint-747",
|
| 5 |
"epoch": 10.0,
|
| 6 |
"eval_steps": 500,
|
|
@@ -11,683 +11,683 @@
|
|
| 11 |
"log_history": [
|
| 12 |
{
|
| 13 |
"epoch": 0.12048192771084337,
|
| 14 |
-
"grad_norm": 6.
|
| 15 |
-
"learning_rate":
|
| 16 |
-
"loss": 1.
|
| 17 |
"step": 10
|
| 18 |
},
|
| 19 |
{
|
| 20 |
"epoch": 0.24096385542168675,
|
| 21 |
-
"grad_norm":
|
| 22 |
-
"learning_rate":
|
| 23 |
-
"loss": 1.
|
| 24 |
"step": 20
|
| 25 |
},
|
| 26 |
{
|
| 27 |
"epoch": 0.3614457831325301,
|
| 28 |
-
"grad_norm":
|
| 29 |
-
"learning_rate":
|
| 30 |
-
"loss": 1.
|
| 31 |
"step": 30
|
| 32 |
},
|
| 33 |
{
|
| 34 |
"epoch": 0.4819277108433735,
|
| 35 |
-
"grad_norm":
|
| 36 |
-
"learning_rate":
|
| 37 |
-
"loss":
|
| 38 |
"step": 40
|
| 39 |
},
|
| 40 |
{
|
| 41 |
"epoch": 0.6024096385542169,
|
| 42 |
-
"grad_norm":
|
| 43 |
-
"learning_rate":
|
| 44 |
-
"loss": 0.
|
| 45 |
"step": 50
|
| 46 |
},
|
| 47 |
{
|
| 48 |
"epoch": 0.7228915662650602,
|
| 49 |
-
"grad_norm":
|
| 50 |
-
"learning_rate":
|
| 51 |
-
"loss": 0.
|
| 52 |
"step": 60
|
| 53 |
},
|
| 54 |
{
|
| 55 |
"epoch": 0.8433734939759037,
|
| 56 |
-
"grad_norm": 9.
|
| 57 |
-
"learning_rate":
|
| 58 |
-
"loss": 0.
|
| 59 |
"step": 70
|
| 60 |
},
|
| 61 |
{
|
| 62 |
"epoch": 0.963855421686747,
|
| 63 |
-
"grad_norm":
|
| 64 |
-
"learning_rate":
|
| 65 |
-
"loss": 0.
|
| 66 |
"step": 80
|
| 67 |
},
|
| 68 |
{
|
| 69 |
"epoch": 1.0,
|
| 70 |
-
"eval_accuracy": 0.
|
| 71 |
-
"eval_loss": 0.
|
| 72 |
-
"eval_runtime": 5.
|
| 73 |
-
"eval_samples_per_second": 103.
|
| 74 |
-
"eval_steps_per_second": 1.
|
| 75 |
"step": 83
|
| 76 |
},
|
| 77 |
{
|
| 78 |
"epoch": 1.0843373493975903,
|
| 79 |
-
"grad_norm":
|
| 80 |
-
"learning_rate":
|
| 81 |
-
"loss": 0.
|
| 82 |
"step": 90
|
| 83 |
},
|
| 84 |
{
|
| 85 |
"epoch": 1.2048192771084336,
|
| 86 |
-
"grad_norm":
|
| 87 |
-
"learning_rate":
|
| 88 |
-
"loss": 0.
|
| 89 |
"step": 100
|
| 90 |
},
|
| 91 |
{
|
| 92 |
"epoch": 1.3253012048192772,
|
| 93 |
-
"grad_norm":
|
| 94 |
-
"learning_rate":
|
| 95 |
-
"loss": 0.
|
| 96 |
"step": 110
|
| 97 |
},
|
| 98 |
{
|
| 99 |
"epoch": 1.4457831325301205,
|
| 100 |
-
"grad_norm":
|
| 101 |
-
"learning_rate":
|
| 102 |
-
"loss": 0.
|
| 103 |
"step": 120
|
| 104 |
},
|
| 105 |
{
|
| 106 |
"epoch": 1.5662650602409638,
|
| 107 |
-
"grad_norm":
|
| 108 |
-
"learning_rate":
|
| 109 |
-
"loss": 0.
|
| 110 |
"step": 130
|
| 111 |
},
|
| 112 |
{
|
| 113 |
"epoch": 1.6867469879518073,
|
| 114 |
-
"grad_norm": 9.
|
| 115 |
-
"learning_rate":
|
| 116 |
-
"loss": 0.
|
| 117 |
"step": 140
|
| 118 |
},
|
| 119 |
{
|
| 120 |
"epoch": 1.8072289156626506,
|
| 121 |
-
"grad_norm":
|
| 122 |
-
"learning_rate":
|
| 123 |
-
"loss": 0.
|
| 124 |
"step": 150
|
| 125 |
},
|
| 126 |
{
|
| 127 |
"epoch": 1.927710843373494,
|
| 128 |
-
"grad_norm": 8.
|
| 129 |
-
"learning_rate":
|
| 130 |
-
"loss": 0.
|
| 131 |
"step": 160
|
| 132 |
},
|
| 133 |
{
|
| 134 |
"epoch": 2.0,
|
| 135 |
-
"eval_accuracy": 0.
|
| 136 |
-
"eval_loss": 0.
|
| 137 |
-
"eval_runtime": 5.
|
| 138 |
-
"eval_samples_per_second":
|
| 139 |
-
"eval_steps_per_second": 1.
|
| 140 |
"step": 166
|
| 141 |
},
|
| 142 |
{
|
| 143 |
"epoch": 2.0481927710843375,
|
| 144 |
-
"grad_norm":
|
| 145 |
-
"learning_rate":
|
| 146 |
-
"loss": 0.
|
| 147 |
"step": 170
|
| 148 |
},
|
| 149 |
{
|
| 150 |
"epoch": 2.1686746987951806,
|
| 151 |
-
"grad_norm":
|
| 152 |
-
"learning_rate":
|
| 153 |
-
"loss": 0.
|
| 154 |
"step": 180
|
| 155 |
},
|
| 156 |
{
|
| 157 |
"epoch": 2.289156626506024,
|
| 158 |
-
"grad_norm":
|
| 159 |
-
"learning_rate":
|
| 160 |
-
"loss": 0.
|
| 161 |
"step": 190
|
| 162 |
},
|
| 163 |
{
|
| 164 |
"epoch": 2.4096385542168672,
|
| 165 |
-
"grad_norm":
|
| 166 |
-
"learning_rate":
|
| 167 |
-
"loss": 0.
|
| 168 |
"step": 200
|
| 169 |
},
|
| 170 |
{
|
| 171 |
"epoch": 2.5301204819277108,
|
| 172 |
-
"grad_norm":
|
| 173 |
-
"learning_rate":
|
| 174 |
-
"loss": 0.
|
| 175 |
"step": 210
|
| 176 |
},
|
| 177 |
{
|
| 178 |
"epoch": 2.6506024096385543,
|
| 179 |
-
"grad_norm":
|
| 180 |
-
"learning_rate":
|
| 181 |
-
"loss": 0.
|
| 182 |
"step": 220
|
| 183 |
},
|
| 184 |
{
|
| 185 |
"epoch": 2.7710843373493974,
|
| 186 |
-
"grad_norm":
|
| 187 |
-
"learning_rate":
|
| 188 |
-
"loss": 0.
|
| 189 |
"step": 230
|
| 190 |
},
|
| 191 |
{
|
| 192 |
"epoch": 2.891566265060241,
|
| 193 |
-
"grad_norm":
|
| 194 |
-
"learning_rate":
|
| 195 |
-
"loss": 0.
|
| 196 |
"step": 240
|
| 197 |
},
|
| 198 |
{
|
| 199 |
"epoch": 3.0,
|
| 200 |
-
"eval_accuracy": 0.
|
| 201 |
-
"eval_loss": 0.
|
| 202 |
-
"eval_runtime":
|
| 203 |
-
"eval_samples_per_second":
|
| 204 |
-
"eval_steps_per_second": 1.
|
| 205 |
"step": 249
|
| 206 |
},
|
| 207 |
{
|
| 208 |
"epoch": 3.0120481927710845,
|
| 209 |
-
"grad_norm":
|
| 210 |
-
"learning_rate":
|
| 211 |
-
"loss": 0.
|
| 212 |
"step": 250
|
| 213 |
},
|
| 214 |
{
|
| 215 |
"epoch": 3.1325301204819276,
|
| 216 |
-
"grad_norm":
|
| 217 |
-
"learning_rate":
|
| 218 |
-
"loss": 0.
|
| 219 |
"step": 260
|
| 220 |
},
|
| 221 |
{
|
| 222 |
"epoch": 3.253012048192771,
|
| 223 |
-
"grad_norm":
|
| 224 |
-
"learning_rate":
|
| 225 |
-
"loss": 0.
|
| 226 |
"step": 270
|
| 227 |
},
|
| 228 |
{
|
| 229 |
"epoch": 3.3734939759036147,
|
| 230 |
-
"grad_norm":
|
| 231 |
-
"learning_rate":
|
| 232 |
-
"loss": 0.
|
| 233 |
"step": 280
|
| 234 |
},
|
| 235 |
{
|
| 236 |
"epoch": 3.4939759036144578,
|
| 237 |
-
"grad_norm":
|
| 238 |
-
"learning_rate":
|
| 239 |
-
"loss": 0.
|
| 240 |
"step": 290
|
| 241 |
},
|
| 242 |
{
|
| 243 |
"epoch": 3.6144578313253013,
|
| 244 |
-
"grad_norm":
|
| 245 |
-
"learning_rate":
|
| 246 |
-
"loss": 0.
|
| 247 |
"step": 300
|
| 248 |
},
|
| 249 |
{
|
| 250 |
"epoch": 3.734939759036145,
|
| 251 |
-
"grad_norm":
|
| 252 |
-
"learning_rate":
|
| 253 |
-
"loss": 0.
|
| 254 |
"step": 310
|
| 255 |
},
|
| 256 |
{
|
| 257 |
"epoch": 3.855421686746988,
|
| 258 |
-
"grad_norm":
|
| 259 |
-
"learning_rate":
|
| 260 |
-
"loss": 0.
|
| 261 |
"step": 320
|
| 262 |
},
|
| 263 |
{
|
| 264 |
"epoch": 3.9759036144578315,
|
| 265 |
-
"grad_norm":
|
| 266 |
-
"learning_rate":
|
| 267 |
-
"loss": 0.
|
| 268 |
"step": 330
|
| 269 |
},
|
| 270 |
{
|
| 271 |
"epoch": 4.0,
|
| 272 |
-
"eval_accuracy": 0.
|
| 273 |
-
"eval_loss": 0.
|
| 274 |
-
"eval_runtime": 5.
|
| 275 |
-
"eval_samples_per_second":
|
| 276 |
-
"eval_steps_per_second": 1.
|
| 277 |
"step": 332
|
| 278 |
},
|
| 279 |
{
|
| 280 |
"epoch": 4.096385542168675,
|
| 281 |
-
"grad_norm":
|
| 282 |
-
"learning_rate":
|
| 283 |
-
"loss": 0.
|
| 284 |
"step": 340
|
| 285 |
},
|
| 286 |
{
|
| 287 |
"epoch": 4.216867469879518,
|
| 288 |
-
"grad_norm":
|
| 289 |
-
"learning_rate":
|
| 290 |
-
"loss": 0.
|
| 291 |
"step": 350
|
| 292 |
},
|
| 293 |
{
|
| 294 |
"epoch": 4.337349397590361,
|
| 295 |
-
"grad_norm":
|
| 296 |
-
"learning_rate":
|
| 297 |
-
"loss": 0.
|
| 298 |
"step": 360
|
| 299 |
},
|
| 300 |
{
|
| 301 |
"epoch": 4.457831325301205,
|
| 302 |
-
"grad_norm":
|
| 303 |
-
"learning_rate":
|
| 304 |
-
"loss": 0.
|
| 305 |
"step": 370
|
| 306 |
},
|
| 307 |
{
|
| 308 |
"epoch": 4.578313253012048,
|
| 309 |
-
"grad_norm":
|
| 310 |
-
"learning_rate":
|
| 311 |
-
"loss": 0.
|
| 312 |
"step": 380
|
| 313 |
},
|
| 314 |
{
|
| 315 |
"epoch": 4.698795180722891,
|
| 316 |
-
"grad_norm":
|
| 317 |
-
"learning_rate":
|
| 318 |
-
"loss": 0.
|
| 319 |
"step": 390
|
| 320 |
},
|
| 321 |
{
|
| 322 |
"epoch": 4.8192771084337345,
|
| 323 |
-
"grad_norm": 9.
|
| 324 |
-
"learning_rate":
|
| 325 |
-
"loss": 0.
|
| 326 |
"step": 400
|
| 327 |
},
|
| 328 |
{
|
| 329 |
"epoch": 4.9397590361445785,
|
| 330 |
-
"grad_norm": 8.
|
| 331 |
-
"learning_rate":
|
| 332 |
-
"loss": 0.
|
| 333 |
"step": 410
|
| 334 |
},
|
| 335 |
{
|
| 336 |
"epoch": 5.0,
|
| 337 |
-
"eval_accuracy": 0.
|
| 338 |
-
"eval_loss": 0.
|
| 339 |
-
"eval_runtime": 5.
|
| 340 |
-
"eval_samples_per_second":
|
| 341 |
-
"eval_steps_per_second": 1.
|
| 342 |
"step": 415
|
| 343 |
},
|
| 344 |
{
|
| 345 |
"epoch": 5.0602409638554215,
|
| 346 |
-
"grad_norm":
|
| 347 |
-
"learning_rate":
|
| 348 |
-
"loss": 0.
|
| 349 |
"step": 420
|
| 350 |
},
|
| 351 |
{
|
| 352 |
"epoch": 5.180722891566265,
|
| 353 |
-
"grad_norm":
|
| 354 |
-
"learning_rate":
|
| 355 |
-
"loss": 0.
|
| 356 |
"step": 430
|
| 357 |
},
|
| 358 |
{
|
| 359 |
"epoch": 5.301204819277109,
|
| 360 |
-
"grad_norm":
|
| 361 |
-
"learning_rate":
|
| 362 |
-
"loss": 0.
|
| 363 |
"step": 440
|
| 364 |
},
|
| 365 |
{
|
| 366 |
"epoch": 5.421686746987952,
|
| 367 |
-
"grad_norm":
|
| 368 |
-
"learning_rate":
|
| 369 |
-
"loss": 0.
|
| 370 |
"step": 450
|
| 371 |
},
|
| 372 |
{
|
| 373 |
"epoch": 5.542168674698795,
|
| 374 |
-
"grad_norm":
|
| 375 |
-
"learning_rate":
|
| 376 |
-
"loss": 0.
|
| 377 |
"step": 460
|
| 378 |
},
|
| 379 |
{
|
| 380 |
"epoch": 5.662650602409639,
|
| 381 |
-
"grad_norm":
|
| 382 |
-
"learning_rate":
|
| 383 |
-
"loss": 0.
|
| 384 |
"step": 470
|
| 385 |
},
|
| 386 |
{
|
| 387 |
"epoch": 5.783132530120482,
|
| 388 |
-
"grad_norm":
|
| 389 |
-
"learning_rate":
|
| 390 |
-
"loss": 0.
|
| 391 |
"step": 480
|
| 392 |
},
|
| 393 |
{
|
| 394 |
"epoch": 5.903614457831325,
|
| 395 |
-
"grad_norm":
|
| 396 |
-
"learning_rate":
|
| 397 |
-
"loss": 0.
|
| 398 |
"step": 490
|
| 399 |
},
|
| 400 |
{
|
| 401 |
"epoch": 6.0,
|
| 402 |
-
"eval_accuracy": 0.
|
| 403 |
-
"eval_loss": 0.
|
| 404 |
-
"eval_runtime":
|
| 405 |
-
"eval_samples_per_second":
|
| 406 |
-
"eval_steps_per_second": 1.
|
| 407 |
"step": 498
|
| 408 |
},
|
| 409 |
{
|
| 410 |
"epoch": 6.024096385542169,
|
| 411 |
-
"grad_norm":
|
| 412 |
-
"learning_rate":
|
| 413 |
-
"loss": 0.
|
| 414 |
"step": 500
|
| 415 |
},
|
| 416 |
{
|
| 417 |
"epoch": 6.144578313253012,
|
| 418 |
-
"grad_norm":
|
| 419 |
-
"learning_rate":
|
| 420 |
-
"loss": 0.
|
| 421 |
"step": 510
|
| 422 |
},
|
| 423 |
{
|
| 424 |
"epoch": 6.265060240963855,
|
| 425 |
-
"grad_norm": 3.
|
| 426 |
-
"learning_rate":
|
| 427 |
-
"loss": 0.
|
| 428 |
"step": 520
|
| 429 |
},
|
| 430 |
{
|
| 431 |
"epoch": 6.385542168674699,
|
| 432 |
-
"grad_norm": 4.
|
| 433 |
-
"learning_rate":
|
| 434 |
-
"loss": 0.
|
| 435 |
"step": 530
|
| 436 |
},
|
| 437 |
{
|
| 438 |
"epoch": 6.506024096385542,
|
| 439 |
-
"grad_norm":
|
| 440 |
-
"learning_rate":
|
| 441 |
-
"loss": 0.
|
| 442 |
"step": 540
|
| 443 |
},
|
| 444 |
{
|
| 445 |
"epoch": 6.626506024096385,
|
| 446 |
-
"grad_norm": 1.
|
| 447 |
-
"learning_rate":
|
| 448 |
-
"loss": 0.
|
| 449 |
"step": 550
|
| 450 |
},
|
| 451 |
{
|
| 452 |
"epoch": 6.746987951807229,
|
| 453 |
-
"grad_norm": 8.
|
| 454 |
-
"learning_rate":
|
| 455 |
-
"loss": 0.
|
| 456 |
"step": 560
|
| 457 |
},
|
| 458 |
{
|
| 459 |
"epoch": 6.867469879518072,
|
| 460 |
-
"grad_norm": 5.
|
| 461 |
-
"learning_rate":
|
| 462 |
-
"loss": 0.
|
| 463 |
"step": 570
|
| 464 |
},
|
| 465 |
{
|
| 466 |
"epoch": 6.9879518072289155,
|
| 467 |
-
"grad_norm":
|
| 468 |
-
"learning_rate":
|
| 469 |
-
"loss": 0.
|
| 470 |
"step": 580
|
| 471 |
},
|
| 472 |
{
|
| 473 |
"epoch": 7.0,
|
| 474 |
-
"eval_accuracy": 0.
|
| 475 |
-
"eval_loss": 0.
|
| 476 |
-
"eval_runtime": 5.
|
| 477 |
-
"eval_samples_per_second":
|
| 478 |
-
"eval_steps_per_second": 1.
|
| 479 |
"step": 581
|
| 480 |
},
|
| 481 |
{
|
| 482 |
"epoch": 7.108433734939759,
|
| 483 |
-
"grad_norm":
|
| 484 |
-
"learning_rate":
|
| 485 |
-
"loss": 0.
|
| 486 |
"step": 590
|
| 487 |
},
|
| 488 |
{
|
| 489 |
"epoch": 7.228915662650603,
|
| 490 |
-
"grad_norm":
|
| 491 |
-
"learning_rate":
|
| 492 |
-
"loss": 0.
|
| 493 |
"step": 600
|
| 494 |
},
|
| 495 |
{
|
| 496 |
"epoch": 7.349397590361446,
|
| 497 |
-
"grad_norm":
|
| 498 |
-
"learning_rate":
|
| 499 |
-
"loss": 0.
|
| 500 |
"step": 610
|
| 501 |
},
|
| 502 |
{
|
| 503 |
"epoch": 7.469879518072289,
|
| 504 |
-
"grad_norm":
|
| 505 |
-
"learning_rate":
|
| 506 |
-
"loss": 0.
|
| 507 |
"step": 620
|
| 508 |
},
|
| 509 |
{
|
| 510 |
"epoch": 7.590361445783133,
|
| 511 |
-
"grad_norm":
|
| 512 |
-
"learning_rate":
|
| 513 |
-
"loss": 0.
|
| 514 |
"step": 630
|
| 515 |
},
|
| 516 |
{
|
| 517 |
"epoch": 7.710843373493976,
|
| 518 |
-
"grad_norm":
|
| 519 |
-
"learning_rate":
|
| 520 |
-
"loss": 0.
|
| 521 |
"step": 640
|
| 522 |
},
|
| 523 |
{
|
| 524 |
"epoch": 7.831325301204819,
|
| 525 |
-
"grad_norm":
|
| 526 |
-
"learning_rate":
|
| 527 |
-
"loss": 0.
|
| 528 |
"step": 650
|
| 529 |
},
|
| 530 |
{
|
| 531 |
"epoch": 7.951807228915663,
|
| 532 |
-
"grad_norm":
|
| 533 |
-
"learning_rate":
|
| 534 |
-
"loss": 0.
|
| 535 |
"step": 660
|
| 536 |
},
|
| 537 |
{
|
| 538 |
"epoch": 8.0,
|
| 539 |
-
"eval_accuracy": 0.
|
| 540 |
-
"eval_loss": 0.
|
| 541 |
-
"eval_runtime": 5.
|
| 542 |
-
"eval_samples_per_second":
|
| 543 |
-
"eval_steps_per_second": 1.
|
| 544 |
"step": 664
|
| 545 |
},
|
| 546 |
{
|
| 547 |
"epoch": 8.072289156626505,
|
| 548 |
-
"grad_norm":
|
| 549 |
-
"learning_rate":
|
| 550 |
-
"loss": 0.
|
| 551 |
"step": 670
|
| 552 |
},
|
| 553 |
{
|
| 554 |
"epoch": 8.19277108433735,
|
| 555 |
-
"grad_norm":
|
| 556 |
-
"learning_rate":
|
| 557 |
-
"loss": 0.
|
| 558 |
"step": 680
|
| 559 |
},
|
| 560 |
{
|
| 561 |
"epoch": 8.313253012048193,
|
| 562 |
-
"grad_norm":
|
| 563 |
-
"learning_rate":
|
| 564 |
-
"loss": 0.
|
| 565 |
"step": 690
|
| 566 |
},
|
| 567 |
{
|
| 568 |
"epoch": 8.433734939759036,
|
| 569 |
-
"grad_norm":
|
| 570 |
-
"learning_rate":
|
| 571 |
-
"loss": 0.
|
| 572 |
"step": 700
|
| 573 |
},
|
| 574 |
{
|
| 575 |
"epoch": 8.55421686746988,
|
| 576 |
-
"grad_norm":
|
| 577 |
-
"learning_rate":
|
| 578 |
-
"loss": 0.
|
| 579 |
"step": 710
|
| 580 |
},
|
| 581 |
{
|
| 582 |
"epoch": 8.674698795180722,
|
| 583 |
-
"grad_norm":
|
| 584 |
-
"learning_rate":
|
| 585 |
-
"loss": 0.
|
| 586 |
"step": 720
|
| 587 |
},
|
| 588 |
{
|
| 589 |
"epoch": 8.795180722891565,
|
| 590 |
-
"grad_norm": 7.
|
| 591 |
-
"learning_rate":
|
| 592 |
-
"loss": 0.
|
| 593 |
"step": 730
|
| 594 |
},
|
| 595 |
{
|
| 596 |
"epoch": 8.91566265060241,
|
| 597 |
-
"grad_norm": 0.
|
| 598 |
-
"learning_rate":
|
| 599 |
-
"loss": 0.
|
| 600 |
"step": 740
|
| 601 |
},
|
| 602 |
{
|
| 603 |
"epoch": 9.0,
|
| 604 |
-
"eval_accuracy": 0.
|
| 605 |
-
"eval_loss": 0.
|
| 606 |
-
"eval_runtime":
|
| 607 |
-
"eval_samples_per_second":
|
| 608 |
-
"eval_steps_per_second": 1.
|
| 609 |
"step": 747
|
| 610 |
},
|
| 611 |
{
|
| 612 |
"epoch": 9.036144578313253,
|
| 613 |
-
"grad_norm":
|
| 614 |
-
"learning_rate":
|
| 615 |
-
"loss": 0.
|
| 616 |
"step": 750
|
| 617 |
},
|
| 618 |
{
|
| 619 |
"epoch": 9.156626506024097,
|
| 620 |
-
"grad_norm":
|
| 621 |
-
"learning_rate":
|
| 622 |
-
"loss": 0.
|
| 623 |
"step": 760
|
| 624 |
},
|
| 625 |
{
|
| 626 |
"epoch": 9.27710843373494,
|
| 627 |
-
"grad_norm":
|
| 628 |
-
"learning_rate":
|
| 629 |
-
"loss": 0.
|
| 630 |
"step": 770
|
| 631 |
},
|
| 632 |
{
|
| 633 |
"epoch": 9.397590361445783,
|
| 634 |
-
"grad_norm":
|
| 635 |
-
"learning_rate":
|
| 636 |
-
"loss": 0.
|
| 637 |
"step": 780
|
| 638 |
},
|
| 639 |
{
|
| 640 |
"epoch": 9.518072289156626,
|
| 641 |
-
"grad_norm":
|
| 642 |
-
"learning_rate":
|
| 643 |
-
"loss": 0.
|
| 644 |
"step": 790
|
| 645 |
},
|
| 646 |
{
|
| 647 |
"epoch": 9.638554216867469,
|
| 648 |
-
"grad_norm":
|
| 649 |
-
"learning_rate":
|
| 650 |
-
"loss": 0.
|
| 651 |
"step": 800
|
| 652 |
},
|
| 653 |
{
|
| 654 |
"epoch": 9.759036144578314,
|
| 655 |
-
"grad_norm":
|
| 656 |
-
"learning_rate":
|
| 657 |
-
"loss": 0.
|
| 658 |
"step": 810
|
| 659 |
},
|
| 660 |
{
|
| 661 |
"epoch": 9.879518072289157,
|
| 662 |
-
"grad_norm":
|
| 663 |
-
"learning_rate":
|
| 664 |
-
"loss": 0.
|
| 665 |
"step": 820
|
| 666 |
},
|
| 667 |
{
|
| 668 |
"epoch": 10.0,
|
| 669 |
-
"grad_norm": 0.
|
| 670 |
-
"learning_rate":
|
| 671 |
-
"loss": 0.
|
| 672 |
"step": 830
|
| 673 |
},
|
| 674 |
{
|
| 675 |
"epoch": 10.0,
|
| 676 |
-
"eval_accuracy": 0.
|
| 677 |
-
"eval_loss": 0.
|
| 678 |
-
"eval_runtime": 5.
|
| 679 |
-
"eval_samples_per_second":
|
| 680 |
-
"eval_steps_per_second": 1.
|
| 681 |
"step": 830
|
| 682 |
},
|
| 683 |
{
|
| 684 |
"epoch": 10.0,
|
| 685 |
"step": 830,
|
| 686 |
"total_flos": 8.593274471605862e+17,
|
| 687 |
-
"train_loss": 0.
|
| 688 |
-
"train_runtime":
|
| 689 |
-
"train_samples_per_second":
|
| 690 |
-
"train_steps_per_second": 0.
|
| 691 |
}
|
| 692 |
],
|
| 693 |
"logging_steps": 10,
|
|
|
|
| 1 |
{
|
| 2 |
"best_global_step": 747,
|
| 3 |
+
"best_metric": 0.9911504424778761,
|
| 4 |
"best_model_checkpoint": "./logs/checkpoint-747",
|
| 5 |
"epoch": 10.0,
|
| 6 |
"eval_steps": 500,
|
|
|
|
| 11 |
"log_history": [
|
| 12 |
{
|
| 13 |
"epoch": 0.12048192771084337,
|
| 14 |
+
"grad_norm": 6.236629486083984,
|
| 15 |
+
"learning_rate": 1.0843373493975904e-05,
|
| 16 |
+
"loss": 1.7779657363891601,
|
| 17 |
"step": 10
|
| 18 |
},
|
| 19 |
{
|
| 20 |
"epoch": 0.24096385542168675,
|
| 21 |
+
"grad_norm": 6.614838123321533,
|
| 22 |
+
"learning_rate": 2.289156626506024e-05,
|
| 23 |
+
"loss": 1.6157239913940429,
|
| 24 |
"step": 20
|
| 25 |
},
|
| 26 |
{
|
| 27 |
"epoch": 0.3614457831325301,
|
| 28 |
+
"grad_norm": 8.87565803527832,
|
| 29 |
+
"learning_rate": 3.4939759036144585e-05,
|
| 30 |
+
"loss": 1.3881938934326172,
|
| 31 |
"step": 30
|
| 32 |
},
|
| 33 |
{
|
| 34 |
"epoch": 0.4819277108433735,
|
| 35 |
+
"grad_norm": 8.774048805236816,
|
| 36 |
+
"learning_rate": 4.698795180722892e-05,
|
| 37 |
+
"loss": 1.0548779487609863,
|
| 38 |
"step": 40
|
| 39 |
},
|
| 40 |
{
|
| 41 |
"epoch": 0.6024096385542169,
|
| 42 |
+
"grad_norm": 10.919977188110352,
|
| 43 |
+
"learning_rate": 5.903614457831326e-05,
|
| 44 |
+
"loss": 0.860891056060791,
|
| 45 |
"step": 50
|
| 46 |
},
|
| 47 |
{
|
| 48 |
"epoch": 0.7228915662650602,
|
| 49 |
+
"grad_norm": 17.067983627319336,
|
| 50 |
+
"learning_rate": 7.108433734939759e-05,
|
| 51 |
+
"loss": 0.7981919765472412,
|
| 52 |
"step": 60
|
| 53 |
},
|
| 54 |
{
|
| 55 |
"epoch": 0.8433734939759037,
|
| 56 |
+
"grad_norm": 9.766481399536133,
|
| 57 |
+
"learning_rate": 8.313253012048194e-05,
|
| 58 |
+
"loss": 0.7360480785369873,
|
| 59 |
"step": 70
|
| 60 |
},
|
| 61 |
{
|
| 62 |
"epoch": 0.963855421686747,
|
| 63 |
+
"grad_norm": 11.55764102935791,
|
| 64 |
+
"learning_rate": 9.518072289156626e-05,
|
| 65 |
+
"loss": 0.636728572845459,
|
| 66 |
"step": 80
|
| 67 |
},
|
| 68 |
{
|
| 69 |
"epoch": 1.0,
|
| 70 |
+
"eval_accuracy": 0.8247787610619469,
|
| 71 |
+
"eval_loss": 0.46115025877952576,
|
| 72 |
+
"eval_runtime": 5.444,
|
| 73 |
+
"eval_samples_per_second": 103.785,
|
| 74 |
+
"eval_steps_per_second": 1.653,
|
| 75 |
"step": 83
|
| 76 |
},
|
| 77 |
{
|
| 78 |
"epoch": 1.0843373493975903,
|
| 79 |
+
"grad_norm": 13.059717178344727,
|
| 80 |
+
"learning_rate": 9.998408238461338e-05,
|
| 81 |
+
"loss": 0.5606242656707764,
|
| 82 |
"step": 90
|
| 83 |
},
|
| 84 |
{
|
| 85 |
"epoch": 1.2048192771084336,
|
| 86 |
+
"grad_norm": 7.8438615798950195,
|
| 87 |
+
"learning_rate": 9.988684476816419e-05,
|
| 88 |
+
"loss": 0.4747779369354248,
|
| 89 |
"step": 100
|
| 90 |
},
|
| 91 |
{
|
| 92 |
"epoch": 1.3253012048192772,
|
| 93 |
+
"grad_norm": 17.873554229736328,
|
| 94 |
+
"learning_rate": 9.970138440692705e-05,
|
| 95 |
+
"loss": 0.513523006439209,
|
| 96 |
"step": 110
|
| 97 |
},
|
| 98 |
{
|
| 99 |
"epoch": 1.4457831325301205,
|
| 100 |
+
"grad_norm": 11.972646713256836,
|
| 101 |
+
"learning_rate": 9.942802927959443e-05,
|
| 102 |
+
"loss": 0.5974394321441651,
|
| 103 |
"step": 120
|
| 104 |
},
|
| 105 |
{
|
| 106 |
"epoch": 1.5662650602409638,
|
| 107 |
+
"grad_norm": 16.943544387817383,
|
| 108 |
+
"learning_rate": 9.906726280298186e-05,
|
| 109 |
+
"loss": 0.452280855178833,
|
| 110 |
"step": 130
|
| 111 |
},
|
| 112 |
{
|
| 113 |
"epoch": 1.6867469879518073,
|
| 114 |
+
"grad_norm": 9.1749906539917,
|
| 115 |
+
"learning_rate": 9.861972297712605e-05,
|
| 116 |
+
"loss": 0.4720293045043945,
|
| 117 |
"step": 140
|
| 118 |
},
|
| 119 |
{
|
| 120 |
"epoch": 1.8072289156626506,
|
| 121 |
+
"grad_norm": 11.010215759277344,
|
| 122 |
+
"learning_rate": 9.808620125700925e-05,
|
| 123 |
+
"loss": 0.4886914253234863,
|
| 124 |
"step": 150
|
| 125 |
},
|
| 126 |
{
|
| 127 |
"epoch": 1.927710843373494,
|
| 128 |
+
"grad_norm": 8.292040824890137,
|
| 129 |
+
"learning_rate": 9.746764115290496e-05,
|
| 130 |
+
"loss": 0.46558895111083987,
|
| 131 |
"step": 160
|
| 132 |
},
|
| 133 |
{
|
| 134 |
"epoch": 2.0,
|
| 135 |
+
"eval_accuracy": 0.8495575221238938,
|
| 136 |
+
"eval_loss": 0.3608015775680542,
|
| 137 |
+
"eval_runtime": 5.678,
|
| 138 |
+
"eval_samples_per_second": 99.507,
|
| 139 |
+
"eval_steps_per_second": 1.585,
|
| 140 |
"step": 166
|
| 141 |
},
|
| 142 |
{
|
| 143 |
"epoch": 2.0481927710843375,
|
| 144 |
+
"grad_norm": 17.205005645751953,
|
| 145 |
+
"learning_rate": 9.676513656182058e-05,
|
| 146 |
+
"loss": 0.36324758529663087,
|
| 147 |
"step": 170
|
| 148 |
},
|
| 149 |
{
|
| 150 |
"epoch": 2.1686746987951806,
|
| 151 |
+
"grad_norm": 11.862001419067383,
|
| 152 |
+
"learning_rate": 9.597992983298747e-05,
|
| 153 |
+
"loss": 0.4520224094390869,
|
| 154 |
"step": 180
|
| 155 |
},
|
| 156 |
{
|
| 157 |
"epoch": 2.289156626506024,
|
| 158 |
+
"grad_norm": 15.979503631591797,
|
| 159 |
+
"learning_rate": 9.511340957081958e-05,
|
| 160 |
+
"loss": 0.40279397964477537,
|
| 161 |
"step": 190
|
| 162 |
},
|
| 163 |
{
|
| 164 |
"epoch": 2.4096385542168672,
|
| 165 |
+
"grad_norm": 14.119962692260742,
|
| 166 |
+
"learning_rate": 9.416710817922615e-05,
|
| 167 |
+
"loss": 0.41336545944213865,
|
| 168 |
"step": 200
|
| 169 |
},
|
| 170 |
{
|
| 171 |
"epoch": 2.5301204819277108,
|
| 172 |
+
"grad_norm": 17.406816482543945,
|
| 173 |
+
"learning_rate": 9.314269915162114e-05,
|
| 174 |
+
"loss": 0.3019423961639404,
|
| 175 |
"step": 210
|
| 176 |
},
|
| 177 |
{
|
| 178 |
"epoch": 2.6506024096385543,
|
| 179 |
+
"grad_norm": 14.716191291809082,
|
| 180 |
+
"learning_rate": 9.204199411142196e-05,
|
| 181 |
+
"loss": 0.3748778820037842,
|
| 182 |
"step": 220
|
| 183 |
},
|
| 184 |
{
|
| 185 |
"epoch": 2.7710843373493974,
|
| 186 |
+
"grad_norm": 21.437185287475586,
|
| 187 |
+
"learning_rate": 9.086693960827105e-05,
|
| 188 |
+
"loss": 0.34201803207397463,
|
| 189 |
"step": 230
|
| 190 |
},
|
| 191 |
{
|
| 192 |
"epoch": 2.891566265060241,
|
| 193 |
+
"grad_norm": 10.061134338378906,
|
| 194 |
+
"learning_rate": 8.961961367564651e-05,
|
| 195 |
+
"loss": 0.49113874435424804,
|
| 196 |
"step": 240
|
| 197 |
},
|
| 198 |
{
|
| 199 |
"epoch": 3.0,
|
| 200 |
+
"eval_accuracy": 0.9646017699115044,
|
| 201 |
+
"eval_loss": 0.1344006061553955,
|
| 202 |
+
"eval_runtime": 6.1864,
|
| 203 |
+
"eval_samples_per_second": 91.329,
|
| 204 |
+
"eval_steps_per_second": 1.455,
|
| 205 |
"step": 249
|
| 206 |
},
|
| 207 |
{
|
| 208 |
"epoch": 3.0120481927710845,
|
| 209 |
+
"grad_norm": 12.825634956359863,
|
| 210 |
+
"learning_rate": 8.83022221559489e-05,
|
| 211 |
+
"loss": 0.27299160957336427,
|
| 212 |
"step": 250
|
| 213 |
},
|
| 214 |
{
|
| 215 |
"epoch": 3.1325301204819276,
|
| 216 |
+
"grad_norm": 10.527647972106934,
|
| 217 |
+
"learning_rate": 8.691709479956373e-05,
|
| 218 |
+
"loss": 0.24414093494415284,
|
| 219 |
"step": 260
|
| 220 |
},
|
| 221 |
{
|
| 222 |
"epoch": 3.253012048192771,
|
| 223 |
+
"grad_norm": 4.087771892547607,
|
| 224 |
+
"learning_rate": 8.546668114479768e-05,
|
| 225 |
+
"loss": 0.18104053735733033,
|
| 226 |
"step": 270
|
| 227 |
},
|
| 228 |
{
|
| 229 |
"epoch": 3.3734939759036147,
|
| 230 |
+
"grad_norm": 9.269667625427246,
|
| 231 |
+
"learning_rate": 8.395354618597533e-05,
|
| 232 |
+
"loss": 0.3339837551116943,
|
| 233 |
"step": 280
|
| 234 |
},
|
| 235 |
{
|
| 236 |
"epoch": 3.4939759036144578,
|
| 237 |
+
"grad_norm": 5.284663200378418,
|
| 238 |
+
"learning_rate": 8.238036583735673e-05,
|
| 239 |
+
"loss": 0.2400984764099121,
|
| 240 |
"step": 290
|
| 241 |
},
|
| 242 |
{
|
| 243 |
"epoch": 3.6144578313253013,
|
| 244 |
+
"grad_norm": 8.479917526245117,
|
| 245 |
+
"learning_rate": 8.074992220089769e-05,
|
| 246 |
+
"loss": 0.22019331455230712,
|
| 247 |
"step": 300
|
| 248 |
},
|
| 249 |
{
|
| 250 |
"epoch": 3.734939759036145,
|
| 251 |
+
"grad_norm": 9.128419876098633,
|
| 252 |
+
"learning_rate": 7.906509864622203e-05,
|
| 253 |
+
"loss": 0.25870823860168457,
|
| 254 |
"step": 310
|
| 255 |
},
|
| 256 |
{
|
| 257 |
"epoch": 3.855421686746988,
|
| 258 |
+
"grad_norm": 12.799089431762695,
|
| 259 |
+
"learning_rate": 7.73288747115059e-05,
|
| 260 |
+
"loss": 0.20678648948669434,
|
| 261 |
"step": 320
|
| 262 |
},
|
| 263 |
{
|
| 264 |
"epoch": 3.9759036144578315,
|
| 265 |
+
"grad_norm": 11.251450538635254,
|
| 266 |
+
"learning_rate": 7.554432083429253e-05,
|
| 267 |
+
"loss": 0.1629856824874878,
|
| 268 |
"step": 330
|
| 269 |
},
|
| 270 |
{
|
| 271 |
"epoch": 4.0,
|
| 272 |
+
"eval_accuracy": 0.9575221238938053,
|
| 273 |
+
"eval_loss": 0.13474561274051666,
|
| 274 |
+
"eval_runtime": 5.6362,
|
| 275 |
+
"eval_samples_per_second": 100.245,
|
| 276 |
+
"eval_steps_per_second": 1.597,
|
| 277 |
"step": 332
|
| 278 |
},
|
| 279 |
{
|
| 280 |
"epoch": 4.096385542168675,
|
| 281 |
+
"grad_norm": 19.876901626586914,
|
| 282 |
+
"learning_rate": 7.3714592921555e-05,
|
| 283 |
+
"loss": 0.24734578132629395,
|
| 284 |
"step": 340
|
| 285 |
},
|
| 286 |
{
|
| 287 |
"epoch": 4.216867469879518,
|
| 288 |
+
"grad_norm": 13.15965461730957,
|
| 289 |
+
"learning_rate": 7.184292676861024e-05,
|
| 290 |
+
"loss": 0.22765071392059327,
|
| 291 |
"step": 350
|
| 292 |
},
|
| 293 |
{
|
| 294 |
"epoch": 4.337349397590361,
|
| 295 |
+
"grad_norm": 12.12977123260498,
|
| 296 |
+
"learning_rate": 6.99326323367538e-05,
|
| 297 |
+
"loss": 0.17916421890258788,
|
| 298 |
"step": 360
|
| 299 |
},
|
| 300 |
{
|
| 301 |
"epoch": 4.457831325301205,
|
| 302 |
+
"grad_norm": 8.979646682739258,
|
| 303 |
+
"learning_rate": 6.798708789973527e-05,
|
| 304 |
+
"loss": 0.1901506304740906,
|
| 305 |
"step": 370
|
| 306 |
},
|
| 307 |
{
|
| 308 |
"epoch": 4.578313253012048,
|
| 309 |
+
"grad_norm": 5.592668056488037,
|
| 310 |
+
"learning_rate": 6.600973406942616e-05,
|
| 311 |
+
"loss": 0.22261853218078614,
|
| 312 |
"step": 380
|
| 313 |
},
|
| 314 |
{
|
| 315 |
"epoch": 4.698795180722891,
|
| 316 |
+
"grad_norm": 12.222548484802246,
|
| 317 |
+
"learning_rate": 6.400406771124536e-05,
|
| 318 |
+
"loss": 0.16046804189682007,
|
| 319 |
"step": 390
|
| 320 |
},
|
| 321 |
{
|
| 322 |
"epoch": 4.8192771084337345,
|
| 323 |
+
"grad_norm": 9.516422271728516,
|
| 324 |
+
"learning_rate": 6.197363576010264e-05,
|
| 325 |
+
"loss": 0.3090466022491455,
|
| 326 |
"step": 400
|
| 327 |
},
|
| 328 |
{
|
| 329 |
"epoch": 4.9397590361445785,
|
| 330 |
+
"grad_norm": 8.311286926269531,
|
| 331 |
+
"learning_rate": 5.992202894779649e-05,
|
| 332 |
+
"loss": 0.18722275495529175,
|
| 333 |
"step": 410
|
| 334 |
},
|
| 335 |
{
|
| 336 |
"epoch": 5.0,
|
| 337 |
+
"eval_accuracy": 0.9628318584070796,
|
| 338 |
+
"eval_loss": 0.11059214919805527,
|
| 339 |
+
"eval_runtime": 5.668,
|
| 340 |
+
"eval_samples_per_second": 99.682,
|
| 341 |
+
"eval_steps_per_second": 1.588,
|
| 342 |
"step": 415
|
| 343 |
},
|
| 344 |
{
|
| 345 |
"epoch": 5.0602409638554215,
|
| 346 |
+
"grad_norm": 6.544140338897705,
|
| 347 |
+
"learning_rate": 5.7852875452958954e-05,
|
| 348 |
+
"loss": 0.1725080966949463,
|
| 349 |
"step": 420
|
| 350 |
},
|
| 351 |
{
|
| 352 |
"epoch": 5.180722891566265,
|
| 353 |
+
"grad_norm": 7.241940975189209,
|
| 354 |
+
"learning_rate": 5.576983448477734e-05,
|
| 355 |
+
"loss": 0.2657145023345947,
|
| 356 |
"step": 430
|
| 357 |
},
|
| 358 |
{
|
| 359 |
"epoch": 5.301204819277109,
|
| 360 |
+
"grad_norm": 2.805722713470459,
|
| 361 |
+
"learning_rate": 5.3676589811839796e-05,
|
| 362 |
+
"loss": 0.16265145540237427,
|
| 363 |
"step": 440
|
| 364 |
},
|
| 365 |
{
|
| 366 |
"epoch": 5.421686746987952,
|
| 367 |
+
"grad_norm": 7.153483867645264,
|
| 368 |
+
"learning_rate": 5.157684324754858e-05,
|
| 369 |
+
"loss": 0.1511433720588684,
|
| 370 |
"step": 450
|
| 371 |
},
|
| 372 |
{
|
| 373 |
"epoch": 5.542168674698795,
|
| 374 |
+
"grad_norm": 3.1414175033569336,
|
| 375 |
+
"learning_rate": 4.9474308103621874e-05,
|
| 376 |
+
"loss": 0.15450478792190553,
|
| 377 |
"step": 460
|
| 378 |
},
|
| 379 |
{
|
| 380 |
"epoch": 5.662650602409639,
|
| 381 |
+
"grad_norm": 3.8960776329040527,
|
| 382 |
+
"learning_rate": 4.737270262326134e-05,
|
| 383 |
+
"loss": 0.13111191987991333,
|
| 384 |
"step": 470
|
| 385 |
},
|
| 386 |
{
|
| 387 |
"epoch": 5.783132530120482,
|
| 388 |
+
"grad_norm": 7.418442726135254,
|
| 389 |
+
"learning_rate": 4.527574340559844e-05,
|
| 390 |
+
"loss": 0.1539200186729431,
|
| 391 |
"step": 480
|
| 392 |
},
|
| 393 |
{
|
| 394 |
"epoch": 5.903614457831325,
|
| 395 |
+
"grad_norm": 8.860248565673828,
|
| 396 |
+
"learning_rate": 4.3187138833048456e-05,
|
| 397 |
+
"loss": 0.1801429271697998,
|
| 398 |
"step": 490
|
| 399 |
},
|
| 400 |
{
|
| 401 |
"epoch": 6.0,
|
| 402 |
+
"eval_accuracy": 0.9823008849557522,
|
| 403 |
+
"eval_loss": 0.09679495543241501,
|
| 404 |
+
"eval_runtime": 6.1076,
|
| 405 |
+
"eval_samples_per_second": 92.508,
|
| 406 |
+
"eval_steps_per_second": 1.474,
|
| 407 |
"step": 498
|
| 408 |
},
|
| 409 |
{
|
| 410 |
"epoch": 6.024096385542169,
|
| 411 |
+
"grad_norm": 11.467449188232422,
|
| 412 |
+
"learning_rate": 4.111058251319516e-05,
|
| 413 |
+
"loss": 0.11156998872756958,
|
| 414 |
"step": 500
|
| 415 |
},
|
| 416 |
{
|
| 417 |
"epoch": 6.144578313253012,
|
| 418 |
+
"grad_norm": 13.2984037399292,
|
| 419 |
+
"learning_rate": 3.904974674680436e-05,
|
| 420 |
+
"loss": 0.11730811595916749,
|
| 421 |
"step": 510
|
| 422 |
},
|
| 423 |
{
|
| 424 |
"epoch": 6.265060240963855,
|
| 425 |
+
"grad_norm": 3.543509006500244,
|
| 426 |
+
"learning_rate": 3.7008276033517396e-05,
|
| 427 |
+
"loss": 0.19998840093612671,
|
| 428 |
"step": 520
|
| 429 |
},
|
| 430 |
{
|
| 431 |
"epoch": 6.385542168674699,
|
| 432 |
+
"grad_norm": 4.334460735321045,
|
| 433 |
+
"learning_rate": 3.49897806267101e-05,
|
| 434 |
+
"loss": 0.09577634930610657,
|
| 435 |
"step": 530
|
| 436 |
},
|
| 437 |
{
|
| 438 |
"epoch": 6.506024096385542,
|
| 439 |
+
"grad_norm": 1.5698552131652832,
|
| 440 |
+
"learning_rate": 3.2997830148914314e-05,
|
| 441 |
+
"loss": 0.11064940690994263,
|
| 442 |
"step": 540
|
| 443 |
},
|
| 444 |
{
|
| 445 |
"epoch": 6.626506024096385,
|
| 446 |
+
"grad_norm": 1.4867029190063477,
|
| 447 |
+
"learning_rate": 3.103594727909385e-05,
|
| 448 |
+
"loss": 0.0978583574295044,
|
| 449 |
"step": 550
|
| 450 |
},
|
| 451 |
{
|
| 452 |
"epoch": 6.746987951807229,
|
| 453 |
+
"grad_norm": 8.762438774108887,
|
| 454 |
+
"learning_rate": 2.910760152293764e-05,
|
| 455 |
+
"loss": 0.08853105902671814,
|
| 456 |
"step": 560
|
| 457 |
},
|
| 458 |
{
|
| 459 |
"epoch": 6.867469879518072,
|
| 460 |
+
"grad_norm": 5.0525360107421875,
|
| 461 |
+
"learning_rate": 2.721620307718793e-05,
|
| 462 |
+
"loss": 0.13467444181442262,
|
| 463 |
"step": 570
|
| 464 |
},
|
| 465 |
{
|
| 466 |
"epoch": 6.9879518072289155,
|
| 467 |
+
"grad_norm": 13.927603721618652,
|
| 468 |
+
"learning_rate": 2.536509679885355e-05,
|
| 469 |
+
"loss": 0.14531443119049073,
|
| 470 |
"step": 580
|
| 471 |
},
|
| 472 |
{
|
| 473 |
"epoch": 7.0,
|
| 474 |
+
"eval_accuracy": 0.9716814159292035,
|
| 475 |
+
"eval_loss": 0.1196078509092331,
|
| 476 |
+
"eval_runtime": 5.7351,
|
| 477 |
+
"eval_samples_per_second": 98.517,
|
| 478 |
+
"eval_steps_per_second": 1.569,
|
| 479 |
"step": 581
|
| 480 |
},
|
| 481 |
{
|
| 482 |
"epoch": 7.108433734939759,
|
| 483 |
+
"grad_norm": 1.3813672065734863,
|
| 484 |
+
"learning_rate": 2.3557556289973838e-05,
|
| 485 |
+
"loss": 0.07141577005386353,
|
| 486 |
"step": 590
|
| 487 |
},
|
| 488 |
{
|
| 489 |
"epoch": 7.228915662650603,
|
| 490 |
+
"grad_norm": 11.30803108215332,
|
| 491 |
+
"learning_rate": 2.179677810839382e-05,
|
| 492 |
+
"loss": 0.08913902044296265,
|
| 493 |
"step": 600
|
| 494 |
},
|
| 495 |
{
|
| 496 |
"epoch": 7.349397590361446,
|
| 497 |
+
"grad_norm": 15.15943717956543,
|
| 498 |
+
"learning_rate": 2.0085876114788937e-05,
|
| 499 |
+
"loss": 0.1786208987236023,
|
| 500 |
"step": 610
|
| 501 |
},
|
| 502 |
{
|
| 503 |
"epoch": 7.469879518072289,
|
| 504 |
+
"grad_norm": 6.659374237060547,
|
| 505 |
+
"learning_rate": 1.8427875965935758e-05,
|
| 506 |
+
"loss": 0.05375434160232544,
|
| 507 |
"step": 620
|
| 508 |
},
|
| 509 |
{
|
| 510 |
"epoch": 7.590361445783133,
|
| 511 |
+
"grad_norm": 3.859622001647949,
|
| 512 |
+
"learning_rate": 1.682570976396811e-05,
|
| 513 |
+
"loss": 0.13732693195343018,
|
| 514 |
"step": 630
|
| 515 |
},
|
| 516 |
{
|
| 517 |
"epoch": 7.710843373493976,
|
| 518 |
+
"grad_norm": 4.593474388122559,
|
| 519 |
+
"learning_rate": 1.5282210871079926e-05,
|
| 520 |
+
"loss": 0.09100980162620545,
|
| 521 |
"step": 640
|
| 522 |
},
|
| 523 |
{
|
| 524 |
"epoch": 7.831325301204819,
|
| 525 |
+
"grad_norm": 13.193694114685059,
|
| 526 |
+
"learning_rate": 1.3800108898846021e-05,
|
| 527 |
+
"loss": 0.09656141400337219,
|
| 528 |
"step": 650
|
| 529 |
},
|
| 530 |
{
|
| 531 |
"epoch": 7.951807228915663,
|
| 532 |
+
"grad_norm": 13.730281829833984,
|
| 533 |
+
"learning_rate": 1.2382024881020937e-05,
|
| 534 |
+
"loss": 0.0786526083946228,
|
| 535 |
"step": 660
|
| 536 |
},
|
| 537 |
{
|
| 538 |
"epoch": 8.0,
|
| 539 |
+
"eval_accuracy": 0.9893805309734514,
|
| 540 |
+
"eval_loss": 0.08379530161619186,
|
| 541 |
+
"eval_runtime": 5.6656,
|
| 542 |
+
"eval_samples_per_second": 99.724,
|
| 543 |
+
"eval_steps_per_second": 1.589,
|
| 544 |
"step": 664
|
| 545 |
},
|
| 546 |
{
|
| 547 |
"epoch": 8.072289156626505,
|
| 548 |
+
"grad_norm": 1.4331773519515991,
|
| 549 |
+
"learning_rate": 1.1030466638353293e-05,
|
| 550 |
+
"loss": 0.0922305703163147,
|
| 551 |
"step": 670
|
| 552 |
},
|
| 553 |
{
|
| 554 |
"epoch": 8.19277108433735,
|
| 555 |
+
"grad_norm": 11.172012329101562,
|
| 556 |
+
"learning_rate": 9.747824343612338e-06,
|
| 557 |
+
"loss": 0.051563167572021486,
|
| 558 |
"step": 680
|
| 559 |
},
|
| 560 |
{
|
| 561 |
"epoch": 8.313253012048193,
|
| 562 |
+
"grad_norm": 9.389365196228027,
|
| 563 |
+
"learning_rate": 8.536366294669978e-06,
|
| 564 |
+
"loss": 0.0976746916770935,
|
| 565 |
"step": 690
|
| 566 |
},
|
| 567 |
{
|
| 568 |
"epoch": 8.433734939759036,
|
| 569 |
+
"grad_norm": 0.6641272902488708,
|
| 570 |
+
"learning_rate": 7.398234903113266e-06,
|
| 571 |
+
"loss": 0.07286246418952942,
|
| 572 |
"step": 700
|
| 573 |
},
|
| 574 |
{
|
| 575 |
"epoch": 8.55421686746988,
|
| 576 |
+
"grad_norm": 10.819772720336914,
|
| 577 |
+
"learning_rate": 6.335442905481442e-06,
|
| 578 |
+
"loss": 0.07259726524353027,
|
| 579 |
"step": 710
|
| 580 |
},
|
| 581 |
{
|
| 582 |
"epoch": 8.674698795180722,
|
| 583 |
+
"grad_norm": 0.8964389562606812,
|
| 584 |
+
"learning_rate": 5.349869803827717e-06,
|
| 585 |
+
"loss": 0.043448707461357115,
|
| 586 |
"step": 720
|
| 587 |
},
|
| 588 |
{
|
| 589 |
"epoch": 8.795180722891565,
|
| 590 |
+
"grad_norm": 7.601110935211182,
|
| 591 |
+
"learning_rate": 4.4432585419005076e-06,
|
| 592 |
+
"loss": 0.10902594327926636,
|
| 593 |
"step": 730
|
| 594 |
},
|
| 595 |
{
|
| 596 |
"epoch": 8.91566265060241,
|
| 597 |
+
"grad_norm": 0.9344149827957153,
|
| 598 |
+
"learning_rate": 3.6172124228221914e-06,
|
| 599 |
+
"loss": 0.03534201383590698,
|
| 600 |
"step": 740
|
| 601 |
},
|
| 602 |
{
|
| 603 |
"epoch": 9.0,
|
| 604 |
+
"eval_accuracy": 0.9911504424778761,
|
| 605 |
+
"eval_loss": 0.0801326259970665,
|
| 606 |
+
"eval_runtime": 6.1806,
|
| 607 |
+
"eval_samples_per_second": 91.415,
|
| 608 |
+
"eval_steps_per_second": 1.456,
|
| 609 |
"step": 747
|
| 610 |
},
|
| 611 |
{
|
| 612 |
"epoch": 9.036144578313253,
|
| 613 |
+
"grad_norm": 2.648090124130249,
|
| 614 |
+
"learning_rate": 2.8731922737163685e-06,
|
| 615 |
+
"loss": 0.024153660237789153,
|
| 616 |
"step": 750
|
| 617 |
},
|
| 618 |
{
|
| 619 |
"epoch": 9.156626506024097,
|
| 620 |
+
"grad_norm": 7.166019439697266,
|
| 621 |
+
"learning_rate": 2.212513862297649e-06,
|
| 622 |
+
"loss": 0.12096415758132935,
|
| 623 |
"step": 760
|
| 624 |
},
|
| 625 |
{
|
| 626 |
"epoch": 9.27710843373494,
|
| 627 |
+
"grad_norm": 9.207470893859863,
|
| 628 |
+
"learning_rate": 1.6363455699930419e-06,
|
| 629 |
+
"loss": 0.10932642221450806,
|
| 630 |
"step": 770
|
| 631 |
},
|
| 632 |
{
|
| 633 |
"epoch": 9.397590361445783,
|
| 634 |
+
"grad_norm": 2.5481743812561035,
|
| 635 |
+
"learning_rate": 1.145706325709389e-06,
|
| 636 |
+
"loss": 0.06989773511886596,
|
| 637 |
"step": 780
|
| 638 |
},
|
| 639 |
{
|
| 640 |
"epoch": 9.518072289156626,
|
| 641 |
+
"grad_norm": 12.317899703979492,
|
| 642 |
+
"learning_rate": 7.414638039014265e-07,
|
| 643 |
+
"loss": 0.08738085627555847,
|
| 644 |
"step": 790
|
| 645 |
},
|
| 646 |
{
|
| 647 |
"epoch": 9.638554216867469,
|
| 648 |
+
"grad_norm": 2.812633752822876,
|
| 649 |
+
"learning_rate": 4.2433289012662194e-07,
|
| 650 |
+
"loss": 0.04424922168254852,
|
| 651 |
"step": 800
|
| 652 |
},
|
| 653 |
{
|
| 654 |
"epoch": 9.759036144578314,
|
| 655 |
+
"grad_norm": 1.378448486328125,
|
| 656 |
+
"learning_rate": 1.9487441680084983e-07,
|
| 657 |
+
"loss": 0.07828723788261413,
|
| 658 |
"step": 810
|
| 659 |
},
|
| 660 |
{
|
| 661 |
"epoch": 9.879518072289157,
|
| 662 |
+
"grad_norm": 9.539847373962402,
|
| 663 |
+
"learning_rate": 5.3494171390228166e-08,
|
| 664 |
+
"loss": 0.06453937888145447,
|
| 665 |
"step": 820
|
| 666 |
},
|
| 667 |
{
|
| 668 |
"epoch": 10.0,
|
| 669 |
+
"grad_norm": 0.7179245948791504,
|
| 670 |
+
"learning_rate": 4.4217879344166103e-10,
|
| 671 |
+
"loss": 0.08780375719070435,
|
| 672 |
"step": 830
|
| 673 |
},
|
| 674 |
{
|
| 675 |
"epoch": 10.0,
|
| 676 |
+
"eval_accuracy": 0.9911504424778761,
|
| 677 |
+
"eval_loss": 0.08178059756755829,
|
| 678 |
+
"eval_runtime": 5.6512,
|
| 679 |
+
"eval_samples_per_second": 99.979,
|
| 680 |
+
"eval_steps_per_second": 1.593,
|
| 681 |
"step": 830
|
| 682 |
},
|
| 683 |
{
|
| 684 |
"epoch": 10.0,
|
| 685 |
"step": 830,
|
| 686 |
"total_flos": 8.593274471605862e+17,
|
| 687 |
+
"train_loss": 0.29410572172288435,
|
| 688 |
+
"train_runtime": 896.4382,
|
| 689 |
+
"train_samples_per_second": 29.461,
|
| 690 |
+
"train_steps_per_second": 0.926
|
| 691 |
}
|
| 692 |
],
|
| 693 |
"logging_steps": 10,
|
training_args.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 5201
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a59329c3833ceef25e02a3057cd6a7cff3d8905c772dabca522835eecbd1b79b
|
| 3 |
size 5201
|