🍻 cheers
Browse files- README.md +6 -5
- all_results.json +12 -12
- eval_results.json +8 -8
- runs/Mar27_22-12-30_ryanserver/events.out.tfevents.1711592832.ryanserver.19364.1 +3 -0
- train_results.json +5 -5
- trainer_state.json +387 -853
README.md
CHANGED
|
@@ -2,6 +2,7 @@
|
|
| 2 |
license: apache-2.0
|
| 3 |
base_model: google/vit-base-patch16-224-in21k
|
| 4 |
tags:
|
|
|
|
| 5 |
- generated_from_trainer
|
| 6 |
model-index:
|
| 7 |
- name: ryan_model3272024
|
|
@@ -13,12 +14,12 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 13 |
|
| 14 |
# ryan_model3272024
|
| 15 |
|
| 16 |
-
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
|
| 17 |
It achieves the following results on the evaluation set:
|
| 18 |
-
- Loss: 0.
|
| 19 |
-
- Na Accuracy: 0.
|
| 20 |
-
- Ordinal Accuracy: 0.
|
| 21 |
-
- Ordinal Mae: 0.
|
| 22 |
|
| 23 |
## Model description
|
| 24 |
|
|
|
|
| 2 |
license: apache-2.0
|
| 3 |
base_model: google/vit-base-patch16-224-in21k
|
| 4 |
tags:
|
| 5 |
+
- image-classification
|
| 6 |
- generated_from_trainer
|
| 7 |
model-index:
|
| 8 |
- name: ryan_model3272024
|
|
|
|
| 14 |
|
| 15 |
# ryan_model3272024
|
| 16 |
|
| 17 |
+
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 beans dataset.
|
| 18 |
It achieves the following results on the evaluation set:
|
| 19 |
+
- Loss: 0.3037
|
| 20 |
+
- Na Accuracy: 0.7297
|
| 21 |
+
- Ordinal Accuracy: 0.5285
|
| 22 |
+
- Ordinal Mae: 0.6723
|
| 23 |
|
| 24 |
## Model description
|
| 25 |
|
all_results.json
CHANGED
|
@@ -1,14 +1,14 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_loss": 0.
|
| 4 |
-
"eval_na_accuracy": 0.
|
| 5 |
-
"eval_ordinal_accuracy": 0.
|
| 6 |
-
"eval_ordinal_mae":
|
| 7 |
-
"eval_runtime":
|
| 8 |
-
"eval_samples_per_second":
|
| 9 |
-
"eval_steps_per_second": 3.
|
| 10 |
-
"train_loss": 0.
|
| 11 |
-
"train_runtime":
|
| 12 |
-
"train_samples_per_second":
|
| 13 |
-
"train_steps_per_second": 0.
|
| 14 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 1.46,
|
| 3 |
+
"eval_loss": 0.3037484288215637,
|
| 4 |
+
"eval_na_accuracy": 0.7297297120094299,
|
| 5 |
+
"eval_ordinal_accuracy": 0.5285171270370483,
|
| 6 |
+
"eval_ordinal_mae": 0.6723113059997559,
|
| 7 |
+
"eval_runtime": 11.1069,
|
| 8 |
+
"eval_samples_per_second": 27.01,
|
| 9 |
+
"eval_steps_per_second": 3.421,
|
| 10 |
+
"train_loss": 0.32911812565543436,
|
| 11 |
+
"train_runtime": 761.2747,
|
| 12 |
+
"train_samples_per_second": 15.763,
|
| 13 |
+
"train_steps_per_second": 0.988
|
| 14 |
}
|
eval_results.json
CHANGED
|
@@ -1,10 +1,10 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_loss": 0.
|
| 4 |
-
"eval_na_accuracy": 0.
|
| 5 |
-
"eval_ordinal_accuracy": 0.
|
| 6 |
-
"eval_ordinal_mae":
|
| 7 |
-
"eval_runtime":
|
| 8 |
-
"eval_samples_per_second":
|
| 9 |
-
"eval_steps_per_second": 3.
|
| 10 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 1.46,
|
| 3 |
+
"eval_loss": 0.3037484288215637,
|
| 4 |
+
"eval_na_accuracy": 0.7297297120094299,
|
| 5 |
+
"eval_ordinal_accuracy": 0.5285171270370483,
|
| 6 |
+
"eval_ordinal_mae": 0.6723113059997559,
|
| 7 |
+
"eval_runtime": 11.1069,
|
| 8 |
+
"eval_samples_per_second": 27.01,
|
| 9 |
+
"eval_steps_per_second": 3.421
|
| 10 |
}
|
runs/Mar27_22-12-30_ryanserver/events.out.tfevents.1711592832.ryanserver.19364.1
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:98d46e1427ce12b776ce603a453568088afd1258d1e6b7d00e4608b6c15502c1
|
| 3 |
+
size 529
|
train_results.json
CHANGED
|
@@ -1,7 +1,7 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"train_loss": 0.
|
| 4 |
-
"train_runtime":
|
| 5 |
-
"train_samples_per_second":
|
| 6 |
-
"train_steps_per_second": 0.
|
| 7 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 1.46,
|
| 3 |
+
"train_loss": 0.32911812565543436,
|
| 4 |
+
"train_runtime": 761.2747,
|
| 5 |
+
"train_samples_per_second": 15.763,
|
| 6 |
+
"train_steps_per_second": 0.988
|
| 7 |
}
|
trainer_state.json
CHANGED
|
@@ -1,1001 +1,535 @@
|
|
| 1 |
{
|
| 2 |
-
"best_metric": 0.
|
| 3 |
-
"best_model_checkpoint": "./ryan_model3272024/checkpoint-
|
| 4 |
-
"epoch":
|
| 5 |
-
"eval_steps":
|
| 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": 0.03,
|
| 13 |
-
"grad_norm":
|
| 14 |
-
"learning_rate": 0.
|
| 15 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 16 |
"step": 10
|
| 17 |
},
|
| 18 |
{
|
| 19 |
-
"epoch": 0.
|
| 20 |
-
"grad_norm":
|
| 21 |
-
"learning_rate": 0.
|
| 22 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 23 |
"step": 20
|
| 24 |
},
|
| 25 |
{
|
| 26 |
-
"epoch": 0.
|
| 27 |
-
"grad_norm": 0.
|
| 28 |
-
"learning_rate": 0.
|
| 29 |
-
"loss": 0.
|
| 30 |
-
"step":
|
| 31 |
},
|
| 32 |
{
|
| 33 |
"epoch": 0.13,
|
| 34 |
-
"
|
| 35 |
-
"
|
| 36 |
-
"
|
| 37 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
| 38 |
},
|
| 39 |
{
|
| 40 |
"epoch": 0.16,
|
| 41 |
-
"grad_norm": 0.
|
| 42 |
-
"learning_rate": 0.
|
| 43 |
-
"loss": 0.
|
| 44 |
-
"step":
|
| 45 |
},
|
| 46 |
{
|
| 47 |
"epoch": 0.19,
|
| 48 |
-
"grad_norm":
|
| 49 |
-
"learning_rate": 0.
|
| 50 |
-
"loss": 0.
|
| 51 |
-
"step":
|
| 52 |
-
},
|
| 53 |
-
{
|
| 54 |
-
"epoch": 0.22,
|
| 55 |
-
"grad_norm": 0.8919397592544556,
|
| 56 |
-
"learning_rate": 0.00018881789137380192,
|
| 57 |
-
"loss": 0.33,
|
| 58 |
-
"step": 70
|
| 59 |
-
},
|
| 60 |
-
{
|
| 61 |
-
"epoch": 0.26,
|
| 62 |
-
"grad_norm": 1.7052876949310303,
|
| 63 |
-
"learning_rate": 0.00018722044728434505,
|
| 64 |
-
"loss": 0.3337,
|
| 65 |
-
"step": 80
|
| 66 |
},
|
| 67 |
{
|
| 68 |
-
"epoch": 0.
|
| 69 |
-
"grad_norm": 0.
|
| 70 |
-
"learning_rate": 0.
|
| 71 |
-
"loss": 0.
|
| 72 |
-
"step":
|
| 73 |
-
},
|
| 74 |
-
{
|
| 75 |
-
"epoch": 0.32,
|
| 76 |
-
"grad_norm": 1.1663854122161865,
|
| 77 |
-
"learning_rate": 0.00018402555910543132,
|
| 78 |
-
"loss": 0.3853,
|
| 79 |
-
"step": 100
|
| 80 |
-
},
|
| 81 |
-
{
|
| 82 |
-
"epoch": 0.32,
|
| 83 |
-
"eval_loss": 0.3272034823894501,
|
| 84 |
-
"eval_na_accuracy": 0.924,
|
| 85 |
-
"eval_ordinal_accuracy": 0.52,
|
| 86 |
-
"eval_ordinal_mae": 1.210578082634343,
|
| 87 |
-
"eval_runtime": 52.9914,
|
| 88 |
-
"eval_samples_per_second": 9.435,
|
| 89 |
-
"eval_steps_per_second": 1.189,
|
| 90 |
-
"step": 100
|
| 91 |
-
},
|
| 92 |
-
{
|
| 93 |
-
"epoch": 0.35,
|
| 94 |
-
"grad_norm": 0.8579528331756592,
|
| 95 |
-
"learning_rate": 0.00018242811501597444,
|
| 96 |
-
"loss": 0.3585,
|
| 97 |
-
"step": 110
|
| 98 |
-
},
|
| 99 |
-
{
|
| 100 |
-
"epoch": 0.38,
|
| 101 |
-
"grad_norm": 1.02351975440979,
|
| 102 |
-
"learning_rate": 0.00018083067092651756,
|
| 103 |
-
"loss": 0.3621,
|
| 104 |
-
"step": 120
|
| 105 |
-
},
|
| 106 |
-
{
|
| 107 |
-
"epoch": 0.42,
|
| 108 |
-
"grad_norm": 1.3286011219024658,
|
| 109 |
-
"learning_rate": 0.00017923322683706071,
|
| 110 |
-
"loss": 0.3714,
|
| 111 |
-
"step": 130
|
| 112 |
-
},
|
| 113 |
-
{
|
| 114 |
-
"epoch": 0.45,
|
| 115 |
-
"grad_norm": 0.6290095448493958,
|
| 116 |
-
"learning_rate": 0.00017763578274760384,
|
| 117 |
-
"loss": 0.3275,
|
| 118 |
-
"step": 140
|
| 119 |
-
},
|
| 120 |
-
{
|
| 121 |
-
"epoch": 0.48,
|
| 122 |
-
"grad_norm": 1.269338846206665,
|
| 123 |
-
"learning_rate": 0.000176038338658147,
|
| 124 |
-
"loss": 0.4287,
|
| 125 |
-
"step": 150
|
| 126 |
-
},
|
| 127 |
-
{
|
| 128 |
-
"epoch": 0.51,
|
| 129 |
-
"grad_norm": 0.6244733333587646,
|
| 130 |
-
"learning_rate": 0.0001744408945686901,
|
| 131 |
-
"loss": 0.3067,
|
| 132 |
-
"step": 160
|
| 133 |
-
},
|
| 134 |
-
{
|
| 135 |
-
"epoch": 0.54,
|
| 136 |
-
"grad_norm": 1.1287596225738525,
|
| 137 |
-
"learning_rate": 0.00017284345047923323,
|
| 138 |
-
"loss": 0.2982,
|
| 139 |
-
"step": 170
|
| 140 |
-
},
|
| 141 |
-
{
|
| 142 |
-
"epoch": 0.58,
|
| 143 |
-
"grad_norm": 1.436303734779358,
|
| 144 |
-
"learning_rate": 0.00017124600638977638,
|
| 145 |
-
"loss": 0.2946,
|
| 146 |
-
"step": 180
|
| 147 |
-
},
|
| 148 |
-
{
|
| 149 |
-
"epoch": 0.61,
|
| 150 |
-
"grad_norm": 0.8159350752830505,
|
| 151 |
-
"learning_rate": 0.00016964856230031948,
|
| 152 |
-
"loss": 0.3514,
|
| 153 |
-
"step": 190
|
| 154 |
-
},
|
| 155 |
-
{
|
| 156 |
-
"epoch": 0.64,
|
| 157 |
-
"grad_norm": 0.7363901138305664,
|
| 158 |
-
"learning_rate": 0.00016805111821086263,
|
| 159 |
-
"loss": 0.3396,
|
| 160 |
-
"step": 200
|
| 161 |
-
},
|
| 162 |
-
{
|
| 163 |
-
"epoch": 0.64,
|
| 164 |
-
"eval_loss": 0.27412503957748413,
|
| 165 |
-
"eval_na_accuracy": 0.94,
|
| 166 |
-
"eval_ordinal_accuracy": 0.5644444444444444,
|
| 167 |
-
"eval_ordinal_mae": 1.1640199238227473,
|
| 168 |
-
"eval_runtime": 21.3186,
|
| 169 |
-
"eval_samples_per_second": 23.454,
|
| 170 |
-
"eval_steps_per_second": 2.955,
|
| 171 |
-
"step": 200
|
| 172 |
-
},
|
| 173 |
-
{
|
| 174 |
-
"epoch": 0.67,
|
| 175 |
-
"grad_norm": 0.6321592330932617,
|
| 176 |
-
"learning_rate": 0.00016645367412140575,
|
| 177 |
-
"loss": 0.3952,
|
| 178 |
-
"step": 210
|
| 179 |
-
},
|
| 180 |
-
{
|
| 181 |
-
"epoch": 0.7,
|
| 182 |
-
"grad_norm": 0.6153714656829834,
|
| 183 |
-
"learning_rate": 0.0001648562300319489,
|
| 184 |
-
"loss": 0.2947,
|
| 185 |
-
"step": 220
|
| 186 |
-
},
|
| 187 |
-
{
|
| 188 |
-
"epoch": 0.73,
|
| 189 |
-
"grad_norm": 1.3031296730041504,
|
| 190 |
-
"learning_rate": 0.00016325878594249202,
|
| 191 |
-
"loss": 0.3556,
|
| 192 |
-
"step": 230
|
| 193 |
-
},
|
| 194 |
-
{
|
| 195 |
-
"epoch": 0.77,
|
| 196 |
-
"grad_norm": 1.058060646057129,
|
| 197 |
-
"learning_rate": 0.00016166134185303515,
|
| 198 |
-
"loss": 0.3432,
|
| 199 |
-
"step": 240
|
| 200 |
-
},
|
| 201 |
-
{
|
| 202 |
-
"epoch": 0.8,
|
| 203 |
-
"grad_norm": 0.957135796546936,
|
| 204 |
-
"learning_rate": 0.0001600638977635783,
|
| 205 |
-
"loss": 0.3675,
|
| 206 |
-
"step": 250
|
| 207 |
-
},
|
| 208 |
-
{
|
| 209 |
-
"epoch": 0.83,
|
| 210 |
-
"grad_norm": 1.6347941160202026,
|
| 211 |
-
"learning_rate": 0.00015846645367412142,
|
| 212 |
-
"loss": 0.3008,
|
| 213 |
-
"step": 260
|
| 214 |
},
|
| 215 |
{
|
| 216 |
-
"epoch": 0.
|
| 217 |
-
"grad_norm":
|
| 218 |
-
"learning_rate": 0.
|
| 219 |
-
"loss": 0.
|
| 220 |
-
"step":
|
| 221 |
},
|
| 222 |
{
|
| 223 |
-
"epoch": 0.
|
| 224 |
-
"grad_norm": 0.
|
| 225 |
-
"learning_rate": 0.
|
| 226 |
-
"loss": 0.
|
| 227 |
-
"step":
|
| 228 |
},
|
| 229 |
{
|
| 230 |
-
"epoch": 0.
|
| 231 |
-
"
|
| 232 |
-
"
|
| 233 |
-
"
|
| 234 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
| 235 |
},
|
| 236 |
{
|
| 237 |
-
"epoch": 0.
|
| 238 |
-
"grad_norm": 0.
|
| 239 |
-
"learning_rate": 0.
|
| 240 |
-
"loss": 0.
|
| 241 |
-
"step":
|
| 242 |
},
|
| 243 |
{
|
| 244 |
-
"epoch": 0.
|
| 245 |
-
"
|
| 246 |
-
"
|
| 247 |
-
"
|
| 248 |
-
"
|
| 249 |
-
"eval_runtime": 20.7347,
|
| 250 |
-
"eval_samples_per_second": 24.114,
|
| 251 |
-
"eval_steps_per_second": 3.038,
|
| 252 |
-
"step": 300
|
| 253 |
},
|
| 254 |
{
|
| 255 |
-
"epoch": 0.
|
| 256 |
-
"grad_norm": 1.
|
| 257 |
-
"learning_rate": 0.
|
| 258 |
-
"loss": 0.
|
| 259 |
-
"step":
|
| 260 |
},
|
| 261 |
{
|
| 262 |
-
"epoch":
|
| 263 |
-
"grad_norm": 0.
|
| 264 |
-
"learning_rate": 0.
|
| 265 |
"loss": 0.3132,
|
| 266 |
-
"step":
|
| 267 |
-
},
|
| 268 |
-
{
|
| 269 |
-
"epoch": 1.05,
|
| 270 |
-
"grad_norm": 1.1288195848464966,
|
| 271 |
-
"learning_rate": 0.00014728434504792333,
|
| 272 |
-
"loss": 0.255,
|
| 273 |
-
"step": 330
|
| 274 |
-
},
|
| 275 |
-
{
|
| 276 |
-
"epoch": 1.09,
|
| 277 |
-
"grad_norm": 0.7704196572303772,
|
| 278 |
-
"learning_rate": 0.00014568690095846646,
|
| 279 |
-
"loss": 0.2415,
|
| 280 |
-
"step": 340
|
| 281 |
-
},
|
| 282 |
-
{
|
| 283 |
-
"epoch": 1.12,
|
| 284 |
-
"grad_norm": 1.9226877689361572,
|
| 285 |
-
"learning_rate": 0.00014408945686900958,
|
| 286 |
-
"loss": 0.1975,
|
| 287 |
-
"step": 350
|
| 288 |
-
},
|
| 289 |
-
{
|
| 290 |
-
"epoch": 1.15,
|
| 291 |
-
"grad_norm": 0.5694310069084167,
|
| 292 |
-
"learning_rate": 0.00014249201277955273,
|
| 293 |
-
"loss": 0.1722,
|
| 294 |
-
"step": 360
|
| 295 |
-
},
|
| 296 |
-
{
|
| 297 |
-
"epoch": 1.18,
|
| 298 |
-
"grad_norm": 1.719147801399231,
|
| 299 |
-
"learning_rate": 0.00014089456869009585,
|
| 300 |
-
"loss": 0.2175,
|
| 301 |
-
"step": 370
|
| 302 |
-
},
|
| 303 |
-
{
|
| 304 |
-
"epoch": 1.21,
|
| 305 |
-
"grad_norm": 0.9247463941574097,
|
| 306 |
-
"learning_rate": 0.000139297124600639,
|
| 307 |
-
"loss": 0.2088,
|
| 308 |
-
"step": 380
|
| 309 |
-
},
|
| 310 |
-
{
|
| 311 |
-
"epoch": 1.25,
|
| 312 |
-
"grad_norm": 1.0941154956817627,
|
| 313 |
-
"learning_rate": 0.00013769968051118212,
|
| 314 |
-
"loss": 0.2854,
|
| 315 |
-
"step": 390
|
| 316 |
-
},
|
| 317 |
-
{
|
| 318 |
-
"epoch": 1.28,
|
| 319 |
-
"grad_norm": 1.0274015665054321,
|
| 320 |
-
"learning_rate": 0.00013610223642172525,
|
| 321 |
-
"loss": 0.196,
|
| 322 |
-
"step": 400
|
| 323 |
-
},
|
| 324 |
-
{
|
| 325 |
-
"epoch": 1.28,
|
| 326 |
-
"eval_loss": 0.273777574300766,
|
| 327 |
-
"eval_na_accuracy": 0.95,
|
| 328 |
-
"eval_ordinal_accuracy": 0.6133333333333333,
|
| 329 |
-
"eval_ordinal_mae": 1.198390154937903,
|
| 330 |
-
"eval_runtime": 20.9145,
|
| 331 |
-
"eval_samples_per_second": 23.907,
|
| 332 |
-
"eval_steps_per_second": 3.012,
|
| 333 |
-
"step": 400
|
| 334 |
-
},
|
| 335 |
-
{
|
| 336 |
-
"epoch": 1.31,
|
| 337 |
-
"grad_norm": 2.912687063217163,
|
| 338 |
-
"learning_rate": 0.00013450479233226837,
|
| 339 |
-
"loss": 0.2156,
|
| 340 |
-
"step": 410
|
| 341 |
-
},
|
| 342 |
-
{
|
| 343 |
-
"epoch": 1.34,
|
| 344 |
-
"grad_norm": 0.6906268000602722,
|
| 345 |
-
"learning_rate": 0.0001329073482428115,
|
| 346 |
-
"loss": 0.1366,
|
| 347 |
-
"step": 420
|
| 348 |
-
},
|
| 349 |
-
{
|
| 350 |
-
"epoch": 1.37,
|
| 351 |
-
"grad_norm": 0.43070048093795776,
|
| 352 |
-
"learning_rate": 0.00013130990415335464,
|
| 353 |
-
"loss": 0.2174,
|
| 354 |
-
"step": 430
|
| 355 |
-
},
|
| 356 |
-
{
|
| 357 |
-
"epoch": 1.41,
|
| 358 |
-
"grad_norm": 0.5173763632774353,
|
| 359 |
-
"learning_rate": 0.00012971246006389777,
|
| 360 |
-
"loss": 0.2016,
|
| 361 |
-
"step": 440
|
| 362 |
-
},
|
| 363 |
-
{
|
| 364 |
-
"epoch": 1.44,
|
| 365 |
-
"grad_norm": 1.04314386844635,
|
| 366 |
-
"learning_rate": 0.00012811501597444092,
|
| 367 |
-
"loss": 0.2233,
|
| 368 |
-
"step": 450
|
| 369 |
-
},
|
| 370 |
-
{
|
| 371 |
-
"epoch": 1.47,
|
| 372 |
-
"grad_norm": 0.523073673248291,
|
| 373 |
-
"learning_rate": 0.00012651757188498404,
|
| 374 |
-
"loss": 0.2231,
|
| 375 |
-
"step": 460
|
| 376 |
-
},
|
| 377 |
-
{
|
| 378 |
-
"epoch": 1.5,
|
| 379 |
-
"grad_norm": 3.259795904159546,
|
| 380 |
-
"learning_rate": 0.00012492012779552716,
|
| 381 |
-
"loss": 0.2366,
|
| 382 |
-
"step": 470
|
| 383 |
-
},
|
| 384 |
-
{
|
| 385 |
-
"epoch": 1.53,
|
| 386 |
-
"grad_norm": 0.6846562027931213,
|
| 387 |
-
"learning_rate": 0.00012332268370607028,
|
| 388 |
-
"loss": 0.2144,
|
| 389 |
-
"step": 480
|
| 390 |
-
},
|
| 391 |
-
{
|
| 392 |
-
"epoch": 1.57,
|
| 393 |
-
"grad_norm": 1.2122007608413696,
|
| 394 |
-
"learning_rate": 0.00012172523961661342,
|
| 395 |
-
"loss": 0.2938,
|
| 396 |
-
"step": 490
|
| 397 |
-
},
|
| 398 |
-
{
|
| 399 |
-
"epoch": 1.6,
|
| 400 |
-
"grad_norm": 1.3790067434310913,
|
| 401 |
-
"learning_rate": 0.00012012779552715656,
|
| 402 |
-
"loss": 0.2228,
|
| 403 |
-
"step": 500
|
| 404 |
-
},
|
| 405 |
-
{
|
| 406 |
-
"epoch": 1.6,
|
| 407 |
-
"eval_loss": 0.26852139830589294,
|
| 408 |
-
"eval_na_accuracy": 0.956,
|
| 409 |
-
"eval_ordinal_accuracy": 0.62,
|
| 410 |
-
"eval_ordinal_mae": 1.1989026491012837,
|
| 411 |
-
"eval_runtime": 20.0158,
|
| 412 |
-
"eval_samples_per_second": 24.98,
|
| 413 |
-
"eval_steps_per_second": 3.148,
|
| 414 |
-
"step": 500
|
| 415 |
-
},
|
| 416 |
-
{
|
| 417 |
-
"epoch": 1.63,
|
| 418 |
-
"grad_norm": 0.7108421921730042,
|
| 419 |
-
"learning_rate": 0.00011853035143769968,
|
| 420 |
-
"loss": 0.1916,
|
| 421 |
-
"step": 510
|
| 422 |
-
},
|
| 423 |
-
{
|
| 424 |
-
"epoch": 1.66,
|
| 425 |
-
"grad_norm": 0.42910462617874146,
|
| 426 |
-
"learning_rate": 0.00011693290734824283,
|
| 427 |
-
"loss": 0.2478,
|
| 428 |
-
"step": 520
|
| 429 |
-
},
|
| 430 |
-
{
|
| 431 |
-
"epoch": 1.69,
|
| 432 |
-
"grad_norm": 0.9730465412139893,
|
| 433 |
-
"learning_rate": 0.00011533546325878595,
|
| 434 |
-
"loss": 0.189,
|
| 435 |
-
"step": 530
|
| 436 |
-
},
|
| 437 |
-
{
|
| 438 |
-
"epoch": 1.73,
|
| 439 |
-
"grad_norm": 0.9566612243652344,
|
| 440 |
-
"learning_rate": 0.00011373801916932908,
|
| 441 |
-
"loss": 0.1768,
|
| 442 |
-
"step": 540
|
| 443 |
-
},
|
| 444 |
-
{
|
| 445 |
-
"epoch": 1.76,
|
| 446 |
-
"grad_norm": 0.5167070627212524,
|
| 447 |
-
"learning_rate": 0.00011214057507987221,
|
| 448 |
-
"loss": 0.1385,
|
| 449 |
-
"step": 550
|
| 450 |
-
},
|
| 451 |
-
{
|
| 452 |
-
"epoch": 1.79,
|
| 453 |
-
"grad_norm": 0.5880122780799866,
|
| 454 |
-
"learning_rate": 0.00011054313099041533,
|
| 455 |
-
"loss": 0.1262,
|
| 456 |
-
"step": 560
|
| 457 |
-
},
|
| 458 |
-
{
|
| 459 |
-
"epoch": 1.82,
|
| 460 |
-
"grad_norm": 1.202286720275879,
|
| 461 |
-
"learning_rate": 0.00010894568690095847,
|
| 462 |
-
"loss": 0.1721,
|
| 463 |
-
"step": 570
|
| 464 |
-
},
|
| 465 |
-
{
|
| 466 |
-
"epoch": 1.85,
|
| 467 |
-
"grad_norm": 2.6997601985931396,
|
| 468 |
-
"learning_rate": 0.0001073482428115016,
|
| 469 |
-
"loss": 0.2128,
|
| 470 |
-
"step": 580
|
| 471 |
-
},
|
| 472 |
-
{
|
| 473 |
-
"epoch": 1.88,
|
| 474 |
-
"grad_norm": 1.1591830253601074,
|
| 475 |
-
"learning_rate": 0.00010575079872204474,
|
| 476 |
-
"loss": 0.2402,
|
| 477 |
-
"step": 590
|
| 478 |
-
},
|
| 479 |
-
{
|
| 480 |
-
"epoch": 1.92,
|
| 481 |
-
"grad_norm": 0.5840221643447876,
|
| 482 |
-
"learning_rate": 0.00010415335463258787,
|
| 483 |
-
"loss": 0.1816,
|
| 484 |
-
"step": 600
|
| 485 |
-
},
|
| 486 |
-
{
|
| 487 |
-
"epoch": 1.92,
|
| 488 |
-
"eval_loss": 0.2552177309989929,
|
| 489 |
-
"eval_na_accuracy": 0.95,
|
| 490 |
-
"eval_ordinal_accuracy": 0.6266666666666667,
|
| 491 |
-
"eval_ordinal_mae": 1.158560517811113,
|
| 492 |
-
"eval_runtime": 19.5011,
|
| 493 |
-
"eval_samples_per_second": 25.64,
|
| 494 |
-
"eval_steps_per_second": 3.231,
|
| 495 |
-
"step": 600
|
| 496 |
-
},
|
| 497 |
-
{
|
| 498 |
-
"epoch": 1.95,
|
| 499 |
-
"grad_norm": 0.7560299634933472,
|
| 500 |
-
"learning_rate": 0.000102555910543131,
|
| 501 |
-
"loss": 0.2021,
|
| 502 |
-
"step": 610
|
| 503 |
-
},
|
| 504 |
-
{
|
| 505 |
-
"epoch": 1.98,
|
| 506 |
-
"grad_norm": 1.8860361576080322,
|
| 507 |
-
"learning_rate": 0.00010095846645367413,
|
| 508 |
-
"loss": 0.2092,
|
| 509 |
-
"step": 620
|
| 510 |
-
},
|
| 511 |
-
{
|
| 512 |
-
"epoch": 2.01,
|
| 513 |
-
"grad_norm": 0.7235255837440491,
|
| 514 |
-
"learning_rate": 9.936102236421726e-05,
|
| 515 |
-
"loss": 0.1131,
|
| 516 |
-
"step": 630
|
| 517 |
-
},
|
| 518 |
-
{
|
| 519 |
-
"epoch": 2.04,
|
| 520 |
-
"grad_norm": 0.3656529486179352,
|
| 521 |
-
"learning_rate": 9.77635782747604e-05,
|
| 522 |
-
"loss": 0.0867,
|
| 523 |
-
"step": 640
|
| 524 |
-
},
|
| 525 |
-
{
|
| 526 |
-
"epoch": 2.08,
|
| 527 |
-
"grad_norm": 0.3450271785259247,
|
| 528 |
-
"learning_rate": 9.616613418530351e-05,
|
| 529 |
-
"loss": 0.0903,
|
| 530 |
-
"step": 650
|
| 531 |
-
},
|
| 532 |
-
{
|
| 533 |
-
"epoch": 2.11,
|
| 534 |
-
"grad_norm": 1.0603750944137573,
|
| 535 |
-
"learning_rate": 9.456869009584664e-05,
|
| 536 |
-
"loss": 0.1234,
|
| 537 |
-
"step": 660
|
| 538 |
-
},
|
| 539 |
-
{
|
| 540 |
-
"epoch": 2.14,
|
| 541 |
-
"grad_norm": 0.6790297031402588,
|
| 542 |
-
"learning_rate": 9.297124600638978e-05,
|
| 543 |
-
"loss": 0.0936,
|
| 544 |
-
"step": 670
|
| 545 |
-
},
|
| 546 |
-
{
|
| 547 |
-
"epoch": 2.17,
|
| 548 |
-
"grad_norm": 0.5596363544464111,
|
| 549 |
-
"learning_rate": 9.137380191693292e-05,
|
| 550 |
-
"loss": 0.0651,
|
| 551 |
-
"step": 680
|
| 552 |
-
},
|
| 553 |
-
{
|
| 554 |
-
"epoch": 2.2,
|
| 555 |
-
"grad_norm": 0.5989049673080444,
|
| 556 |
-
"learning_rate": 8.977635782747604e-05,
|
| 557 |
-
"loss": 0.1218,
|
| 558 |
-
"step": 690
|
| 559 |
-
},
|
| 560 |
-
{
|
| 561 |
-
"epoch": 2.24,
|
| 562 |
-
"grad_norm": 0.9003208875656128,
|
| 563 |
-
"learning_rate": 8.817891373801918e-05,
|
| 564 |
-
"loss": 0.0682,
|
| 565 |
-
"step": 700
|
| 566 |
-
},
|
| 567 |
-
{
|
| 568 |
-
"epoch": 2.24,
|
| 569 |
-
"eval_loss": 0.27212005853652954,
|
| 570 |
-
"eval_na_accuracy": 0.952,
|
| 571 |
-
"eval_ordinal_accuracy": 0.6577777777777778,
|
| 572 |
-
"eval_ordinal_mae": 1.1557789803379113,
|
| 573 |
-
"eval_runtime": 19.5966,
|
| 574 |
-
"eval_samples_per_second": 25.515,
|
| 575 |
-
"eval_steps_per_second": 3.215,
|
| 576 |
-
"step": 700
|
| 577 |
-
},
|
| 578 |
-
{
|
| 579 |
-
"epoch": 2.27,
|
| 580 |
-
"grad_norm": 0.6663013100624084,
|
| 581 |
-
"learning_rate": 8.658146964856231e-05,
|
| 582 |
-
"loss": 0.0714,
|
| 583 |
-
"step": 710
|
| 584 |
-
},
|
| 585 |
-
{
|
| 586 |
-
"epoch": 2.3,
|
| 587 |
-
"grad_norm": 1.0458776950836182,
|
| 588 |
-
"learning_rate": 8.498402555910544e-05,
|
| 589 |
-
"loss": 0.102,
|
| 590 |
-
"step": 720
|
| 591 |
-
},
|
| 592 |
-
{
|
| 593 |
-
"epoch": 2.33,
|
| 594 |
-
"grad_norm": 0.9246501922607422,
|
| 595 |
-
"learning_rate": 8.338658146964856e-05,
|
| 596 |
-
"loss": 0.1623,
|
| 597 |
-
"step": 730
|
| 598 |
-
},
|
| 599 |
-
{
|
| 600 |
-
"epoch": 2.36,
|
| 601 |
-
"grad_norm": 1.0837684869766235,
|
| 602 |
-
"learning_rate": 8.17891373801917e-05,
|
| 603 |
-
"loss": 0.0934,
|
| 604 |
-
"step": 740
|
| 605 |
-
},
|
| 606 |
-
{
|
| 607 |
-
"epoch": 2.4,
|
| 608 |
-
"grad_norm": 0.564241349697113,
|
| 609 |
-
"learning_rate": 8.019169329073483e-05,
|
| 610 |
-
"loss": 0.0853,
|
| 611 |
-
"step": 750
|
| 612 |
},
|
| 613 |
{
|
| 614 |
-
"epoch":
|
| 615 |
-
"grad_norm":
|
| 616 |
-
"learning_rate":
|
| 617 |
-
"loss": 0.
|
| 618 |
-
"step":
|
| 619 |
},
|
| 620 |
{
|
| 621 |
-
"epoch":
|
| 622 |
-
"
|
| 623 |
-
"
|
| 624 |
-
"
|
| 625 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
| 626 |
},
|
| 627 |
{
|
| 628 |
-
"epoch":
|
| 629 |
-
"grad_norm":
|
| 630 |
-
"learning_rate":
|
| 631 |
-
"loss": 0.
|
| 632 |
-
"step":
|
| 633 |
},
|
| 634 |
{
|
| 635 |
-
"epoch":
|
| 636 |
-
"grad_norm":
|
| 637 |
-
"learning_rate":
|
| 638 |
-
"loss": 0.
|
| 639 |
-
"step":
|
| 640 |
},
|
| 641 |
{
|
| 642 |
-
"epoch":
|
| 643 |
-
"grad_norm": 0.
|
| 644 |
-
"learning_rate":
|
| 645 |
-
"loss": 0.
|
| 646 |
-
"step":
|
| 647 |
},
|
| 648 |
{
|
| 649 |
-
"epoch":
|
| 650 |
-
"
|
| 651 |
-
"
|
| 652 |
-
"
|
| 653 |
-
"
|
| 654 |
-
"eval_runtime": 20.0506,
|
| 655 |
-
"eval_samples_per_second": 24.937,
|
| 656 |
-
"eval_steps_per_second": 3.142,
|
| 657 |
-
"step": 800
|
| 658 |
},
|
| 659 |
{
|
| 660 |
-
"epoch":
|
| 661 |
-
"grad_norm":
|
| 662 |
-
"learning_rate":
|
| 663 |
-
"loss": 0.
|
| 664 |
-
"step":
|
| 665 |
},
|
| 666 |
{
|
| 667 |
-
"epoch":
|
| 668 |
-
"
|
| 669 |
-
"
|
| 670 |
-
"
|
| 671 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
| 672 |
},
|
| 673 |
{
|
| 674 |
-
"epoch":
|
| 675 |
-
"grad_norm":
|
| 676 |
-
"learning_rate":
|
| 677 |
-
"loss": 0.
|
| 678 |
-
"step":
|
| 679 |
},
|
| 680 |
{
|
| 681 |
-
"epoch":
|
| 682 |
-
"grad_norm": 0.
|
| 683 |
-
"learning_rate":
|
| 684 |
-
"loss": 0.
|
| 685 |
-
"step":
|
| 686 |
},
|
| 687 |
{
|
| 688 |
-
"epoch":
|
| 689 |
-
"grad_norm": 0.
|
| 690 |
-
"learning_rate":
|
| 691 |
-
"loss": 0.
|
| 692 |
-
"step":
|
| 693 |
},
|
| 694 |
{
|
| 695 |
-
"epoch":
|
| 696 |
-
"grad_norm":
|
| 697 |
-
"learning_rate":
|
| 698 |
-
"loss": 0.
|
| 699 |
-
"step":
|
| 700 |
},
|
| 701 |
{
|
| 702 |
-
"epoch":
|
| 703 |
-
"grad_norm": 0.
|
| 704 |
-
"learning_rate":
|
| 705 |
-
"loss": 0.
|
| 706 |
-
"step":
|
| 707 |
},
|
| 708 |
{
|
| 709 |
-
"epoch":
|
| 710 |
-
"
|
| 711 |
-
"
|
| 712 |
-
"
|
| 713 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
| 714 |
},
|
| 715 |
{
|
| 716 |
-
"epoch":
|
| 717 |
-
"grad_norm": 0.
|
| 718 |
-
"learning_rate":
|
| 719 |
-
"loss": 0.
|
| 720 |
-
"step":
|
| 721 |
},
|
| 722 |
{
|
| 723 |
-
"epoch":
|
| 724 |
-
"grad_norm":
|
| 725 |
-
"learning_rate":
|
| 726 |
-
"loss": 0.
|
| 727 |
-
"step":
|
| 728 |
},
|
| 729 |
{
|
| 730 |
-
"epoch":
|
| 731 |
-
"
|
| 732 |
-
"
|
| 733 |
-
"
|
| 734 |
-
"
|
| 735 |
-
"eval_runtime": 20.0091,
|
| 736 |
-
"eval_samples_per_second": 24.989,
|
| 737 |
-
"eval_steps_per_second": 3.149,
|
| 738 |
-
"step": 900
|
| 739 |
},
|
| 740 |
{
|
| 741 |
-
"epoch":
|
| 742 |
-
"grad_norm":
|
| 743 |
-
"learning_rate":
|
| 744 |
-
"loss": 0.
|
| 745 |
-
"step":
|
| 746 |
},
|
| 747 |
{
|
| 748 |
-
"epoch":
|
| 749 |
-
"grad_norm":
|
| 750 |
-
"learning_rate":
|
| 751 |
-
"loss": 0.
|
| 752 |
-
"step":
|
| 753 |
},
|
| 754 |
{
|
| 755 |
-
"epoch":
|
| 756 |
-
"
|
| 757 |
-
"
|
| 758 |
-
"
|
| 759 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
| 760 |
},
|
| 761 |
{
|
| 762 |
-
"epoch":
|
| 763 |
-
"grad_norm": 0.
|
| 764 |
-
"learning_rate":
|
| 765 |
-
"loss": 0.
|
| 766 |
-
"step":
|
| 767 |
},
|
| 768 |
{
|
| 769 |
-
"epoch":
|
| 770 |
-
"grad_norm":
|
| 771 |
-
"learning_rate":
|
| 772 |
-
"loss": 0.
|
| 773 |
-
"step":
|
| 774 |
},
|
| 775 |
{
|
| 776 |
-
"epoch":
|
| 777 |
-
"grad_norm": 0.
|
| 778 |
-
"learning_rate":
|
| 779 |
-
"loss": 0.
|
| 780 |
-
"step":
|
| 781 |
},
|
| 782 |
{
|
| 783 |
-
"epoch":
|
| 784 |
-
"grad_norm": 0.
|
| 785 |
-
"learning_rate":
|
| 786 |
-
"loss": 0.
|
| 787 |
-
"step":
|
| 788 |
},
|
| 789 |
{
|
| 790 |
-
"epoch":
|
| 791 |
-
"grad_norm": 0.
|
| 792 |
-
"learning_rate":
|
| 793 |
-
"loss": 0.
|
| 794 |
-
"step":
|
| 795 |
},
|
| 796 |
{
|
| 797 |
-
"epoch":
|
| 798 |
-
"
|
| 799 |
-
"
|
| 800 |
-
"
|
| 801 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
| 802 |
},
|
| 803 |
{
|
| 804 |
-
"epoch":
|
| 805 |
-
"grad_norm": 0.
|
| 806 |
-
"learning_rate":
|
| 807 |
-
"loss": 0.
|
| 808 |
-
"step":
|
| 809 |
},
|
| 810 |
{
|
| 811 |
-
"epoch":
|
| 812 |
-
"
|
| 813 |
-
"
|
| 814 |
-
"
|
| 815 |
-
"
|
| 816 |
-
"eval_runtime": 19.3226,
|
| 817 |
-
"eval_samples_per_second": 25.876,
|
| 818 |
-
"eval_steps_per_second": 3.26,
|
| 819 |
-
"step": 1000
|
| 820 |
},
|
| 821 |
{
|
| 822 |
-
"epoch":
|
| 823 |
-
"grad_norm":
|
| 824 |
-
"learning_rate":
|
| 825 |
-
"loss": 0.
|
| 826 |
-
"step":
|
| 827 |
},
|
| 828 |
{
|
| 829 |
-
"epoch":
|
| 830 |
-
"grad_norm": 0.
|
| 831 |
-
"learning_rate":
|
| 832 |
-
"loss": 0.
|
| 833 |
-
"step":
|
| 834 |
},
|
| 835 |
{
|
| 836 |
-
"epoch":
|
| 837 |
-
"grad_norm":
|
| 838 |
-
"learning_rate":
|
| 839 |
-
"loss": 0.
|
| 840 |
-
"step":
|
| 841 |
},
|
| 842 |
{
|
| 843 |
-
"epoch":
|
| 844 |
-
"
|
| 845 |
-
"
|
| 846 |
-
"
|
| 847 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
| 848 |
},
|
| 849 |
{
|
| 850 |
-
"epoch":
|
| 851 |
-
"grad_norm": 0.
|
| 852 |
-
"learning_rate":
|
| 853 |
-
"loss": 0.
|
| 854 |
-
"step":
|
| 855 |
},
|
| 856 |
{
|
| 857 |
-
"epoch":
|
| 858 |
-
"grad_norm": 0.
|
| 859 |
-
"learning_rate":
|
| 860 |
-
"loss": 0.
|
| 861 |
-
"step":
|
| 862 |
},
|
| 863 |
{
|
| 864 |
-
"epoch":
|
| 865 |
-
"grad_norm": 0.
|
| 866 |
-
"learning_rate":
|
| 867 |
-
"loss": 0.
|
| 868 |
-
"step":
|
| 869 |
},
|
| 870 |
{
|
| 871 |
-
"epoch":
|
| 872 |
-
"grad_norm":
|
| 873 |
-
"learning_rate":
|
| 874 |
-
"loss": 0.
|
| 875 |
-
"step":
|
| 876 |
},
|
| 877 |
{
|
| 878 |
-
"epoch":
|
| 879 |
-
"grad_norm":
|
| 880 |
-
"learning_rate":
|
| 881 |
-
"loss": 0.
|
| 882 |
-
"step":
|
| 883 |
},
|
| 884 |
{
|
| 885 |
-
"epoch":
|
| 886 |
-
"
|
| 887 |
-
"
|
| 888 |
-
"
|
| 889 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
| 890 |
},
|
| 891 |
{
|
| 892 |
-
"epoch":
|
| 893 |
-
"
|
| 894 |
-
"
|
| 895 |
-
"
|
| 896 |
-
"
|
| 897 |
-
"eval_runtime": 19.9365,
|
| 898 |
-
"eval_samples_per_second": 25.08,
|
| 899 |
-
"eval_steps_per_second": 3.16,
|
| 900 |
-
"step": 1100
|
| 901 |
},
|
| 902 |
{
|
| 903 |
-
"epoch":
|
| 904 |
-
"grad_norm": 0.
|
| 905 |
-
"learning_rate":
|
| 906 |
-
"loss": 0.
|
| 907 |
-
"step":
|
| 908 |
},
|
| 909 |
{
|
| 910 |
-
"epoch":
|
| 911 |
-
"grad_norm":
|
| 912 |
-
"learning_rate":
|
| 913 |
-
"loss": 0.
|
| 914 |
-
"step":
|
| 915 |
},
|
| 916 |
{
|
| 917 |
-
"epoch": 3
|
| 918 |
-
"grad_norm": 0.
|
| 919 |
-
"learning_rate":
|
| 920 |
-
"loss": 0.
|
| 921 |
-
"step":
|
| 922 |
},
|
| 923 |
{
|
| 924 |
-
"epoch":
|
| 925 |
-
"grad_norm": 0.
|
| 926 |
-
"learning_rate":
|
| 927 |
-
"loss": 0.
|
| 928 |
-
"step":
|
| 929 |
},
|
| 930 |
{
|
| 931 |
-
"epoch":
|
| 932 |
-
"
|
| 933 |
-
"
|
| 934 |
-
"
|
| 935 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
| 936 |
},
|
| 937 |
{
|
| 938 |
-
"epoch":
|
| 939 |
-
"grad_norm": 0.
|
| 940 |
-
"learning_rate":
|
| 941 |
-
"loss": 0.
|
| 942 |
-
"step":
|
| 943 |
},
|
| 944 |
{
|
| 945 |
-
"epoch":
|
| 946 |
-
"grad_norm": 0.
|
| 947 |
-
"learning_rate":
|
| 948 |
-
"loss": 0.
|
| 949 |
-
"step":
|
| 950 |
},
|
| 951 |
{
|
| 952 |
-
"epoch":
|
| 953 |
-
"grad_norm":
|
| 954 |
-
"learning_rate":
|
| 955 |
-
"loss": 0.
|
| 956 |
-
"step":
|
| 957 |
},
|
| 958 |
{
|
| 959 |
-
"epoch":
|
| 960 |
-
"grad_norm": 0.
|
| 961 |
-
"learning_rate":
|
| 962 |
-
"loss": 0.
|
| 963 |
-
"step":
|
| 964 |
},
|
| 965 |
{
|
| 966 |
-
"epoch":
|
| 967 |
-
"grad_norm":
|
| 968 |
-
"learning_rate":
|
| 969 |
-
"loss": 0.
|
| 970 |
-
"step":
|
| 971 |
},
|
| 972 |
{
|
| 973 |
-
"epoch":
|
| 974 |
-
"eval_loss": 0.
|
| 975 |
-
"eval_na_accuracy": 0.
|
| 976 |
-
"eval_ordinal_accuracy": 0.
|
| 977 |
-
"eval_ordinal_mae":
|
| 978 |
-
"eval_runtime":
|
| 979 |
-
"eval_samples_per_second":
|
| 980 |
-
"eval_steps_per_second": 3.
|
| 981 |
-
"step":
|
| 982 |
},
|
| 983 |
{
|
| 984 |
-
"epoch":
|
| 985 |
-
"step":
|
| 986 |
-
"total_flos":
|
| 987 |
-
"train_loss": 0.
|
| 988 |
-
"train_runtime":
|
| 989 |
-
"train_samples_per_second":
|
| 990 |
-
"train_steps_per_second": 0.
|
| 991 |
}
|
| 992 |
],
|
| 993 |
-
"logging_steps":
|
| 994 |
-
"max_steps":
|
| 995 |
"num_input_tokens_seen": 0,
|
| 996 |
"num_train_epochs": 4,
|
| 997 |
-
"save_steps":
|
| 998 |
-
"total_flos":
|
| 999 |
"train_batch_size": 16,
|
| 1000 |
"trial_name": null,
|
| 1001 |
"trial_params": null
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_metric": 0.3037484288215637,
|
| 3 |
+
"best_model_checkpoint": "./ryan_model3272024/checkpoint-250",
|
| 4 |
+
"epoch": 1.4627659574468086,
|
| 5 |
+
"eval_steps": 25,
|
| 6 |
+
"global_step": 275,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
"epoch": 0.03,
|
| 13 |
+
"grad_norm": 0.5357832312583923,
|
| 14 |
+
"learning_rate": 0.00019867021276595746,
|
| 15 |
+
"loss": 0.4838,
|
| 16 |
+
"step": 5
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.05,
|
| 20 |
+
"grad_norm": 0.3301275372505188,
|
| 21 |
+
"learning_rate": 0.00019734042553191489,
|
| 22 |
+
"loss": 0.3457,
|
| 23 |
"step": 10
|
| 24 |
},
|
| 25 |
{
|
| 26 |
+
"epoch": 0.08,
|
| 27 |
+
"grad_norm": 0.829723060131073,
|
| 28 |
+
"learning_rate": 0.00019601063829787234,
|
| 29 |
+
"loss": 0.4508,
|
| 30 |
+
"step": 15
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.11,
|
| 34 |
+
"grad_norm": 0.43481916189193726,
|
| 35 |
+
"learning_rate": 0.00019468085106382982,
|
| 36 |
+
"loss": 0.3205,
|
| 37 |
"step": 20
|
| 38 |
},
|
| 39 |
{
|
| 40 |
+
"epoch": 0.13,
|
| 41 |
+
"grad_norm": 0.9304245114326477,
|
| 42 |
+
"learning_rate": 0.00019335106382978724,
|
| 43 |
+
"loss": 0.4062,
|
| 44 |
+
"step": 25
|
| 45 |
},
|
| 46 |
{
|
| 47 |
"epoch": 0.13,
|
| 48 |
+
"eval_loss": 0.3798711597919464,
|
| 49 |
+
"eval_na_accuracy": 0.6216216087341309,
|
| 50 |
+
"eval_ordinal_accuracy": 0.23954372107982635,
|
| 51 |
+
"eval_ordinal_mae": 0.9244347810745239,
|
| 52 |
+
"eval_runtime": 31.9197,
|
| 53 |
+
"eval_samples_per_second": 9.399,
|
| 54 |
+
"eval_steps_per_second": 1.19,
|
| 55 |
+
"step": 25
|
| 56 |
},
|
| 57 |
{
|
| 58 |
"epoch": 0.16,
|
| 59 |
+
"grad_norm": 0.8552330136299133,
|
| 60 |
+
"learning_rate": 0.0001920212765957447,
|
| 61 |
+
"loss": 0.3029,
|
| 62 |
+
"step": 30
|
| 63 |
},
|
| 64 |
{
|
| 65 |
"epoch": 0.19,
|
| 66 |
+
"grad_norm": 0.43845850229263306,
|
| 67 |
+
"learning_rate": 0.00019069148936170214,
|
| 68 |
+
"loss": 0.3261,
|
| 69 |
+
"step": 35
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 70 |
},
|
| 71 |
{
|
| 72 |
+
"epoch": 0.21,
|
| 73 |
+
"grad_norm": 0.18216854333877563,
|
| 74 |
+
"learning_rate": 0.00018936170212765957,
|
| 75 |
+
"loss": 0.2895,
|
| 76 |
+
"step": 40
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 77 |
},
|
| 78 |
{
|
| 79 |
+
"epoch": 0.24,
|
| 80 |
+
"grad_norm": 0.7537259459495544,
|
| 81 |
+
"learning_rate": 0.00018829787234042554,
|
| 82 |
+
"loss": 0.4943,
|
| 83 |
+
"step": 45
|
| 84 |
},
|
| 85 |
{
|
| 86 |
+
"epoch": 0.27,
|
| 87 |
+
"grad_norm": 0.4183604419231415,
|
| 88 |
+
"learning_rate": 0.000186968085106383,
|
| 89 |
+
"loss": 0.3536,
|
| 90 |
+
"step": 50
|
| 91 |
},
|
| 92 |
{
|
| 93 |
+
"epoch": 0.27,
|
| 94 |
+
"eval_loss": 0.3699643015861511,
|
| 95 |
+
"eval_na_accuracy": 0.6756756901741028,
|
| 96 |
+
"eval_ordinal_accuracy": 0.38403043150901794,
|
| 97 |
+
"eval_ordinal_mae": 0.9066693782806396,
|
| 98 |
+
"eval_runtime": 12.0769,
|
| 99 |
+
"eval_samples_per_second": 24.841,
|
| 100 |
+
"eval_steps_per_second": 3.147,
|
| 101 |
+
"step": 50
|
| 102 |
},
|
| 103 |
{
|
| 104 |
+
"epoch": 0.29,
|
| 105 |
+
"grad_norm": 0.8079000115394592,
|
| 106 |
+
"learning_rate": 0.00018563829787234044,
|
| 107 |
+
"loss": 0.3601,
|
| 108 |
+
"step": 55
|
| 109 |
},
|
| 110 |
{
|
| 111 |
+
"epoch": 0.32,
|
| 112 |
+
"grad_norm": 0.9893763065338135,
|
| 113 |
+
"learning_rate": 0.0001843085106382979,
|
| 114 |
+
"loss": 0.4152,
|
| 115 |
+
"step": 60
|
|
|
|
|
|
|
|
|
|
|
|
|
| 116 |
},
|
| 117 |
{
|
| 118 |
+
"epoch": 0.35,
|
| 119 |
+
"grad_norm": 1.3993595838546753,
|
| 120 |
+
"learning_rate": 0.00018297872340425532,
|
| 121 |
+
"loss": 0.448,
|
| 122 |
+
"step": 65
|
| 123 |
},
|
| 124 |
{
|
| 125 |
+
"epoch": 0.37,
|
| 126 |
+
"grad_norm": 0.45827603340148926,
|
| 127 |
+
"learning_rate": 0.00018164893617021277,
|
| 128 |
"loss": 0.3132,
|
| 129 |
+
"step": 70
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 130 |
},
|
| 131 |
{
|
| 132 |
+
"epoch": 0.4,
|
| 133 |
+
"grad_norm": 1.4677202701568604,
|
| 134 |
+
"learning_rate": 0.00018031914893617022,
|
| 135 |
+
"loss": 0.4295,
|
| 136 |
+
"step": 75
|
| 137 |
},
|
| 138 |
{
|
| 139 |
+
"epoch": 0.4,
|
| 140 |
+
"eval_loss": 0.34051504731178284,
|
| 141 |
+
"eval_na_accuracy": 0.7837837934494019,
|
| 142 |
+
"eval_ordinal_accuracy": 0.29657796025276184,
|
| 143 |
+
"eval_ordinal_mae": 0.8797782063484192,
|
| 144 |
+
"eval_runtime": 11.5094,
|
| 145 |
+
"eval_samples_per_second": 26.066,
|
| 146 |
+
"eval_steps_per_second": 3.302,
|
| 147 |
+
"step": 75
|
| 148 |
},
|
| 149 |
{
|
| 150 |
+
"epoch": 0.43,
|
| 151 |
+
"grad_norm": 1.1136772632598877,
|
| 152 |
+
"learning_rate": 0.00017898936170212767,
|
| 153 |
+
"loss": 0.2444,
|
| 154 |
+
"step": 80
|
| 155 |
},
|
| 156 |
{
|
| 157 |
+
"epoch": 0.45,
|
| 158 |
+
"grad_norm": 3.5127644538879395,
|
| 159 |
+
"learning_rate": 0.00017765957446808512,
|
| 160 |
+
"loss": 0.4542,
|
| 161 |
+
"step": 85
|
| 162 |
},
|
| 163 |
{
|
| 164 |
+
"epoch": 0.48,
|
| 165 |
+
"grad_norm": 0.6393898725509644,
|
| 166 |
+
"learning_rate": 0.00017632978723404257,
|
| 167 |
+
"loss": 0.3411,
|
| 168 |
+
"step": 90
|
| 169 |
},
|
| 170 |
{
|
| 171 |
+
"epoch": 0.51,
|
| 172 |
+
"grad_norm": 0.8198608160018921,
|
| 173 |
+
"learning_rate": 0.000175,
|
| 174 |
+
"loss": 0.3184,
|
| 175 |
+
"step": 95
|
|
|
|
|
|
|
|
|
|
|
|
|
| 176 |
},
|
| 177 |
{
|
| 178 |
+
"epoch": 0.53,
|
| 179 |
+
"grad_norm": 1.1194807291030884,
|
| 180 |
+
"learning_rate": 0.00017367021276595745,
|
| 181 |
+
"loss": 0.4114,
|
| 182 |
+
"step": 100
|
| 183 |
},
|
| 184 |
{
|
| 185 |
+
"epoch": 0.53,
|
| 186 |
+
"eval_loss": 0.39057785272598267,
|
| 187 |
+
"eval_na_accuracy": 0.7297297120094299,
|
| 188 |
+
"eval_ordinal_accuracy": 0.35361215472221375,
|
| 189 |
+
"eval_ordinal_mae": 0.8806185126304626,
|
| 190 |
+
"eval_runtime": 12.2664,
|
| 191 |
+
"eval_samples_per_second": 24.457,
|
| 192 |
+
"eval_steps_per_second": 3.098,
|
| 193 |
+
"step": 100
|
| 194 |
},
|
| 195 |
{
|
| 196 |
+
"epoch": 0.56,
|
| 197 |
+
"grad_norm": 3.365068197250366,
|
| 198 |
+
"learning_rate": 0.0001723404255319149,
|
| 199 |
+
"loss": 0.469,
|
| 200 |
+
"step": 105
|
| 201 |
},
|
| 202 |
{
|
| 203 |
+
"epoch": 0.59,
|
| 204 |
+
"grad_norm": 0.6197894215583801,
|
| 205 |
+
"learning_rate": 0.00017101063829787233,
|
| 206 |
+
"loss": 0.2738,
|
| 207 |
+
"step": 110
|
| 208 |
},
|
| 209 |
{
|
| 210 |
+
"epoch": 0.61,
|
| 211 |
+
"grad_norm": 0.5341880917549133,
|
| 212 |
+
"learning_rate": 0.0001696808510638298,
|
| 213 |
+
"loss": 0.2867,
|
| 214 |
+
"step": 115
|
| 215 |
},
|
| 216 |
{
|
| 217 |
+
"epoch": 0.64,
|
| 218 |
+
"grad_norm": 1.0934885740280151,
|
| 219 |
+
"learning_rate": 0.00016835106382978726,
|
| 220 |
+
"loss": 0.4105,
|
| 221 |
+
"step": 120
|
| 222 |
},
|
| 223 |
{
|
| 224 |
+
"epoch": 0.66,
|
| 225 |
+
"grad_norm": 0.5764520764350891,
|
| 226 |
+
"learning_rate": 0.00016702127659574468,
|
| 227 |
+
"loss": 0.3521,
|
| 228 |
+
"step": 125
|
| 229 |
},
|
| 230 |
{
|
| 231 |
+
"epoch": 0.66,
|
| 232 |
+
"eval_loss": 0.35300251841545105,
|
| 233 |
+
"eval_na_accuracy": 0.8108108043670654,
|
| 234 |
+
"eval_ordinal_accuracy": 0.42585551738739014,
|
| 235 |
+
"eval_ordinal_mae": 0.8441764116287231,
|
| 236 |
+
"eval_runtime": 12.3015,
|
| 237 |
+
"eval_samples_per_second": 24.387,
|
| 238 |
+
"eval_steps_per_second": 3.089,
|
| 239 |
+
"step": 125
|
| 240 |
},
|
| 241 |
{
|
| 242 |
+
"epoch": 0.69,
|
| 243 |
+
"grad_norm": 0.2090584933757782,
|
| 244 |
+
"learning_rate": 0.00016569148936170213,
|
| 245 |
+
"loss": 0.294,
|
| 246 |
+
"step": 130
|
| 247 |
},
|
| 248 |
{
|
| 249 |
+
"epoch": 0.72,
|
| 250 |
+
"grad_norm": 0.2995198369026184,
|
| 251 |
+
"learning_rate": 0.00016436170212765958,
|
| 252 |
+
"loss": 0.3099,
|
| 253 |
+
"step": 135
|
| 254 |
},
|
| 255 |
{
|
| 256 |
+
"epoch": 0.74,
|
| 257 |
+
"grad_norm": 0.41820451617240906,
|
| 258 |
+
"learning_rate": 0.00016303191489361703,
|
| 259 |
+
"loss": 0.4392,
|
| 260 |
+
"step": 140
|
|
|
|
|
|
|
|
|
|
|
|
|
| 261 |
},
|
| 262 |
{
|
| 263 |
+
"epoch": 0.77,
|
| 264 |
+
"grad_norm": 1.1886084079742432,
|
| 265 |
+
"learning_rate": 0.00016170212765957446,
|
| 266 |
+
"loss": 0.3725,
|
| 267 |
+
"step": 145
|
| 268 |
},
|
| 269 |
{
|
| 270 |
+
"epoch": 0.8,
|
| 271 |
+
"grad_norm": 0.8490511178970337,
|
| 272 |
+
"learning_rate": 0.00016037234042553194,
|
| 273 |
+
"loss": 0.3349,
|
| 274 |
+
"step": 150
|
| 275 |
},
|
| 276 |
{
|
| 277 |
+
"epoch": 0.8,
|
| 278 |
+
"eval_loss": 0.34123122692108154,
|
| 279 |
+
"eval_na_accuracy": 0.7297297120094299,
|
| 280 |
+
"eval_ordinal_accuracy": 0.4752851724624634,
|
| 281 |
+
"eval_ordinal_mae": 0.8015652298927307,
|
| 282 |
+
"eval_runtime": 12.509,
|
| 283 |
+
"eval_samples_per_second": 23.983,
|
| 284 |
+
"eval_steps_per_second": 3.038,
|
| 285 |
+
"step": 150
|
| 286 |
},
|
| 287 |
{
|
| 288 |
+
"epoch": 0.82,
|
| 289 |
+
"grad_norm": 0.4098907709121704,
|
| 290 |
+
"learning_rate": 0.00015904255319148936,
|
| 291 |
+
"loss": 0.2306,
|
| 292 |
+
"step": 155
|
| 293 |
},
|
| 294 |
{
|
| 295 |
+
"epoch": 0.85,
|
| 296 |
+
"grad_norm": 1.5454349517822266,
|
| 297 |
+
"learning_rate": 0.0001577127659574468,
|
| 298 |
+
"loss": 0.2382,
|
| 299 |
+
"step": 160
|
| 300 |
},
|
| 301 |
{
|
| 302 |
+
"epoch": 0.88,
|
| 303 |
+
"grad_norm": 0.61043381690979,
|
| 304 |
+
"learning_rate": 0.00015638297872340426,
|
| 305 |
+
"loss": 0.3448,
|
| 306 |
+
"step": 165
|
| 307 |
},
|
| 308 |
{
|
| 309 |
+
"epoch": 0.9,
|
| 310 |
+
"grad_norm": 0.7741652727127075,
|
| 311 |
+
"learning_rate": 0.00015505319148936171,
|
| 312 |
+
"loss": 0.2037,
|
| 313 |
+
"step": 170
|
| 314 |
},
|
| 315 |
{
|
| 316 |
+
"epoch": 0.93,
|
| 317 |
+
"grad_norm": 0.5108156204223633,
|
| 318 |
+
"learning_rate": 0.00015372340425531914,
|
| 319 |
+
"loss": 0.4612,
|
| 320 |
+
"step": 175
|
| 321 |
},
|
| 322 |
{
|
| 323 |
+
"epoch": 0.93,
|
| 324 |
+
"eval_loss": 0.36386463046073914,
|
| 325 |
+
"eval_na_accuracy": 0.5405405163764954,
|
| 326 |
+
"eval_ordinal_accuracy": 0.46768060326576233,
|
| 327 |
+
"eval_ordinal_mae": 0.7603853940963745,
|
| 328 |
+
"eval_runtime": 12.3807,
|
| 329 |
+
"eval_samples_per_second": 24.231,
|
| 330 |
+
"eval_steps_per_second": 3.069,
|
| 331 |
+
"step": 175
|
| 332 |
},
|
| 333 |
{
|
| 334 |
+
"epoch": 0.96,
|
| 335 |
+
"grad_norm": 0.8889250159263611,
|
| 336 |
+
"learning_rate": 0.00015239361702127662,
|
| 337 |
+
"loss": 0.4401,
|
| 338 |
+
"step": 180
|
| 339 |
},
|
| 340 |
{
|
| 341 |
+
"epoch": 0.98,
|
| 342 |
+
"grad_norm": 1.047706127166748,
|
| 343 |
+
"learning_rate": 0.00015106382978723407,
|
| 344 |
+
"loss": 0.2771,
|
| 345 |
+
"step": 185
|
|
|
|
|
|
|
|
|
|
|
|
|
| 346 |
},
|
| 347 |
{
|
| 348 |
+
"epoch": 1.01,
|
| 349 |
+
"grad_norm": 0.6727350354194641,
|
| 350 |
+
"learning_rate": 0.0001497340425531915,
|
| 351 |
+
"loss": 0.3726,
|
| 352 |
+
"step": 190
|
| 353 |
},
|
| 354 |
{
|
| 355 |
+
"epoch": 1.04,
|
| 356 |
+
"grad_norm": 0.4043642282485962,
|
| 357 |
+
"learning_rate": 0.00014840425531914894,
|
| 358 |
+
"loss": 0.3504,
|
| 359 |
+
"step": 195
|
| 360 |
},
|
| 361 |
{
|
| 362 |
+
"epoch": 1.06,
|
| 363 |
+
"grad_norm": 1.089250922203064,
|
| 364 |
+
"learning_rate": 0.0001470744680851064,
|
| 365 |
+
"loss": 0.2424,
|
| 366 |
+
"step": 200
|
| 367 |
},
|
| 368 |
{
|
| 369 |
+
"epoch": 1.06,
|
| 370 |
+
"eval_loss": 0.32972484827041626,
|
| 371 |
+
"eval_na_accuracy": 0.7027027010917664,
|
| 372 |
+
"eval_ordinal_accuracy": 0.48669201135635376,
|
| 373 |
+
"eval_ordinal_mae": 0.7117426991462708,
|
| 374 |
+
"eval_runtime": 11.8333,
|
| 375 |
+
"eval_samples_per_second": 25.352,
|
| 376 |
+
"eval_steps_per_second": 3.211,
|
| 377 |
+
"step": 200
|
| 378 |
},
|
| 379 |
{
|
| 380 |
+
"epoch": 1.09,
|
| 381 |
+
"grad_norm": 0.6453298926353455,
|
| 382 |
+
"learning_rate": 0.00014574468085106382,
|
| 383 |
+
"loss": 0.2115,
|
| 384 |
+
"step": 205
|
| 385 |
},
|
| 386 |
{
|
| 387 |
+
"epoch": 1.12,
|
| 388 |
+
"grad_norm": 0.7021524310112,
|
| 389 |
+
"learning_rate": 0.00014441489361702127,
|
| 390 |
+
"loss": 0.2281,
|
| 391 |
+
"step": 210
|
| 392 |
},
|
| 393 |
{
|
| 394 |
+
"epoch": 1.14,
|
| 395 |
+
"grad_norm": 0.7665510773658752,
|
| 396 |
+
"learning_rate": 0.00014308510638297875,
|
| 397 |
+
"loss": 0.2048,
|
| 398 |
+
"step": 215
|
| 399 |
},
|
| 400 |
{
|
| 401 |
+
"epoch": 1.17,
|
| 402 |
+
"grad_norm": 1.2339574098587036,
|
| 403 |
+
"learning_rate": 0.00014175531914893617,
|
| 404 |
+
"loss": 0.2344,
|
| 405 |
+
"step": 220
|
| 406 |
},
|
| 407 |
{
|
| 408 |
+
"epoch": 1.2,
|
| 409 |
+
"grad_norm": 2.540107011795044,
|
| 410 |
+
"learning_rate": 0.00014042553191489363,
|
| 411 |
+
"loss": 0.2928,
|
| 412 |
+
"step": 225
|
| 413 |
},
|
| 414 |
{
|
| 415 |
+
"epoch": 1.2,
|
| 416 |
+
"eval_loss": 0.3493916690349579,
|
| 417 |
+
"eval_na_accuracy": 0.6756756901741028,
|
| 418 |
+
"eval_ordinal_accuracy": 0.5285171270370483,
|
| 419 |
+
"eval_ordinal_mae": 0.6955077052116394,
|
| 420 |
+
"eval_runtime": 10.8667,
|
| 421 |
+
"eval_samples_per_second": 27.607,
|
| 422 |
+
"eval_steps_per_second": 3.497,
|
| 423 |
+
"step": 225
|
| 424 |
},
|
| 425 |
{
|
| 426 |
+
"epoch": 1.22,
|
| 427 |
+
"grad_norm": 0.7262032628059387,
|
| 428 |
+
"learning_rate": 0.00013909574468085108,
|
| 429 |
+
"loss": 0.3973,
|
| 430 |
+
"step": 230
|
|
|
|
|
|
|
|
|
|
|
|
|
| 431 |
},
|
| 432 |
{
|
| 433 |
+
"epoch": 1.25,
|
| 434 |
+
"grad_norm": 0.28402724862098694,
|
| 435 |
+
"learning_rate": 0.0001377659574468085,
|
| 436 |
+
"loss": 0.218,
|
| 437 |
+
"step": 235
|
| 438 |
},
|
| 439 |
{
|
| 440 |
+
"epoch": 1.28,
|
| 441 |
+
"grad_norm": 1.5170676708221436,
|
| 442 |
+
"learning_rate": 0.00013643617021276595,
|
| 443 |
+
"loss": 0.227,
|
| 444 |
+
"step": 240
|
| 445 |
},
|
| 446 |
{
|
| 447 |
+
"epoch": 1.3,
|
| 448 |
+
"grad_norm": 0.35739636421203613,
|
| 449 |
+
"learning_rate": 0.0001351063829787234,
|
| 450 |
+
"loss": 0.2275,
|
| 451 |
+
"step": 245
|
| 452 |
},
|
| 453 |
{
|
| 454 |
+
"epoch": 1.33,
|
| 455 |
+
"grad_norm": 0.5471745133399963,
|
| 456 |
+
"learning_rate": 0.00013377659574468086,
|
| 457 |
+
"loss": 0.2436,
|
| 458 |
+
"step": 250
|
| 459 |
},
|
| 460 |
{
|
| 461 |
+
"epoch": 1.33,
|
| 462 |
+
"eval_loss": 0.3037484288215637,
|
| 463 |
+
"eval_na_accuracy": 0.7297297120094299,
|
| 464 |
+
"eval_ordinal_accuracy": 0.5285171270370483,
|
| 465 |
+
"eval_ordinal_mae": 0.6723113059997559,
|
| 466 |
+
"eval_runtime": 12.062,
|
| 467 |
+
"eval_samples_per_second": 24.871,
|
| 468 |
+
"eval_steps_per_second": 3.15,
|
| 469 |
+
"step": 250
|
| 470 |
},
|
| 471 |
{
|
| 472 |
+
"epoch": 1.36,
|
| 473 |
+
"grad_norm": 0.41173043847084045,
|
| 474 |
+
"learning_rate": 0.0001324468085106383,
|
| 475 |
+
"loss": 0.2408,
|
| 476 |
+
"step": 255
|
| 477 |
},
|
| 478 |
{
|
| 479 |
+
"epoch": 1.38,
|
| 480 |
+
"grad_norm": 0.8529615998268127,
|
| 481 |
+
"learning_rate": 0.00013111702127659576,
|
| 482 |
+
"loss": 0.2935,
|
| 483 |
+
"step": 260
|
| 484 |
},
|
| 485 |
{
|
| 486 |
+
"epoch": 1.41,
|
| 487 |
+
"grad_norm": 1.3896653652191162,
|
| 488 |
+
"learning_rate": 0.00012978723404255318,
|
| 489 |
+
"loss": 0.2547,
|
| 490 |
+
"step": 265
|
| 491 |
},
|
| 492 |
{
|
| 493 |
+
"epoch": 1.44,
|
| 494 |
+
"grad_norm": 0.30819597840309143,
|
| 495 |
+
"learning_rate": 0.00012845744680851063,
|
| 496 |
+
"loss": 0.3372,
|
| 497 |
+
"step": 270
|
| 498 |
},
|
| 499 |
{
|
| 500 |
+
"epoch": 1.46,
|
| 501 |
+
"grad_norm": 2.0342652797698975,
|
| 502 |
+
"learning_rate": 0.00012712765957446809,
|
| 503 |
+
"loss": 0.2776,
|
| 504 |
+
"step": 275
|
| 505 |
},
|
| 506 |
{
|
| 507 |
+
"epoch": 1.46,
|
| 508 |
+
"eval_loss": 0.3365646004676819,
|
| 509 |
+
"eval_na_accuracy": 0.5945945978164673,
|
| 510 |
+
"eval_ordinal_accuracy": 0.517110288143158,
|
| 511 |
+
"eval_ordinal_mae": 0.672748863697052,
|
| 512 |
+
"eval_runtime": 11.4536,
|
| 513 |
+
"eval_samples_per_second": 26.193,
|
| 514 |
+
"eval_steps_per_second": 3.318,
|
| 515 |
+
"step": 275
|
| 516 |
},
|
| 517 |
{
|
| 518 |
+
"epoch": 1.46,
|
| 519 |
+
"step": 275,
|
| 520 |
+
"total_flos": 3.403570199991091e+17,
|
| 521 |
+
"train_loss": 0.32911812565543436,
|
| 522 |
+
"train_runtime": 761.2747,
|
| 523 |
+
"train_samples_per_second": 15.763,
|
| 524 |
+
"train_steps_per_second": 0.988
|
| 525 |
}
|
| 526 |
],
|
| 527 |
+
"logging_steps": 5,
|
| 528 |
+
"max_steps": 752,
|
| 529 |
"num_input_tokens_seen": 0,
|
| 530 |
"num_train_epochs": 4,
|
| 531 |
+
"save_steps": 25,
|
| 532 |
+
"total_flos": 3.403570199991091e+17,
|
| 533 |
"train_batch_size": 16,
|
| 534 |
"trial_name": null,
|
| 535 |
"trial_params": null
|