🍻 cheers
Browse files- README.md +6 -5
- all_results.json +9 -9
- eval_results.json +5 -5
- runs/May09_07-16-20_361db62a36de/events.out.tfevents.1715259917.361db62a36de.34.1 +3 -0
- train_results.json +4 -4
- trainer_state.json +1880 -356
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 |
datasets:
|
| 7 |
- imagefolder
|
|
@@ -14,7 +15,7 @@ model-index:
|
|
| 14 |
name: Image Classification
|
| 15 |
type: image-classification
|
| 16 |
dataset:
|
| 17 |
-
name:
|
| 18 |
type: imagefolder
|
| 19 |
config: default
|
| 20 |
split: train
|
|
@@ -22,7 +23,7 @@ model-index:
|
|
| 22 |
metrics:
|
| 23 |
- name: Accuracy
|
| 24 |
type: accuracy
|
| 25 |
-
value: 0.
|
| 26 |
---
|
| 27 |
|
| 28 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
@@ -30,10 +31,10 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 30 |
|
| 31 |
# Action_model
|
| 32 |
|
| 33 |
-
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
|
| 34 |
It achieves the following results on the evaluation set:
|
| 35 |
-
- Loss: 0.
|
| 36 |
-
- Accuracy: 0.
|
| 37 |
|
| 38 |
## Model description
|
| 39 |
|
|
|
|
| 2 |
license: apache-2.0
|
| 3 |
base_model: google/vit-base-patch16-224-in21k
|
| 4 |
tags:
|
| 5 |
+
- image-classification
|
| 6 |
- generated_from_trainer
|
| 7 |
datasets:
|
| 8 |
- imagefolder
|
|
|
|
| 15 |
name: Image Classification
|
| 16 |
type: image-classification
|
| 17 |
dataset:
|
| 18 |
+
name: action_class
|
| 19 |
type: imagefolder
|
| 20 |
config: default
|
| 21 |
split: train
|
|
|
|
| 23 |
metrics:
|
| 24 |
- name: Accuracy
|
| 25 |
type: accuracy
|
| 26 |
+
value: 0.8576449912126538
|
| 27 |
---
|
| 28 |
|
| 29 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
|
|
| 31 |
|
| 32 |
# Action_model
|
| 33 |
|
| 34 |
+
This model is a fine-tuned version of [google/vit-base-patch16-224-in21k](https://huggingface.co/google/vit-base-patch16-224-in21k) on the action_class dataset.
|
| 35 |
It achieves the following results on the evaluation set:
|
| 36 |
+
- Loss: 0.4589
|
| 37 |
+
- Accuracy: 0.8576
|
| 38 |
|
| 39 |
## Model description
|
| 40 |
|
all_results.json
CHANGED
|
@@ -1,13 +1,13 @@
|
|
| 1 |
{
|
| 2 |
"epoch": 10.0,
|
| 3 |
-
"eval_accuracy": 0.
|
| 4 |
-
"eval_loss": 0.
|
| 5 |
-
"eval_runtime": 7.
|
| 6 |
-
"eval_samples_per_second":
|
| 7 |
-
"eval_steps_per_second": 9.
|
| 8 |
"total_flos": 3.3230947683690086e+18,
|
| 9 |
-
"train_loss": 0.
|
| 10 |
-
"train_runtime":
|
| 11 |
-
"train_samples_per_second":
|
| 12 |
-
"train_steps_per_second": 1.
|
| 13 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"epoch": 10.0,
|
| 3 |
+
"eval_accuracy": 0.8576449912126538,
|
| 4 |
+
"eval_loss": 0.4589254856109619,
|
| 5 |
+
"eval_runtime": 7.9329,
|
| 6 |
+
"eval_samples_per_second": 71.727,
|
| 7 |
+
"eval_steps_per_second": 9.076,
|
| 8 |
"total_flos": 3.3230947683690086e+18,
|
| 9 |
+
"train_loss": 0.45543073504718384,
|
| 10 |
+
"train_runtime": 1353.2313,
|
| 11 |
+
"train_samples_per_second": 31.687,
|
| 12 |
+
"train_steps_per_second": 1.98
|
| 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": 7.
|
| 6 |
-
"eval_samples_per_second":
|
| 7 |
-
"eval_steps_per_second": 9.
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"epoch": 10.0,
|
| 3 |
+
"eval_accuracy": 0.8576449912126538,
|
| 4 |
+
"eval_loss": 0.4589254856109619,
|
| 5 |
+
"eval_runtime": 7.9329,
|
| 6 |
+
"eval_samples_per_second": 71.727,
|
| 7 |
+
"eval_steps_per_second": 9.076
|
| 8 |
}
|
runs/May09_07-16-20_361db62a36de/events.out.tfevents.1715259917.361db62a36de.34.1
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:649da8ed98001a4af8b5268f38716de03351fd30b732596dea56d6bc9be2f61e
|
| 3 |
+
size 411
|
train_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
"epoch": 10.0,
|
| 3 |
"total_flos": 3.3230947683690086e+18,
|
| 4 |
-
"train_loss": 0.
|
| 5 |
-
"train_runtime":
|
| 6 |
-
"train_samples_per_second":
|
| 7 |
-
"train_steps_per_second": 1.
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"epoch": 10.0,
|
| 3 |
"total_flos": 3.3230947683690086e+18,
|
| 4 |
+
"train_loss": 0.45543073504718384,
|
| 5 |
+
"train_runtime": 1353.2313,
|
| 6 |
+
"train_samples_per_second": 31.687,
|
| 7 |
+
"train_steps_per_second": 1.98
|
| 8 |
}
|
trainer_state.json
CHANGED
|
@@ -1,616 +1,2140 @@
|
|
| 1 |
{
|
| 2 |
-
"best_metric": 0.
|
| 3 |
-
"best_model_checkpoint": "Action_model/checkpoint-
|
| 4 |
"epoch": 10.0,
|
| 5 |
"eval_steps": 100,
|
| 6 |
-
"global_step":
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
-
"epoch": 0.
|
| 13 |
-
"grad_norm": 1.
|
| 14 |
-
"learning_rate": 9.
|
| 15 |
-
"loss":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 16 |
"step": 20
|
| 17 |
},
|
| 18 |
{
|
| 19 |
-
"epoch": 0.
|
| 20 |
-
"grad_norm":
|
| 21 |
-
"learning_rate": 9.
|
| 22 |
-
"loss":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 23 |
"step": 40
|
| 24 |
},
|
| 25 |
{
|
| 26 |
-
"epoch": 0.
|
| 27 |
-
"grad_norm": 2.
|
| 28 |
-
"learning_rate": 9.
|
| 29 |
-
"loss":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 30 |
"step": 60
|
| 31 |
},
|
| 32 |
{
|
| 33 |
-
"epoch": 0.
|
| 34 |
-
"grad_norm": 2.
|
| 35 |
-
"learning_rate": 9.
|
| 36 |
-
"loss":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 37 |
"step": 80
|
| 38 |
},
|
| 39 |
{
|
| 40 |
-
"epoch": 0.
|
| 41 |
-
"grad_norm":
|
| 42 |
-
"learning_rate": 9.
|
| 43 |
-
"loss":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 44 |
"step": 100
|
| 45 |
},
|
| 46 |
{
|
| 47 |
-
"epoch": 0.
|
| 48 |
-
"eval_accuracy": 0.
|
| 49 |
-
"eval_loss":
|
| 50 |
-
"eval_runtime":
|
| 51 |
-
"eval_samples_per_second":
|
| 52 |
-
"eval_steps_per_second":
|
| 53 |
"step": 100
|
| 54 |
},
|
| 55 |
{
|
| 56 |
-
"epoch": 0.
|
| 57 |
-
"grad_norm":
|
| 58 |
-
"learning_rate": 9.
|
| 59 |
-
"loss":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 60 |
"step": 120
|
| 61 |
},
|
| 62 |
{
|
| 63 |
-
"epoch":
|
| 64 |
-
"grad_norm": 3.
|
| 65 |
-
"learning_rate":
|
| 66 |
-
"loss":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 67 |
"step": 140
|
| 68 |
},
|
| 69 |
{
|
| 70 |
-
"epoch":
|
| 71 |
-
"grad_norm":
|
| 72 |
-
"learning_rate":
|
| 73 |
-
"loss":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 74 |
"step": 160
|
| 75 |
},
|
| 76 |
{
|
| 77 |
-
"epoch":
|
| 78 |
-
"grad_norm": 4.
|
| 79 |
-
"learning_rate":
|
| 80 |
-
"loss":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 81 |
"step": 180
|
| 82 |
},
|
| 83 |
{
|
| 84 |
-
"epoch":
|
| 85 |
-
"grad_norm":
|
| 86 |
-
"learning_rate":
|
| 87 |
-
"loss":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 88 |
"step": 200
|
| 89 |
},
|
| 90 |
{
|
| 91 |
-
"epoch":
|
| 92 |
-
"eval_accuracy": 0.
|
| 93 |
-
"eval_loss": 0.
|
| 94 |
-
"eval_runtime": 7.
|
| 95 |
-
"eval_samples_per_second":
|
| 96 |
-
"eval_steps_per_second": 9.
|
| 97 |
"step": 200
|
| 98 |
},
|
| 99 |
{
|
| 100 |
-
"epoch":
|
| 101 |
-
"grad_norm":
|
| 102 |
-
"learning_rate":
|
| 103 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 104 |
"step": 220
|
| 105 |
},
|
| 106 |
{
|
| 107 |
-
"epoch":
|
| 108 |
-
"grad_norm": 3.
|
| 109 |
-
"learning_rate":
|
| 110 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 111 |
"step": 240
|
| 112 |
},
|
| 113 |
{
|
| 114 |
-
"epoch":
|
| 115 |
-
"grad_norm": 4.
|
| 116 |
-
"learning_rate":
|
| 117 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 118 |
"step": 260
|
| 119 |
},
|
| 120 |
{
|
| 121 |
-
"epoch":
|
| 122 |
-
"grad_norm":
|
| 123 |
-
"learning_rate":
|
| 124 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 125 |
"step": 280
|
| 126 |
},
|
| 127 |
{
|
| 128 |
-
"epoch":
|
| 129 |
-
"grad_norm":
|
| 130 |
-
"learning_rate":
|
| 131 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 132 |
"step": 300
|
| 133 |
},
|
| 134 |
{
|
| 135 |
-
"epoch":
|
| 136 |
-
"eval_accuracy": 0.
|
| 137 |
-
"eval_loss": 0.
|
| 138 |
-
"eval_runtime": 7.
|
| 139 |
-
"eval_samples_per_second":
|
| 140 |
-
"eval_steps_per_second": 9.
|
| 141 |
"step": 300
|
| 142 |
},
|
| 143 |
{
|
| 144 |
-
"epoch":
|
| 145 |
-
"grad_norm": 3.
|
| 146 |
-
"learning_rate":
|
| 147 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 148 |
"step": 320
|
| 149 |
},
|
| 150 |
{
|
| 151 |
-
"epoch":
|
| 152 |
-
"grad_norm": 3.
|
| 153 |
-
"learning_rate":
|
| 154 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 155 |
"step": 340
|
| 156 |
},
|
| 157 |
{
|
| 158 |
-
"epoch":
|
| 159 |
-
"grad_norm":
|
| 160 |
-
"learning_rate":
|
| 161 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 162 |
"step": 360
|
| 163 |
},
|
| 164 |
{
|
| 165 |
-
"epoch":
|
| 166 |
-
"grad_norm": 3.
|
| 167 |
-
"learning_rate":
|
| 168 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 169 |
"step": 380
|
| 170 |
},
|
| 171 |
{
|
| 172 |
-
"epoch":
|
| 173 |
-
"grad_norm":
|
| 174 |
-
"learning_rate":
|
| 175 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 176 |
"step": 400
|
| 177 |
},
|
| 178 |
{
|
| 179 |
-
"epoch":
|
| 180 |
-
"eval_accuracy": 0.
|
| 181 |
-
"eval_loss": 0.
|
| 182 |
-
"eval_runtime": 7.
|
| 183 |
-
"eval_samples_per_second":
|
| 184 |
-
"eval_steps_per_second": 9.
|
| 185 |
"step": 400
|
| 186 |
},
|
| 187 |
{
|
| 188 |
-
"epoch":
|
| 189 |
-
"grad_norm":
|
| 190 |
-
"learning_rate":
|
| 191 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 192 |
"step": 420
|
| 193 |
},
|
| 194 |
{
|
| 195 |
-
"epoch":
|
| 196 |
-
"grad_norm":
|
| 197 |
-
"learning_rate":
|
| 198 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 199 |
"step": 440
|
| 200 |
},
|
| 201 |
{
|
| 202 |
-
"epoch":
|
| 203 |
-
"grad_norm":
|
| 204 |
-
"learning_rate":
|
| 205 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 206 |
"step": 460
|
| 207 |
},
|
| 208 |
{
|
| 209 |
-
"epoch":
|
| 210 |
-
"grad_norm":
|
| 211 |
-
"learning_rate":
|
| 212 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 213 |
"step": 480
|
| 214 |
},
|
| 215 |
{
|
| 216 |
-
"epoch":
|
| 217 |
-
"grad_norm":
|
| 218 |
-
"learning_rate":
|
| 219 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 220 |
"step": 500
|
| 221 |
},
|
| 222 |
{
|
| 223 |
-
"epoch":
|
| 224 |
-
"eval_accuracy": 0.
|
| 225 |
-
"eval_loss": 0.
|
| 226 |
-
"eval_runtime": 7.
|
| 227 |
-
"eval_samples_per_second":
|
| 228 |
-
"eval_steps_per_second": 9.
|
| 229 |
"step": 500
|
| 230 |
},
|
| 231 |
{
|
| 232 |
-
"epoch":
|
| 233 |
-
"grad_norm":
|
| 234 |
-
"learning_rate":
|
| 235 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 236 |
"step": 520
|
| 237 |
},
|
| 238 |
{
|
| 239 |
-
"epoch":
|
| 240 |
-
"grad_norm":
|
| 241 |
-
"learning_rate":
|
| 242 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 243 |
"step": 540
|
| 244 |
},
|
| 245 |
{
|
| 246 |
-
"epoch":
|
| 247 |
-
"grad_norm":
|
| 248 |
-
"learning_rate":
|
| 249 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 250 |
"step": 560
|
| 251 |
},
|
| 252 |
{
|
| 253 |
-
"epoch":
|
| 254 |
-
"grad_norm":
|
| 255 |
-
"learning_rate":
|
| 256 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 257 |
"step": 580
|
| 258 |
},
|
| 259 |
{
|
| 260 |
-
"epoch":
|
| 261 |
-
"grad_norm":
|
| 262 |
-
"learning_rate":
|
| 263 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 264 |
"step": 600
|
| 265 |
},
|
| 266 |
{
|
| 267 |
-
"epoch":
|
| 268 |
-
"eval_accuracy": 0.
|
| 269 |
-
"eval_loss": 0.
|
| 270 |
-
"eval_runtime": 7.
|
| 271 |
-
"eval_samples_per_second":
|
| 272 |
-
"eval_steps_per_second": 9.
|
| 273 |
"step": 600
|
| 274 |
},
|
| 275 |
{
|
| 276 |
-
"epoch":
|
| 277 |
-
"grad_norm":
|
| 278 |
-
"learning_rate":
|
| 279 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 280 |
"step": 620
|
| 281 |
},
|
| 282 |
{
|
| 283 |
-
"epoch":
|
| 284 |
-
"grad_norm":
|
| 285 |
-
"learning_rate":
|
| 286 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 287 |
"step": 640
|
| 288 |
},
|
| 289 |
{
|
| 290 |
-
"epoch":
|
| 291 |
-
"grad_norm":
|
| 292 |
-
"learning_rate":
|
| 293 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 294 |
"step": 660
|
| 295 |
},
|
| 296 |
{
|
| 297 |
-
"epoch": 5
|
| 298 |
-
"grad_norm": 3.
|
| 299 |
-
"learning_rate":
|
| 300 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 301 |
"step": 680
|
| 302 |
},
|
| 303 |
{
|
| 304 |
-
"epoch":
|
| 305 |
-
"grad_norm": 4.
|
| 306 |
-
"learning_rate":
|
| 307 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 308 |
"step": 700
|
| 309 |
},
|
| 310 |
{
|
| 311 |
-
"epoch":
|
| 312 |
-
"eval_accuracy": 0.
|
| 313 |
-
"eval_loss": 0.
|
| 314 |
-
"eval_runtime": 7.
|
| 315 |
-
"eval_samples_per_second": 72.
|
| 316 |
-
"eval_steps_per_second": 9.
|
| 317 |
"step": 700
|
| 318 |
},
|
| 319 |
{
|
| 320 |
-
"epoch":
|
| 321 |
-
"grad_norm":
|
| 322 |
-
"learning_rate":
|
| 323 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 324 |
"step": 720
|
| 325 |
},
|
| 326 |
{
|
| 327 |
-
"epoch":
|
| 328 |
-
"grad_norm":
|
| 329 |
-
"learning_rate":
|
| 330 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 331 |
"step": 740
|
| 332 |
},
|
| 333 |
{
|
| 334 |
-
"epoch":
|
| 335 |
-
"grad_norm":
|
| 336 |
-
"learning_rate":
|
| 337 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 338 |
"step": 760
|
| 339 |
},
|
| 340 |
{
|
| 341 |
-
"epoch":
|
| 342 |
-
"grad_norm":
|
| 343 |
-
"learning_rate":
|
| 344 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 345 |
"step": 780
|
| 346 |
},
|
| 347 |
{
|
| 348 |
-
"epoch":
|
| 349 |
-
"grad_norm":
|
| 350 |
-
"learning_rate":
|
| 351 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 352 |
"step": 800
|
| 353 |
},
|
| 354 |
{
|
| 355 |
-
"epoch":
|
| 356 |
-
"eval_accuracy": 0.
|
| 357 |
-
"eval_loss": 0.
|
| 358 |
-
"eval_runtime": 7.
|
| 359 |
-
"eval_samples_per_second":
|
| 360 |
-
"eval_steps_per_second": 9.
|
| 361 |
"step": 800
|
| 362 |
},
|
| 363 |
{
|
| 364 |
-
"epoch":
|
| 365 |
-
"grad_norm":
|
| 366 |
-
"learning_rate":
|
| 367 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 368 |
"step": 820
|
| 369 |
},
|
| 370 |
{
|
| 371 |
-
"epoch":
|
| 372 |
-
"grad_norm":
|
| 373 |
-
"learning_rate":
|
| 374 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 375 |
"step": 840
|
| 376 |
},
|
| 377 |
{
|
| 378 |
-
"epoch":
|
| 379 |
-
"grad_norm":
|
| 380 |
-
"learning_rate":
|
| 381 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 382 |
"step": 860
|
| 383 |
},
|
| 384 |
{
|
| 385 |
-
"epoch":
|
| 386 |
-
"grad_norm": 2.
|
| 387 |
-
"learning_rate":
|
| 388 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 389 |
"step": 880
|
| 390 |
},
|
| 391 |
{
|
| 392 |
-
"epoch":
|
| 393 |
-
"grad_norm": 6.
|
| 394 |
-
"learning_rate":
|
| 395 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 396 |
"step": 900
|
| 397 |
},
|
| 398 |
{
|
| 399 |
-
"epoch":
|
| 400 |
-
"eval_accuracy": 0.
|
| 401 |
-
"eval_loss": 0.
|
| 402 |
-
"eval_runtime": 7.
|
| 403 |
-
"eval_samples_per_second": 73.
|
| 404 |
-
"eval_steps_per_second": 9.
|
| 405 |
"step": 900
|
| 406 |
},
|
| 407 |
{
|
| 408 |
-
"epoch":
|
| 409 |
-
"grad_norm":
|
| 410 |
-
"learning_rate":
|
| 411 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 412 |
"step": 920
|
| 413 |
},
|
| 414 |
{
|
| 415 |
-
"epoch":
|
| 416 |
-
"grad_norm":
|
| 417 |
-
"learning_rate":
|
| 418 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 419 |
"step": 940
|
| 420 |
},
|
| 421 |
{
|
| 422 |
-
"epoch":
|
| 423 |
-
"grad_norm":
|
| 424 |
-
"learning_rate":
|
| 425 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 426 |
"step": 960
|
| 427 |
},
|
| 428 |
{
|
| 429 |
-
"epoch":
|
| 430 |
-
"grad_norm":
|
| 431 |
-
"learning_rate":
|
| 432 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 433 |
"step": 980
|
| 434 |
},
|
| 435 |
{
|
| 436 |
-
"epoch":
|
| 437 |
-
"grad_norm":
|
| 438 |
-
"learning_rate":
|
| 439 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 440 |
"step": 1000
|
| 441 |
},
|
| 442 |
{
|
| 443 |
-
"epoch":
|
| 444 |
-
"eval_accuracy": 0.
|
| 445 |
-
"eval_loss": 0.
|
| 446 |
-
"eval_runtime": 7.
|
| 447 |
-
"eval_samples_per_second":
|
| 448 |
-
"eval_steps_per_second": 9.
|
| 449 |
"step": 1000
|
| 450 |
},
|
| 451 |
{
|
| 452 |
-
"epoch":
|
| 453 |
-
"grad_norm":
|
| 454 |
-
"learning_rate":
|
| 455 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 456 |
"step": 1020
|
| 457 |
},
|
| 458 |
{
|
| 459 |
-
"epoch":
|
| 460 |
-
"grad_norm": 4.
|
| 461 |
-
"learning_rate":
|
| 462 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 463 |
"step": 1040
|
| 464 |
},
|
| 465 |
{
|
| 466 |
-
"epoch":
|
| 467 |
-
"grad_norm":
|
| 468 |
-
"learning_rate":
|
| 469 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 470 |
"step": 1060
|
| 471 |
},
|
| 472 |
{
|
| 473 |
-
"epoch":
|
| 474 |
-
"grad_norm":
|
| 475 |
-
"learning_rate":
|
| 476 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 477 |
"step": 1080
|
| 478 |
},
|
| 479 |
{
|
| 480 |
-
"epoch":
|
| 481 |
-
"grad_norm":
|
| 482 |
-
"learning_rate":
|
| 483 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 484 |
"step": 1100
|
| 485 |
},
|
| 486 |
{
|
| 487 |
-
"epoch":
|
| 488 |
-
"eval_accuracy": 0.
|
| 489 |
-
"eval_loss": 0.
|
| 490 |
-
"eval_runtime": 7.
|
| 491 |
-
"eval_samples_per_second":
|
| 492 |
-
"eval_steps_per_second": 9.
|
| 493 |
"step": 1100
|
| 494 |
},
|
| 495 |
{
|
| 496 |
-
"epoch":
|
| 497 |
-
"grad_norm":
|
| 498 |
-
"learning_rate":
|
| 499 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 500 |
"step": 1120
|
| 501 |
},
|
| 502 |
{
|
| 503 |
-
"epoch":
|
| 504 |
-
"grad_norm": 1.
|
| 505 |
-
"learning_rate":
|
| 506 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 507 |
"step": 1140
|
| 508 |
},
|
| 509 |
{
|
| 510 |
-
"epoch":
|
| 511 |
-
"grad_norm":
|
| 512 |
-
"learning_rate":
|
| 513 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 514 |
"step": 1160
|
| 515 |
},
|
| 516 |
{
|
| 517 |
-
"epoch":
|
| 518 |
-
"grad_norm":
|
| 519 |
-
"learning_rate":
|
| 520 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 521 |
"step": 1180
|
| 522 |
},
|
| 523 |
{
|
| 524 |
-
"epoch":
|
| 525 |
-
"grad_norm": 4.
|
| 526 |
-
"learning_rate":
|
| 527 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 528 |
"step": 1200
|
| 529 |
},
|
| 530 |
{
|
| 531 |
-
"epoch":
|
| 532 |
"eval_accuracy": 0.8576449912126538,
|
| 533 |
-
"eval_loss": 0.
|
| 534 |
-
"eval_runtime": 7.
|
| 535 |
-
"eval_samples_per_second":
|
| 536 |
-
"eval_steps_per_second": 9.
|
| 537 |
"step": 1200
|
| 538 |
},
|
| 539 |
{
|
| 540 |
-
"epoch":
|
| 541 |
-
"grad_norm":
|
| 542 |
-
"learning_rate":
|
| 543 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 544 |
"step": 1220
|
| 545 |
},
|
| 546 |
{
|
| 547 |
-
"epoch":
|
| 548 |
-
"grad_norm":
|
| 549 |
-
"learning_rate":
|
| 550 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 551 |
"step": 1240
|
| 552 |
},
|
| 553 |
{
|
| 554 |
-
"epoch":
|
| 555 |
-
"grad_norm":
|
| 556 |
-
"learning_rate":
|
| 557 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 558 |
"step": 1260
|
| 559 |
},
|
| 560 |
{
|
| 561 |
-
"epoch":
|
| 562 |
-
"grad_norm":
|
| 563 |
-
"learning_rate":
|
| 564 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 565 |
"step": 1280
|
| 566 |
},
|
| 567 |
{
|
| 568 |
-
"epoch":
|
| 569 |
-
"grad_norm":
|
| 570 |
-
"learning_rate":
|
| 571 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 572 |
"step": 1300
|
| 573 |
},
|
| 574 |
{
|
| 575 |
-
"epoch":
|
| 576 |
-
"eval_accuracy": 0.
|
| 577 |
-
"eval_loss": 0.
|
| 578 |
-
"eval_runtime": 7.
|
| 579 |
-
"eval_samples_per_second":
|
| 580 |
-
"eval_steps_per_second": 9.
|
| 581 |
"step": 1300
|
| 582 |
},
|
| 583 |
{
|
| 584 |
-
"epoch":
|
| 585 |
-
"grad_norm": 2.
|
| 586 |
-
"learning_rate":
|
| 587 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 588 |
"step": 1320
|
| 589 |
},
|
| 590 |
{
|
| 591 |
-
"epoch":
|
| 592 |
-
"grad_norm":
|
| 593 |
-
"learning_rate":
|
| 594 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 595 |
"step": 1340
|
| 596 |
},
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 597 |
{
|
| 598 |
"epoch": 10.0,
|
| 599 |
-
"step":
|
| 600 |
"total_flos": 3.3230947683690086e+18,
|
| 601 |
-
"train_loss": 0.
|
| 602 |
-
"train_runtime":
|
| 603 |
-
"train_samples_per_second":
|
| 604 |
-
"train_steps_per_second": 1.
|
| 605 |
}
|
| 606 |
],
|
| 607 |
-
"logging_steps":
|
| 608 |
-
"max_steps":
|
| 609 |
"num_input_tokens_seen": 0,
|
| 610 |
"num_train_epochs": 10,
|
| 611 |
"save_steps": 100,
|
| 612 |
"total_flos": 3.3230947683690086e+18,
|
| 613 |
-
"train_batch_size":
|
| 614 |
"trial_name": null,
|
| 615 |
"trial_params": null
|
| 616 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_metric": 0.4589254856109619,
|
| 3 |
+
"best_model_checkpoint": "Action_model/checkpoint-1500",
|
| 4 |
"epoch": 10.0,
|
| 5 |
"eval_steps": 100,
|
| 6 |
+
"global_step": 2680,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
+
"epoch": 0.04,
|
| 13 |
+
"grad_norm": 1.7369908094406128,
|
| 14 |
+
"learning_rate": 9.96268656716418e-05,
|
| 15 |
+
"loss": 2.2759,
|
| 16 |
+
"step": 10
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.07,
|
| 20 |
+
"grad_norm": 1.753720998764038,
|
| 21 |
+
"learning_rate": 9.925373134328359e-05,
|
| 22 |
+
"loss": 2.1743,
|
| 23 |
"step": 20
|
| 24 |
},
|
| 25 |
{
|
| 26 |
+
"epoch": 0.11,
|
| 27 |
+
"grad_norm": 1.8532754182815552,
|
| 28 |
+
"learning_rate": 9.888059701492539e-05,
|
| 29 |
+
"loss": 2.0233,
|
| 30 |
+
"step": 30
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.15,
|
| 34 |
+
"grad_norm": 2.195688486099243,
|
| 35 |
+
"learning_rate": 9.850746268656717e-05,
|
| 36 |
+
"loss": 1.8293,
|
| 37 |
"step": 40
|
| 38 |
},
|
| 39 |
{
|
| 40 |
+
"epoch": 0.19,
|
| 41 |
+
"grad_norm": 2.392077684402466,
|
| 42 |
+
"learning_rate": 9.813432835820896e-05,
|
| 43 |
+
"loss": 1.7307,
|
| 44 |
+
"step": 50
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.22,
|
| 48 |
+
"grad_norm": 2.851775646209717,
|
| 49 |
+
"learning_rate": 9.776119402985075e-05,
|
| 50 |
+
"loss": 1.5716,
|
| 51 |
"step": 60
|
| 52 |
},
|
| 53 |
{
|
| 54 |
+
"epoch": 0.26,
|
| 55 |
+
"grad_norm": 2.2557411193847656,
|
| 56 |
+
"learning_rate": 9.738805970149254e-05,
|
| 57 |
+
"loss": 1.4694,
|
| 58 |
+
"step": 70
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.3,
|
| 62 |
+
"grad_norm": 2.4612302780151367,
|
| 63 |
+
"learning_rate": 9.701492537313434e-05,
|
| 64 |
+
"loss": 1.3609,
|
| 65 |
"step": 80
|
| 66 |
},
|
| 67 |
{
|
| 68 |
+
"epoch": 0.34,
|
| 69 |
+
"grad_norm": 2.7514560222625732,
|
| 70 |
+
"learning_rate": 9.664179104477612e-05,
|
| 71 |
+
"loss": 1.2871,
|
| 72 |
+
"step": 90
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.37,
|
| 76 |
+
"grad_norm": 3.6256659030914307,
|
| 77 |
+
"learning_rate": 9.626865671641792e-05,
|
| 78 |
+
"loss": 1.2754,
|
| 79 |
"step": 100
|
| 80 |
},
|
| 81 |
{
|
| 82 |
+
"epoch": 0.37,
|
| 83 |
+
"eval_accuracy": 0.7328646748681898,
|
| 84 |
+
"eval_loss": 1.1163370609283447,
|
| 85 |
+
"eval_runtime": 12.5514,
|
| 86 |
+
"eval_samples_per_second": 45.333,
|
| 87 |
+
"eval_steps_per_second": 5.736,
|
| 88 |
"step": 100
|
| 89 |
},
|
| 90 |
{
|
| 91 |
+
"epoch": 0.41,
|
| 92 |
+
"grad_norm": 2.642601728439331,
|
| 93 |
+
"learning_rate": 9.58955223880597e-05,
|
| 94 |
+
"loss": 1.2354,
|
| 95 |
+
"step": 110
|
| 96 |
+
},
|
| 97 |
+
{
|
| 98 |
+
"epoch": 0.45,
|
| 99 |
+
"grad_norm": 2.4862725734710693,
|
| 100 |
+
"learning_rate": 9.552238805970149e-05,
|
| 101 |
+
"loss": 1.169,
|
| 102 |
"step": 120
|
| 103 |
},
|
| 104 |
{
|
| 105 |
+
"epoch": 0.49,
|
| 106 |
+
"grad_norm": 3.962764263153076,
|
| 107 |
+
"learning_rate": 9.514925373134329e-05,
|
| 108 |
+
"loss": 1.2546,
|
| 109 |
+
"step": 130
|
| 110 |
+
},
|
| 111 |
+
{
|
| 112 |
+
"epoch": 0.52,
|
| 113 |
+
"grad_norm": 2.9388816356658936,
|
| 114 |
+
"learning_rate": 9.477611940298507e-05,
|
| 115 |
+
"loss": 1.1702,
|
| 116 |
"step": 140
|
| 117 |
},
|
| 118 |
{
|
| 119 |
+
"epoch": 0.56,
|
| 120 |
+
"grad_norm": 4.958592414855957,
|
| 121 |
+
"learning_rate": 9.440298507462687e-05,
|
| 122 |
+
"loss": 1.0865,
|
| 123 |
+
"step": 150
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"epoch": 0.6,
|
| 127 |
+
"grad_norm": 3.4470815658569336,
|
| 128 |
+
"learning_rate": 9.402985074626867e-05,
|
| 129 |
+
"loss": 1.0097,
|
| 130 |
"step": 160
|
| 131 |
},
|
| 132 |
{
|
| 133 |
+
"epoch": 0.63,
|
| 134 |
+
"grad_norm": 4.423004627227783,
|
| 135 |
+
"learning_rate": 9.365671641791045e-05,
|
| 136 |
+
"loss": 1.0749,
|
| 137 |
+
"step": 170
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 0.67,
|
| 141 |
+
"grad_norm": 2.808164358139038,
|
| 142 |
+
"learning_rate": 9.328358208955224e-05,
|
| 143 |
+
"loss": 0.9732,
|
| 144 |
"step": 180
|
| 145 |
},
|
| 146 |
{
|
| 147 |
+
"epoch": 0.71,
|
| 148 |
+
"grad_norm": 6.00456428527832,
|
| 149 |
+
"learning_rate": 9.291044776119402e-05,
|
| 150 |
+
"loss": 1.0009,
|
| 151 |
+
"step": 190
|
| 152 |
+
},
|
| 153 |
+
{
|
| 154 |
+
"epoch": 0.75,
|
| 155 |
+
"grad_norm": 5.091552734375,
|
| 156 |
+
"learning_rate": 9.253731343283582e-05,
|
| 157 |
+
"loss": 0.9345,
|
| 158 |
"step": 200
|
| 159 |
},
|
| 160 |
{
|
| 161 |
+
"epoch": 0.75,
|
| 162 |
+
"eval_accuracy": 0.7996485061511424,
|
| 163 |
+
"eval_loss": 0.8296495079994202,
|
| 164 |
+
"eval_runtime": 7.8912,
|
| 165 |
+
"eval_samples_per_second": 72.105,
|
| 166 |
+
"eval_steps_per_second": 9.124,
|
| 167 |
"step": 200
|
| 168 |
},
|
| 169 |
{
|
| 170 |
+
"epoch": 0.78,
|
| 171 |
+
"grad_norm": 3.2533326148986816,
|
| 172 |
+
"learning_rate": 9.216417910447762e-05,
|
| 173 |
+
"loss": 0.793,
|
| 174 |
+
"step": 210
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"epoch": 0.82,
|
| 178 |
+
"grad_norm": 6.073918342590332,
|
| 179 |
+
"learning_rate": 9.17910447761194e-05,
|
| 180 |
+
"loss": 0.9835,
|
| 181 |
"step": 220
|
| 182 |
},
|
| 183 |
{
|
| 184 |
+
"epoch": 0.86,
|
| 185 |
+
"grad_norm": 3.6311192512512207,
|
| 186 |
+
"learning_rate": 9.14179104477612e-05,
|
| 187 |
+
"loss": 0.8801,
|
| 188 |
+
"step": 230
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 0.9,
|
| 192 |
+
"grad_norm": 4.446895599365234,
|
| 193 |
+
"learning_rate": 9.104477611940299e-05,
|
| 194 |
+
"loss": 1.0534,
|
| 195 |
"step": 240
|
| 196 |
},
|
| 197 |
{
|
| 198 |
+
"epoch": 0.93,
|
| 199 |
+
"grad_norm": 4.668705463409424,
|
| 200 |
+
"learning_rate": 9.067164179104479e-05,
|
| 201 |
+
"loss": 0.9396,
|
| 202 |
+
"step": 250
|
| 203 |
+
},
|
| 204 |
+
{
|
| 205 |
+
"epoch": 0.97,
|
| 206 |
+
"grad_norm": 6.191302299499512,
|
| 207 |
+
"learning_rate": 9.029850746268657e-05,
|
| 208 |
+
"loss": 0.9275,
|
| 209 |
"step": 260
|
| 210 |
},
|
| 211 |
{
|
| 212 |
+
"epoch": 1.01,
|
| 213 |
+
"grad_norm": 3.170959711074829,
|
| 214 |
+
"learning_rate": 8.992537313432836e-05,
|
| 215 |
+
"loss": 0.8595,
|
| 216 |
+
"step": 270
|
| 217 |
+
},
|
| 218 |
+
{
|
| 219 |
+
"epoch": 1.04,
|
| 220 |
+
"grad_norm": 3.690964460372925,
|
| 221 |
+
"learning_rate": 8.955223880597016e-05,
|
| 222 |
+
"loss": 0.733,
|
| 223 |
"step": 280
|
| 224 |
},
|
| 225 |
{
|
| 226 |
+
"epoch": 1.08,
|
| 227 |
+
"grad_norm": 4.871851444244385,
|
| 228 |
+
"learning_rate": 8.917910447761194e-05,
|
| 229 |
+
"loss": 0.7623,
|
| 230 |
+
"step": 290
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 1.12,
|
| 234 |
+
"grad_norm": 3.3851799964904785,
|
| 235 |
+
"learning_rate": 8.880597014925374e-05,
|
| 236 |
+
"loss": 0.8816,
|
| 237 |
"step": 300
|
| 238 |
},
|
| 239 |
{
|
| 240 |
+
"epoch": 1.12,
|
| 241 |
+
"eval_accuracy": 0.8101933216168717,
|
| 242 |
+
"eval_loss": 0.7156229615211487,
|
| 243 |
+
"eval_runtime": 7.8519,
|
| 244 |
+
"eval_samples_per_second": 72.467,
|
| 245 |
+
"eval_steps_per_second": 9.17,
|
| 246 |
"step": 300
|
| 247 |
},
|
| 248 |
{
|
| 249 |
+
"epoch": 1.16,
|
| 250 |
+
"grad_norm": 3.334380865097046,
|
| 251 |
+
"learning_rate": 8.843283582089554e-05,
|
| 252 |
+
"loss": 0.8567,
|
| 253 |
+
"step": 310
|
| 254 |
+
},
|
| 255 |
+
{
|
| 256 |
+
"epoch": 1.19,
|
| 257 |
+
"grad_norm": 4.673859596252441,
|
| 258 |
+
"learning_rate": 8.805970149253732e-05,
|
| 259 |
+
"loss": 0.7926,
|
| 260 |
"step": 320
|
| 261 |
},
|
| 262 |
{
|
| 263 |
+
"epoch": 1.23,
|
| 264 |
+
"grad_norm": 3.3042550086975098,
|
| 265 |
+
"learning_rate": 8.76865671641791e-05,
|
| 266 |
+
"loss": 0.6847,
|
| 267 |
+
"step": 330
|
| 268 |
+
},
|
| 269 |
+
{
|
| 270 |
+
"epoch": 1.27,
|
| 271 |
+
"grad_norm": 5.4356513023376465,
|
| 272 |
+
"learning_rate": 8.731343283582089e-05,
|
| 273 |
+
"loss": 0.7656,
|
| 274 |
"step": 340
|
| 275 |
},
|
| 276 |
{
|
| 277 |
+
"epoch": 1.31,
|
| 278 |
+
"grad_norm": 7.050413131713867,
|
| 279 |
+
"learning_rate": 8.694029850746269e-05,
|
| 280 |
+
"loss": 0.6658,
|
| 281 |
+
"step": 350
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 1.34,
|
| 285 |
+
"grad_norm": 5.980592727661133,
|
| 286 |
+
"learning_rate": 8.656716417910447e-05,
|
| 287 |
+
"loss": 0.7948,
|
| 288 |
"step": 360
|
| 289 |
},
|
| 290 |
{
|
| 291 |
+
"epoch": 1.38,
|
| 292 |
+
"grad_norm": 3.894716739654541,
|
| 293 |
+
"learning_rate": 8.619402985074627e-05,
|
| 294 |
+
"loss": 0.8381,
|
| 295 |
+
"step": 370
|
| 296 |
+
},
|
| 297 |
+
{
|
| 298 |
+
"epoch": 1.42,
|
| 299 |
+
"grad_norm": 7.189664363861084,
|
| 300 |
+
"learning_rate": 8.582089552238807e-05,
|
| 301 |
+
"loss": 0.6532,
|
| 302 |
"step": 380
|
| 303 |
},
|
| 304 |
{
|
| 305 |
+
"epoch": 1.46,
|
| 306 |
+
"grad_norm": 4.317276477813721,
|
| 307 |
+
"learning_rate": 8.544776119402986e-05,
|
| 308 |
+
"loss": 0.7763,
|
| 309 |
+
"step": 390
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 1.49,
|
| 313 |
+
"grad_norm": 4.480589866638184,
|
| 314 |
+
"learning_rate": 8.511194029850747e-05,
|
| 315 |
+
"loss": 0.7425,
|
| 316 |
"step": 400
|
| 317 |
},
|
| 318 |
{
|
| 319 |
+
"epoch": 1.49,
|
| 320 |
+
"eval_accuracy": 0.8066783831282952,
|
| 321 |
+
"eval_loss": 0.6529447436332703,
|
| 322 |
+
"eval_runtime": 7.793,
|
| 323 |
+
"eval_samples_per_second": 73.014,
|
| 324 |
+
"eval_steps_per_second": 9.239,
|
| 325 |
"step": 400
|
| 326 |
},
|
| 327 |
{
|
| 328 |
+
"epoch": 1.53,
|
| 329 |
+
"grad_norm": 4.1799163818359375,
|
| 330 |
+
"learning_rate": 8.473880597014926e-05,
|
| 331 |
+
"loss": 0.6928,
|
| 332 |
+
"step": 410
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 1.57,
|
| 336 |
+
"grad_norm": 4.81996488571167,
|
| 337 |
+
"learning_rate": 8.436567164179105e-05,
|
| 338 |
+
"loss": 0.7769,
|
| 339 |
"step": 420
|
| 340 |
},
|
| 341 |
{
|
| 342 |
+
"epoch": 1.6,
|
| 343 |
+
"grad_norm": 7.18645715713501,
|
| 344 |
+
"learning_rate": 8.399253731343283e-05,
|
| 345 |
+
"loss": 0.6848,
|
| 346 |
+
"step": 430
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 1.64,
|
| 350 |
+
"grad_norm": 3.888197660446167,
|
| 351 |
+
"learning_rate": 8.361940298507463e-05,
|
| 352 |
+
"loss": 0.5977,
|
| 353 |
"step": 440
|
| 354 |
},
|
| 355 |
{
|
| 356 |
+
"epoch": 1.68,
|
| 357 |
+
"grad_norm": 7.374312877655029,
|
| 358 |
+
"learning_rate": 8.324626865671642e-05,
|
| 359 |
+
"loss": 0.6001,
|
| 360 |
+
"step": 450
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 1.72,
|
| 364 |
+
"grad_norm": 6.553064823150635,
|
| 365 |
+
"learning_rate": 8.287313432835821e-05,
|
| 366 |
+
"loss": 0.6683,
|
| 367 |
"step": 460
|
| 368 |
},
|
| 369 |
{
|
| 370 |
+
"epoch": 1.75,
|
| 371 |
+
"grad_norm": 3.466761589050293,
|
| 372 |
+
"learning_rate": 8.25e-05,
|
| 373 |
+
"loss": 0.6484,
|
| 374 |
+
"step": 470
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 1.79,
|
| 378 |
+
"grad_norm": 3.534076690673828,
|
| 379 |
+
"learning_rate": 8.21268656716418e-05,
|
| 380 |
+
"loss": 0.6589,
|
| 381 |
"step": 480
|
| 382 |
},
|
| 383 |
{
|
| 384 |
+
"epoch": 1.83,
|
| 385 |
+
"grad_norm": 3.581280469894409,
|
| 386 |
+
"learning_rate": 8.17537313432836e-05,
|
| 387 |
+
"loss": 0.6173,
|
| 388 |
+
"step": 490
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 1.87,
|
| 392 |
+
"grad_norm": 6.162041664123535,
|
| 393 |
+
"learning_rate": 8.138059701492538e-05,
|
| 394 |
+
"loss": 0.6883,
|
| 395 |
"step": 500
|
| 396 |
},
|
| 397 |
{
|
| 398 |
+
"epoch": 1.87,
|
| 399 |
+
"eval_accuracy": 0.8242530755711776,
|
| 400 |
+
"eval_loss": 0.6078779697418213,
|
| 401 |
+
"eval_runtime": 7.6716,
|
| 402 |
+
"eval_samples_per_second": 74.169,
|
| 403 |
+
"eval_steps_per_second": 9.385,
|
| 404 |
"step": 500
|
| 405 |
},
|
| 406 |
{
|
| 407 |
+
"epoch": 1.9,
|
| 408 |
+
"grad_norm": 5.477086067199707,
|
| 409 |
+
"learning_rate": 8.100746268656717e-05,
|
| 410 |
+
"loss": 0.5952,
|
| 411 |
+
"step": 510
|
| 412 |
+
},
|
| 413 |
+
{
|
| 414 |
+
"epoch": 1.94,
|
| 415 |
+
"grad_norm": 2.389667510986328,
|
| 416 |
+
"learning_rate": 8.063432835820895e-05,
|
| 417 |
+
"loss": 0.5193,
|
| 418 |
"step": 520
|
| 419 |
},
|
| 420 |
{
|
| 421 |
+
"epoch": 1.98,
|
| 422 |
+
"grad_norm": 5.730781555175781,
|
| 423 |
+
"learning_rate": 8.026119402985075e-05,
|
| 424 |
+
"loss": 0.6818,
|
| 425 |
+
"step": 530
|
| 426 |
+
},
|
| 427 |
+
{
|
| 428 |
+
"epoch": 2.01,
|
| 429 |
+
"grad_norm": 6.305990219116211,
|
| 430 |
+
"learning_rate": 7.992537313432836e-05,
|
| 431 |
+
"loss": 0.5738,
|
| 432 |
"step": 540
|
| 433 |
},
|
| 434 |
{
|
| 435 |
+
"epoch": 2.05,
|
| 436 |
+
"grad_norm": 3.507434368133545,
|
| 437 |
+
"learning_rate": 7.955223880597016e-05,
|
| 438 |
+
"loss": 0.5685,
|
| 439 |
+
"step": 550
|
| 440 |
+
},
|
| 441 |
+
{
|
| 442 |
+
"epoch": 2.09,
|
| 443 |
+
"grad_norm": 12.683993339538574,
|
| 444 |
+
"learning_rate": 7.917910447761194e-05,
|
| 445 |
+
"loss": 0.6684,
|
| 446 |
"step": 560
|
| 447 |
},
|
| 448 |
{
|
| 449 |
+
"epoch": 2.13,
|
| 450 |
+
"grad_norm": 5.5166916847229,
|
| 451 |
+
"learning_rate": 7.880597014925374e-05,
|
| 452 |
+
"loss": 0.4787,
|
| 453 |
+
"step": 570
|
| 454 |
+
},
|
| 455 |
+
{
|
| 456 |
+
"epoch": 2.16,
|
| 457 |
+
"grad_norm": 6.427499294281006,
|
| 458 |
+
"learning_rate": 7.843283582089552e-05,
|
| 459 |
+
"loss": 0.5818,
|
| 460 |
"step": 580
|
| 461 |
},
|
| 462 |
{
|
| 463 |
+
"epoch": 2.2,
|
| 464 |
+
"grad_norm": 5.062973976135254,
|
| 465 |
+
"learning_rate": 7.805970149253732e-05,
|
| 466 |
+
"loss": 0.4766,
|
| 467 |
+
"step": 590
|
| 468 |
+
},
|
| 469 |
+
{
|
| 470 |
+
"epoch": 2.24,
|
| 471 |
+
"grad_norm": 5.720675945281982,
|
| 472 |
+
"learning_rate": 7.768656716417911e-05,
|
| 473 |
+
"loss": 0.5454,
|
| 474 |
"step": 600
|
| 475 |
},
|
| 476 |
{
|
| 477 |
+
"epoch": 2.24,
|
| 478 |
+
"eval_accuracy": 0.8347978910369068,
|
| 479 |
+
"eval_loss": 0.5604887008666992,
|
| 480 |
+
"eval_runtime": 7.7133,
|
| 481 |
+
"eval_samples_per_second": 73.769,
|
| 482 |
+
"eval_steps_per_second": 9.335,
|
| 483 |
"step": 600
|
| 484 |
},
|
| 485 |
{
|
| 486 |
+
"epoch": 2.28,
|
| 487 |
+
"grad_norm": 7.875051021575928,
|
| 488 |
+
"learning_rate": 7.731343283582089e-05,
|
| 489 |
+
"loss": 0.5935,
|
| 490 |
+
"step": 610
|
| 491 |
+
},
|
| 492 |
+
{
|
| 493 |
+
"epoch": 2.31,
|
| 494 |
+
"grad_norm": 4.378401756286621,
|
| 495 |
+
"learning_rate": 7.694029850746269e-05,
|
| 496 |
+
"loss": 0.4639,
|
| 497 |
"step": 620
|
| 498 |
},
|
| 499 |
{
|
| 500 |
+
"epoch": 2.35,
|
| 501 |
+
"grad_norm": 7.522930145263672,
|
| 502 |
+
"learning_rate": 7.656716417910448e-05,
|
| 503 |
+
"loss": 0.4867,
|
| 504 |
+
"step": 630
|
| 505 |
+
},
|
| 506 |
+
{
|
| 507 |
+
"epoch": 2.39,
|
| 508 |
+
"grad_norm": 6.3615288734436035,
|
| 509 |
+
"learning_rate": 7.619402985074627e-05,
|
| 510 |
+
"loss": 0.5302,
|
| 511 |
"step": 640
|
| 512 |
},
|
| 513 |
{
|
| 514 |
+
"epoch": 2.43,
|
| 515 |
+
"grad_norm": 3.8204784393310547,
|
| 516 |
+
"learning_rate": 7.582089552238806e-05,
|
| 517 |
+
"loss": 0.3864,
|
| 518 |
+
"step": 650
|
| 519 |
+
},
|
| 520 |
+
{
|
| 521 |
+
"epoch": 2.46,
|
| 522 |
+
"grad_norm": 2.3520662784576416,
|
| 523 |
+
"learning_rate": 7.544776119402986e-05,
|
| 524 |
+
"loss": 0.6458,
|
| 525 |
"step": 660
|
| 526 |
},
|
| 527 |
{
|
| 528 |
+
"epoch": 2.5,
|
| 529 |
+
"grad_norm": 3.9832942485809326,
|
| 530 |
+
"learning_rate": 7.507462686567166e-05,
|
| 531 |
+
"loss": 0.494,
|
| 532 |
+
"step": 670
|
| 533 |
+
},
|
| 534 |
+
{
|
| 535 |
+
"epoch": 2.54,
|
| 536 |
+
"grad_norm": 3.6783320903778076,
|
| 537 |
+
"learning_rate": 7.470149253731343e-05,
|
| 538 |
+
"loss": 0.6213,
|
| 539 |
"step": 680
|
| 540 |
},
|
| 541 |
{
|
| 542 |
+
"epoch": 2.57,
|
| 543 |
+
"grad_norm": 4.528789520263672,
|
| 544 |
+
"learning_rate": 7.432835820895523e-05,
|
| 545 |
+
"loss": 0.615,
|
| 546 |
+
"step": 690
|
| 547 |
+
},
|
| 548 |
+
{
|
| 549 |
+
"epoch": 2.61,
|
| 550 |
+
"grad_norm": 5.556227207183838,
|
| 551 |
+
"learning_rate": 7.395522388059701e-05,
|
| 552 |
+
"loss": 0.5383,
|
| 553 |
"step": 700
|
| 554 |
},
|
| 555 |
{
|
| 556 |
+
"epoch": 2.61,
|
| 557 |
+
"eval_accuracy": 0.8295254833040422,
|
| 558 |
+
"eval_loss": 0.5571200251579285,
|
| 559 |
+
"eval_runtime": 7.8934,
|
| 560 |
+
"eval_samples_per_second": 72.085,
|
| 561 |
+
"eval_steps_per_second": 9.122,
|
| 562 |
"step": 700
|
| 563 |
},
|
| 564 |
{
|
| 565 |
+
"epoch": 2.65,
|
| 566 |
+
"grad_norm": 4.617480754852295,
|
| 567 |
+
"learning_rate": 7.358208955223881e-05,
|
| 568 |
+
"loss": 0.4987,
|
| 569 |
+
"step": 710
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 2.69,
|
| 573 |
+
"grad_norm": 4.6940412521362305,
|
| 574 |
+
"learning_rate": 7.32089552238806e-05,
|
| 575 |
+
"loss": 0.5466,
|
| 576 |
"step": 720
|
| 577 |
},
|
| 578 |
{
|
| 579 |
+
"epoch": 2.72,
|
| 580 |
+
"grad_norm": 3.8839175701141357,
|
| 581 |
+
"learning_rate": 7.283582089552239e-05,
|
| 582 |
+
"loss": 0.5409,
|
| 583 |
+
"step": 730
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 2.76,
|
| 587 |
+
"grad_norm": 6.855696201324463,
|
| 588 |
+
"learning_rate": 7.246268656716419e-05,
|
| 589 |
+
"loss": 0.3972,
|
| 590 |
"step": 740
|
| 591 |
},
|
| 592 |
{
|
| 593 |
+
"epoch": 2.8,
|
| 594 |
+
"grad_norm": 3.9779269695281982,
|
| 595 |
+
"learning_rate": 7.208955223880597e-05,
|
| 596 |
+
"loss": 0.4719,
|
| 597 |
+
"step": 750
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 2.84,
|
| 601 |
+
"grad_norm": 10.327420234680176,
|
| 602 |
+
"learning_rate": 7.171641791044776e-05,
|
| 603 |
+
"loss": 0.668,
|
| 604 |
"step": 760
|
| 605 |
},
|
| 606 |
{
|
| 607 |
+
"epoch": 2.87,
|
| 608 |
+
"grad_norm": 5.06951904296875,
|
| 609 |
+
"learning_rate": 7.134328358208956e-05,
|
| 610 |
+
"loss": 0.5899,
|
| 611 |
+
"step": 770
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 2.91,
|
| 615 |
+
"grad_norm": 5.539373397827148,
|
| 616 |
+
"learning_rate": 7.097014925373134e-05,
|
| 617 |
+
"loss": 0.5813,
|
| 618 |
"step": 780
|
| 619 |
},
|
| 620 |
{
|
| 621 |
+
"epoch": 2.95,
|
| 622 |
+
"grad_norm": 4.622121334075928,
|
| 623 |
+
"learning_rate": 7.059701492537314e-05,
|
| 624 |
+
"loss": 0.5294,
|
| 625 |
+
"step": 790
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 2.99,
|
| 629 |
+
"grad_norm": 2.6457552909851074,
|
| 630 |
+
"learning_rate": 7.022388059701493e-05,
|
| 631 |
+
"loss": 0.5442,
|
| 632 |
"step": 800
|
| 633 |
},
|
| 634 |
{
|
| 635 |
+
"epoch": 2.99,
|
| 636 |
+
"eval_accuracy": 0.8189806678383128,
|
| 637 |
+
"eval_loss": 0.5864126682281494,
|
| 638 |
+
"eval_runtime": 7.8507,
|
| 639 |
+
"eval_samples_per_second": 72.478,
|
| 640 |
+
"eval_steps_per_second": 9.171,
|
| 641 |
"step": 800
|
| 642 |
},
|
| 643 |
{
|
| 644 |
+
"epoch": 3.02,
|
| 645 |
+
"grad_norm": 3.373798370361328,
|
| 646 |
+
"learning_rate": 6.985074626865672e-05,
|
| 647 |
+
"loss": 0.4183,
|
| 648 |
+
"step": 810
|
| 649 |
+
},
|
| 650 |
+
{
|
| 651 |
+
"epoch": 3.06,
|
| 652 |
+
"grad_norm": 4.0179667472839355,
|
| 653 |
+
"learning_rate": 6.947761194029851e-05,
|
| 654 |
+
"loss": 0.3611,
|
| 655 |
"step": 820
|
| 656 |
},
|
| 657 |
{
|
| 658 |
+
"epoch": 3.1,
|
| 659 |
+
"grad_norm": 7.72437858581543,
|
| 660 |
+
"learning_rate": 6.91044776119403e-05,
|
| 661 |
+
"loss": 0.4543,
|
| 662 |
+
"step": 830
|
| 663 |
+
},
|
| 664 |
+
{
|
| 665 |
+
"epoch": 3.13,
|
| 666 |
+
"grad_norm": 3.1097893714904785,
|
| 667 |
+
"learning_rate": 6.873134328358209e-05,
|
| 668 |
+
"loss": 0.5194,
|
| 669 |
"step": 840
|
| 670 |
},
|
| 671 |
{
|
| 672 |
+
"epoch": 3.17,
|
| 673 |
+
"grad_norm": 6.581250190734863,
|
| 674 |
+
"learning_rate": 6.835820895522388e-05,
|
| 675 |
+
"loss": 0.3839,
|
| 676 |
+
"step": 850
|
| 677 |
+
},
|
| 678 |
+
{
|
| 679 |
+
"epoch": 3.21,
|
| 680 |
+
"grad_norm": 5.605171203613281,
|
| 681 |
+
"learning_rate": 6.798507462686568e-05,
|
| 682 |
+
"loss": 0.4499,
|
| 683 |
"step": 860
|
| 684 |
},
|
| 685 |
{
|
| 686 |
+
"epoch": 3.25,
|
| 687 |
+
"grad_norm": 2.834651231765747,
|
| 688 |
+
"learning_rate": 6.761194029850747e-05,
|
| 689 |
+
"loss": 0.5067,
|
| 690 |
+
"step": 870
|
| 691 |
+
},
|
| 692 |
+
{
|
| 693 |
+
"epoch": 3.28,
|
| 694 |
+
"grad_norm": 4.615099906921387,
|
| 695 |
+
"learning_rate": 6.723880597014926e-05,
|
| 696 |
+
"loss": 0.4869,
|
| 697 |
"step": 880
|
| 698 |
},
|
| 699 |
{
|
| 700 |
+
"epoch": 3.32,
|
| 701 |
+
"grad_norm": 6.115981101989746,
|
| 702 |
+
"learning_rate": 6.686567164179106e-05,
|
| 703 |
+
"loss": 0.4793,
|
| 704 |
+
"step": 890
|
| 705 |
+
},
|
| 706 |
+
{
|
| 707 |
+
"epoch": 3.36,
|
| 708 |
+
"grad_norm": 1.1021697521209717,
|
| 709 |
+
"learning_rate": 6.649253731343283e-05,
|
| 710 |
+
"loss": 0.3986,
|
| 711 |
"step": 900
|
| 712 |
},
|
| 713 |
{
|
| 714 |
+
"epoch": 3.36,
|
| 715 |
+
"eval_accuracy": 0.8312829525483304,
|
| 716 |
+
"eval_loss": 0.5632173418998718,
|
| 717 |
+
"eval_runtime": 7.731,
|
| 718 |
+
"eval_samples_per_second": 73.6,
|
| 719 |
+
"eval_steps_per_second": 9.313,
|
| 720 |
"step": 900
|
| 721 |
},
|
| 722 |
{
|
| 723 |
+
"epoch": 3.4,
|
| 724 |
+
"grad_norm": 7.019008159637451,
|
| 725 |
+
"learning_rate": 6.611940298507463e-05,
|
| 726 |
+
"loss": 0.383,
|
| 727 |
+
"step": 910
|
| 728 |
+
},
|
| 729 |
+
{
|
| 730 |
+
"epoch": 3.43,
|
| 731 |
+
"grad_norm": 2.586031913757324,
|
| 732 |
+
"learning_rate": 6.574626865671642e-05,
|
| 733 |
+
"loss": 0.2752,
|
| 734 |
"step": 920
|
| 735 |
},
|
| 736 |
{
|
| 737 |
+
"epoch": 3.47,
|
| 738 |
+
"grad_norm": 2.5189669132232666,
|
| 739 |
+
"learning_rate": 6.537313432835821e-05,
|
| 740 |
+
"loss": 0.2944,
|
| 741 |
+
"step": 930
|
| 742 |
+
},
|
| 743 |
+
{
|
| 744 |
+
"epoch": 3.51,
|
| 745 |
+
"grad_norm": 10.028382301330566,
|
| 746 |
+
"learning_rate": 6.500000000000001e-05,
|
| 747 |
+
"loss": 0.4378,
|
| 748 |
"step": 940
|
| 749 |
},
|
| 750 |
{
|
| 751 |
+
"epoch": 3.54,
|
| 752 |
+
"grad_norm": 1.8697803020477295,
|
| 753 |
+
"learning_rate": 6.462686567164179e-05,
|
| 754 |
+
"loss": 0.3956,
|
| 755 |
+
"step": 950
|
| 756 |
+
},
|
| 757 |
+
{
|
| 758 |
+
"epoch": 3.58,
|
| 759 |
+
"grad_norm": 5.872415065765381,
|
| 760 |
+
"learning_rate": 6.425373134328359e-05,
|
| 761 |
+
"loss": 0.338,
|
| 762 |
"step": 960
|
| 763 |
},
|
| 764 |
{
|
| 765 |
+
"epoch": 3.62,
|
| 766 |
+
"grad_norm": 8.272451400756836,
|
| 767 |
+
"learning_rate": 6.388059701492538e-05,
|
| 768 |
+
"loss": 0.4264,
|
| 769 |
+
"step": 970
|
| 770 |
+
},
|
| 771 |
+
{
|
| 772 |
+
"epoch": 3.66,
|
| 773 |
+
"grad_norm": 9.422249794006348,
|
| 774 |
+
"learning_rate": 6.350746268656716e-05,
|
| 775 |
+
"loss": 0.4258,
|
| 776 |
"step": 980
|
| 777 |
},
|
| 778 |
{
|
| 779 |
+
"epoch": 3.69,
|
| 780 |
+
"grad_norm": 8.768738746643066,
|
| 781 |
+
"learning_rate": 6.313432835820896e-05,
|
| 782 |
+
"loss": 0.3308,
|
| 783 |
+
"step": 990
|
| 784 |
+
},
|
| 785 |
+
{
|
| 786 |
+
"epoch": 3.73,
|
| 787 |
+
"grad_norm": 6.355968475341797,
|
| 788 |
+
"learning_rate": 6.276119402985074e-05,
|
| 789 |
+
"loss": 0.3438,
|
| 790 |
"step": 1000
|
| 791 |
},
|
| 792 |
{
|
| 793 |
+
"epoch": 3.73,
|
| 794 |
+
"eval_accuracy": 0.836555360281195,
|
| 795 |
+
"eval_loss": 0.5606371760368347,
|
| 796 |
+
"eval_runtime": 7.818,
|
| 797 |
+
"eval_samples_per_second": 72.781,
|
| 798 |
+
"eval_steps_per_second": 9.21,
|
| 799 |
"step": 1000
|
| 800 |
},
|
| 801 |
{
|
| 802 |
+
"epoch": 3.77,
|
| 803 |
+
"grad_norm": 3.973480463027954,
|
| 804 |
+
"learning_rate": 6.238805970149254e-05,
|
| 805 |
+
"loss": 0.5042,
|
| 806 |
+
"step": 1010
|
| 807 |
+
},
|
| 808 |
+
{
|
| 809 |
+
"epoch": 3.81,
|
| 810 |
+
"grad_norm": 5.739313125610352,
|
| 811 |
+
"learning_rate": 6.201492537313434e-05,
|
| 812 |
+
"loss": 0.4515,
|
| 813 |
"step": 1020
|
| 814 |
},
|
| 815 |
{
|
| 816 |
+
"epoch": 3.84,
|
| 817 |
+
"grad_norm": 4.196649074554443,
|
| 818 |
+
"learning_rate": 6.164179104477613e-05,
|
| 819 |
+
"loss": 0.4404,
|
| 820 |
+
"step": 1030
|
| 821 |
+
},
|
| 822 |
+
{
|
| 823 |
+
"epoch": 3.88,
|
| 824 |
+
"grad_norm": 4.671971321105957,
|
| 825 |
+
"learning_rate": 6.126865671641791e-05,
|
| 826 |
+
"loss": 0.4746,
|
| 827 |
"step": 1040
|
| 828 |
},
|
| 829 |
{
|
| 830 |
+
"epoch": 3.92,
|
| 831 |
+
"grad_norm": 6.87581205368042,
|
| 832 |
+
"learning_rate": 6.08955223880597e-05,
|
| 833 |
+
"loss": 0.4637,
|
| 834 |
+
"step": 1050
|
| 835 |
+
},
|
| 836 |
+
{
|
| 837 |
+
"epoch": 3.96,
|
| 838 |
+
"grad_norm": 7.224815368652344,
|
| 839 |
+
"learning_rate": 6.052238805970149e-05,
|
| 840 |
+
"loss": 0.4754,
|
| 841 |
"step": 1060
|
| 842 |
},
|
| 843 |
{
|
| 844 |
+
"epoch": 3.99,
|
| 845 |
+
"grad_norm": 4.4340314865112305,
|
| 846 |
+
"learning_rate": 6.014925373134329e-05,
|
| 847 |
+
"loss": 0.4165,
|
| 848 |
+
"step": 1070
|
| 849 |
+
},
|
| 850 |
+
{
|
| 851 |
+
"epoch": 4.03,
|
| 852 |
+
"grad_norm": 1.151932716369629,
|
| 853 |
+
"learning_rate": 5.9776119402985076e-05,
|
| 854 |
+
"loss": 0.3498,
|
| 855 |
"step": 1080
|
| 856 |
},
|
| 857 |
{
|
| 858 |
+
"epoch": 4.07,
|
| 859 |
+
"grad_norm": 6.31879997253418,
|
| 860 |
+
"learning_rate": 5.940298507462687e-05,
|
| 861 |
+
"loss": 0.3505,
|
| 862 |
+
"step": 1090
|
| 863 |
+
},
|
| 864 |
+
{
|
| 865 |
+
"epoch": 4.1,
|
| 866 |
+
"grad_norm": 4.674696445465088,
|
| 867 |
+
"learning_rate": 5.902985074626865e-05,
|
| 868 |
+
"loss": 0.4345,
|
| 869 |
"step": 1100
|
| 870 |
},
|
| 871 |
{
|
| 872 |
+
"epoch": 4.1,
|
| 873 |
+
"eval_accuracy": 0.836555360281195,
|
| 874 |
+
"eval_loss": 0.5353797674179077,
|
| 875 |
+
"eval_runtime": 7.9559,
|
| 876 |
+
"eval_samples_per_second": 71.519,
|
| 877 |
+
"eval_steps_per_second": 9.05,
|
| 878 |
"step": 1100
|
| 879 |
},
|
| 880 |
{
|
| 881 |
+
"epoch": 4.14,
|
| 882 |
+
"grad_norm": 6.790203094482422,
|
| 883 |
+
"learning_rate": 5.865671641791045e-05,
|
| 884 |
+
"loss": 0.3189,
|
| 885 |
+
"step": 1110
|
| 886 |
+
},
|
| 887 |
+
{
|
| 888 |
+
"epoch": 4.18,
|
| 889 |
+
"grad_norm": 5.554905414581299,
|
| 890 |
+
"learning_rate": 5.828358208955225e-05,
|
| 891 |
+
"loss": 0.3255,
|
| 892 |
"step": 1120
|
| 893 |
},
|
| 894 |
{
|
| 895 |
+
"epoch": 4.22,
|
| 896 |
+
"grad_norm": 1.87189781665802,
|
| 897 |
+
"learning_rate": 5.7910447761194034e-05,
|
| 898 |
+
"loss": 0.2613,
|
| 899 |
+
"step": 1130
|
| 900 |
+
},
|
| 901 |
+
{
|
| 902 |
+
"epoch": 4.25,
|
| 903 |
+
"grad_norm": 3.4729249477386475,
|
| 904 |
+
"learning_rate": 5.7537313432835826e-05,
|
| 905 |
+
"loss": 0.4037,
|
| 906 |
"step": 1140
|
| 907 |
},
|
| 908 |
{
|
| 909 |
+
"epoch": 4.29,
|
| 910 |
+
"grad_norm": 3.2373063564300537,
|
| 911 |
+
"learning_rate": 5.716417910447761e-05,
|
| 912 |
+
"loss": 0.384,
|
| 913 |
+
"step": 1150
|
| 914 |
+
},
|
| 915 |
+
{
|
| 916 |
+
"epoch": 4.33,
|
| 917 |
+
"grad_norm": 1.8042526245117188,
|
| 918 |
+
"learning_rate": 5.679104477611941e-05,
|
| 919 |
+
"loss": 0.4024,
|
| 920 |
"step": 1160
|
| 921 |
},
|
| 922 |
{
|
| 923 |
+
"epoch": 4.37,
|
| 924 |
+
"grad_norm": 0.9592193365097046,
|
| 925 |
+
"learning_rate": 5.64179104477612e-05,
|
| 926 |
+
"loss": 0.3646,
|
| 927 |
+
"step": 1170
|
| 928 |
+
},
|
| 929 |
+
{
|
| 930 |
+
"epoch": 4.4,
|
| 931 |
+
"grad_norm": 4.0469584465026855,
|
| 932 |
+
"learning_rate": 5.6044776119402986e-05,
|
| 933 |
+
"loss": 0.3622,
|
| 934 |
"step": 1180
|
| 935 |
},
|
| 936 |
{
|
| 937 |
+
"epoch": 4.44,
|
| 938 |
+
"grad_norm": 4.470405578613281,
|
| 939 |
+
"learning_rate": 5.5671641791044784e-05,
|
| 940 |
+
"loss": 0.2996,
|
| 941 |
+
"step": 1190
|
| 942 |
+
},
|
| 943 |
+
{
|
| 944 |
+
"epoch": 4.48,
|
| 945 |
+
"grad_norm": 6.086768627166748,
|
| 946 |
+
"learning_rate": 5.529850746268657e-05,
|
| 947 |
+
"loss": 0.4523,
|
| 948 |
"step": 1200
|
| 949 |
},
|
| 950 |
{
|
| 951 |
+
"epoch": 4.48,
|
| 952 |
"eval_accuracy": 0.8576449912126538,
|
| 953 |
+
"eval_loss": 0.49876561760902405,
|
| 954 |
+
"eval_runtime": 7.8527,
|
| 955 |
+
"eval_samples_per_second": 72.459,
|
| 956 |
+
"eval_steps_per_second": 9.169,
|
| 957 |
"step": 1200
|
| 958 |
},
|
| 959 |
{
|
| 960 |
+
"epoch": 4.51,
|
| 961 |
+
"grad_norm": 3.478428363800049,
|
| 962 |
+
"learning_rate": 5.492537313432836e-05,
|
| 963 |
+
"loss": 0.4198,
|
| 964 |
+
"step": 1210
|
| 965 |
+
},
|
| 966 |
+
{
|
| 967 |
+
"epoch": 4.55,
|
| 968 |
+
"grad_norm": 4.539990425109863,
|
| 969 |
+
"learning_rate": 5.455223880597016e-05,
|
| 970 |
+
"loss": 0.3125,
|
| 971 |
"step": 1220
|
| 972 |
},
|
| 973 |
{
|
| 974 |
+
"epoch": 4.59,
|
| 975 |
+
"grad_norm": 3.971435070037842,
|
| 976 |
+
"learning_rate": 5.4179104477611943e-05,
|
| 977 |
+
"loss": 0.2773,
|
| 978 |
+
"step": 1230
|
| 979 |
+
},
|
| 980 |
+
{
|
| 981 |
+
"epoch": 4.63,
|
| 982 |
+
"grad_norm": 7.168191909790039,
|
| 983 |
+
"learning_rate": 5.3805970149253735e-05,
|
| 984 |
+
"loss": 0.4852,
|
| 985 |
"step": 1240
|
| 986 |
},
|
| 987 |
{
|
| 988 |
+
"epoch": 4.66,
|
| 989 |
+
"grad_norm": 2.896576166152954,
|
| 990 |
+
"learning_rate": 5.343283582089552e-05,
|
| 991 |
+
"loss": 0.3425,
|
| 992 |
+
"step": 1250
|
| 993 |
+
},
|
| 994 |
+
{
|
| 995 |
+
"epoch": 4.7,
|
| 996 |
+
"grad_norm": 1.4190607070922852,
|
| 997 |
+
"learning_rate": 5.305970149253732e-05,
|
| 998 |
+
"loss": 0.2219,
|
| 999 |
"step": 1260
|
| 1000 |
},
|
| 1001 |
{
|
| 1002 |
+
"epoch": 4.74,
|
| 1003 |
+
"grad_norm": 5.066045761108398,
|
| 1004 |
+
"learning_rate": 5.268656716417911e-05,
|
| 1005 |
+
"loss": 0.3447,
|
| 1006 |
+
"step": 1270
|
| 1007 |
+
},
|
| 1008 |
+
{
|
| 1009 |
+
"epoch": 4.78,
|
| 1010 |
+
"grad_norm": 4.2649126052856445,
|
| 1011 |
+
"learning_rate": 5.2313432835820895e-05,
|
| 1012 |
+
"loss": 0.3931,
|
| 1013 |
"step": 1280
|
| 1014 |
},
|
| 1015 |
{
|
| 1016 |
+
"epoch": 4.81,
|
| 1017 |
+
"grad_norm": 5.704684734344482,
|
| 1018 |
+
"learning_rate": 5.197761194029851e-05,
|
| 1019 |
+
"loss": 0.4274,
|
| 1020 |
+
"step": 1290
|
| 1021 |
+
},
|
| 1022 |
+
{
|
| 1023 |
+
"epoch": 4.85,
|
| 1024 |
+
"grad_norm": 6.395939350128174,
|
| 1025 |
+
"learning_rate": 5.16044776119403e-05,
|
| 1026 |
+
"loss": 0.3162,
|
| 1027 |
"step": 1300
|
| 1028 |
},
|
| 1029 |
{
|
| 1030 |
+
"epoch": 4.85,
|
| 1031 |
+
"eval_accuracy": 0.8541300527240774,
|
| 1032 |
+
"eval_loss": 0.5099390745162964,
|
| 1033 |
+
"eval_runtime": 7.9919,
|
| 1034 |
+
"eval_samples_per_second": 71.197,
|
| 1035 |
+
"eval_steps_per_second": 9.009,
|
| 1036 |
"step": 1300
|
| 1037 |
},
|
| 1038 |
{
|
| 1039 |
+
"epoch": 4.89,
|
| 1040 |
+
"grad_norm": 2.4717729091644287,
|
| 1041 |
+
"learning_rate": 5.123134328358209e-05,
|
| 1042 |
+
"loss": 0.3442,
|
| 1043 |
+
"step": 1310
|
| 1044 |
+
},
|
| 1045 |
+
{
|
| 1046 |
+
"epoch": 4.93,
|
| 1047 |
+
"grad_norm": 0.6504545211791992,
|
| 1048 |
+
"learning_rate": 5.0858208955223885e-05,
|
| 1049 |
+
"loss": 0.3313,
|
| 1050 |
"step": 1320
|
| 1051 |
},
|
| 1052 |
{
|
| 1053 |
+
"epoch": 4.96,
|
| 1054 |
+
"grad_norm": 4.316141128540039,
|
| 1055 |
+
"learning_rate": 5.048507462686567e-05,
|
| 1056 |
+
"loss": 0.3787,
|
| 1057 |
+
"step": 1330
|
| 1058 |
+
},
|
| 1059 |
+
{
|
| 1060 |
+
"epoch": 5.0,
|
| 1061 |
+
"grad_norm": 4.9243998527526855,
|
| 1062 |
+
"learning_rate": 5.011194029850746e-05,
|
| 1063 |
+
"loss": 0.38,
|
| 1064 |
"step": 1340
|
| 1065 |
},
|
| 1066 |
+
{
|
| 1067 |
+
"epoch": 5.04,
|
| 1068 |
+
"grad_norm": 5.312038421630859,
|
| 1069 |
+
"learning_rate": 4.973880597014925e-05,
|
| 1070 |
+
"loss": 0.3268,
|
| 1071 |
+
"step": 1350
|
| 1072 |
+
},
|
| 1073 |
+
{
|
| 1074 |
+
"epoch": 5.07,
|
| 1075 |
+
"grad_norm": 3.5483176708221436,
|
| 1076 |
+
"learning_rate": 4.9365671641791045e-05,
|
| 1077 |
+
"loss": 0.3423,
|
| 1078 |
+
"step": 1360
|
| 1079 |
+
},
|
| 1080 |
+
{
|
| 1081 |
+
"epoch": 5.11,
|
| 1082 |
+
"grad_norm": 4.414547920227051,
|
| 1083 |
+
"learning_rate": 4.899253731343284e-05,
|
| 1084 |
+
"loss": 0.2421,
|
| 1085 |
+
"step": 1370
|
| 1086 |
+
},
|
| 1087 |
+
{
|
| 1088 |
+
"epoch": 5.15,
|
| 1089 |
+
"grad_norm": 5.7323689460754395,
|
| 1090 |
+
"learning_rate": 4.861940298507463e-05,
|
| 1091 |
+
"loss": 0.2795,
|
| 1092 |
+
"step": 1380
|
| 1093 |
+
},
|
| 1094 |
+
{
|
| 1095 |
+
"epoch": 5.19,
|
| 1096 |
+
"grad_norm": 4.2763471603393555,
|
| 1097 |
+
"learning_rate": 4.824626865671642e-05,
|
| 1098 |
+
"loss": 0.2402,
|
| 1099 |
+
"step": 1390
|
| 1100 |
+
},
|
| 1101 |
+
{
|
| 1102 |
+
"epoch": 5.22,
|
| 1103 |
+
"grad_norm": 9.259199142456055,
|
| 1104 |
+
"learning_rate": 4.787313432835821e-05,
|
| 1105 |
+
"loss": 0.3793,
|
| 1106 |
+
"step": 1400
|
| 1107 |
+
},
|
| 1108 |
+
{
|
| 1109 |
+
"epoch": 5.22,
|
| 1110 |
+
"eval_accuracy": 0.843585237258348,
|
| 1111 |
+
"eval_loss": 0.5190387964248657,
|
| 1112 |
+
"eval_runtime": 7.7562,
|
| 1113 |
+
"eval_samples_per_second": 73.361,
|
| 1114 |
+
"eval_steps_per_second": 9.283,
|
| 1115 |
+
"step": 1400
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 5.26,
|
| 1119 |
+
"grad_norm": 4.773892402648926,
|
| 1120 |
+
"learning_rate": 4.75e-05,
|
| 1121 |
+
"loss": 0.3476,
|
| 1122 |
+
"step": 1410
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 5.3,
|
| 1126 |
+
"grad_norm": 1.1271159648895264,
|
| 1127 |
+
"learning_rate": 4.7126865671641794e-05,
|
| 1128 |
+
"loss": 0.1949,
|
| 1129 |
+
"step": 1420
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 5.34,
|
| 1133 |
+
"grad_norm": 2.823958158493042,
|
| 1134 |
+
"learning_rate": 4.6753731343283586e-05,
|
| 1135 |
+
"loss": 0.3009,
|
| 1136 |
+
"step": 1430
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 5.37,
|
| 1140 |
+
"grad_norm": 0.35977163910865784,
|
| 1141 |
+
"learning_rate": 4.638059701492538e-05,
|
| 1142 |
+
"loss": 0.1821,
|
| 1143 |
+
"step": 1440
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 5.41,
|
| 1147 |
+
"grad_norm": 3.380308151245117,
|
| 1148 |
+
"learning_rate": 4.600746268656716e-05,
|
| 1149 |
+
"loss": 0.323,
|
| 1150 |
+
"step": 1450
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 5.45,
|
| 1154 |
+
"grad_norm": 5.946179389953613,
|
| 1155 |
+
"learning_rate": 4.5634328358208954e-05,
|
| 1156 |
+
"loss": 0.5344,
|
| 1157 |
+
"step": 1460
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 5.49,
|
| 1161 |
+
"grad_norm": 8.254781723022461,
|
| 1162 |
+
"learning_rate": 4.526119402985075e-05,
|
| 1163 |
+
"loss": 0.2799,
|
| 1164 |
+
"step": 1470
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 5.52,
|
| 1168 |
+
"grad_norm": 6.808130741119385,
|
| 1169 |
+
"learning_rate": 4.4888059701492544e-05,
|
| 1170 |
+
"loss": 0.3173,
|
| 1171 |
+
"step": 1480
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 5.56,
|
| 1175 |
+
"grad_norm": 17.452037811279297,
|
| 1176 |
+
"learning_rate": 4.451492537313433e-05,
|
| 1177 |
+
"loss": 0.3251,
|
| 1178 |
+
"step": 1490
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 5.6,
|
| 1182 |
+
"grad_norm": 2.3097095489501953,
|
| 1183 |
+
"learning_rate": 4.414179104477612e-05,
|
| 1184 |
+
"loss": 0.3228,
|
| 1185 |
+
"step": 1500
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 5.6,
|
| 1189 |
+
"eval_accuracy": 0.8576449912126538,
|
| 1190 |
+
"eval_loss": 0.4589254856109619,
|
| 1191 |
+
"eval_runtime": 8.0547,
|
| 1192 |
+
"eval_samples_per_second": 70.642,
|
| 1193 |
+
"eval_steps_per_second": 8.939,
|
| 1194 |
+
"step": 1500
|
| 1195 |
+
},
|
| 1196 |
+
{
|
| 1197 |
+
"epoch": 5.63,
|
| 1198 |
+
"grad_norm": 3.337970018386841,
|
| 1199 |
+
"learning_rate": 4.376865671641791e-05,
|
| 1200 |
+
"loss": 0.2528,
|
| 1201 |
+
"step": 1510
|
| 1202 |
+
},
|
| 1203 |
+
{
|
| 1204 |
+
"epoch": 5.67,
|
| 1205 |
+
"grad_norm": 0.5921415090560913,
|
| 1206 |
+
"learning_rate": 4.33955223880597e-05,
|
| 1207 |
+
"loss": 0.2459,
|
| 1208 |
+
"step": 1520
|
| 1209 |
+
},
|
| 1210 |
+
{
|
| 1211 |
+
"epoch": 5.71,
|
| 1212 |
+
"grad_norm": 4.148998260498047,
|
| 1213 |
+
"learning_rate": 4.3022388059701495e-05,
|
| 1214 |
+
"loss": 0.2927,
|
| 1215 |
+
"step": 1530
|
| 1216 |
+
},
|
| 1217 |
+
{
|
| 1218 |
+
"epoch": 5.75,
|
| 1219 |
+
"grad_norm": 5.740537166595459,
|
| 1220 |
+
"learning_rate": 4.2649253731343286e-05,
|
| 1221 |
+
"loss": 0.423,
|
| 1222 |
+
"step": 1540
|
| 1223 |
+
},
|
| 1224 |
+
{
|
| 1225 |
+
"epoch": 5.78,
|
| 1226 |
+
"grad_norm": 5.316250324249268,
|
| 1227 |
+
"learning_rate": 4.227611940298508e-05,
|
| 1228 |
+
"loss": 0.3735,
|
| 1229 |
+
"step": 1550
|
| 1230 |
+
},
|
| 1231 |
+
{
|
| 1232 |
+
"epoch": 5.82,
|
| 1233 |
+
"grad_norm": 5.52378511428833,
|
| 1234 |
+
"learning_rate": 4.190298507462686e-05,
|
| 1235 |
+
"loss": 0.3613,
|
| 1236 |
+
"step": 1560
|
| 1237 |
+
},
|
| 1238 |
+
{
|
| 1239 |
+
"epoch": 5.86,
|
| 1240 |
+
"grad_norm": 2.1002511978149414,
|
| 1241 |
+
"learning_rate": 4.152985074626866e-05,
|
| 1242 |
+
"loss": 0.259,
|
| 1243 |
+
"step": 1570
|
| 1244 |
+
},
|
| 1245 |
+
{
|
| 1246 |
+
"epoch": 5.9,
|
| 1247 |
+
"grad_norm": 5.339119911193848,
|
| 1248 |
+
"learning_rate": 4.115671641791045e-05,
|
| 1249 |
+
"loss": 0.3355,
|
| 1250 |
+
"step": 1580
|
| 1251 |
+
},
|
| 1252 |
+
{
|
| 1253 |
+
"epoch": 5.93,
|
| 1254 |
+
"grad_norm": 3.0551536083221436,
|
| 1255 |
+
"learning_rate": 4.0783582089552244e-05,
|
| 1256 |
+
"loss": 0.4342,
|
| 1257 |
+
"step": 1590
|
| 1258 |
+
},
|
| 1259 |
+
{
|
| 1260 |
+
"epoch": 5.97,
|
| 1261 |
+
"grad_norm": 6.549235820770264,
|
| 1262 |
+
"learning_rate": 4.041044776119403e-05,
|
| 1263 |
+
"loss": 0.1795,
|
| 1264 |
+
"step": 1600
|
| 1265 |
+
},
|
| 1266 |
+
{
|
| 1267 |
+
"epoch": 5.97,
|
| 1268 |
+
"eval_accuracy": 0.8488576449912126,
|
| 1269 |
+
"eval_loss": 0.5095508694648743,
|
| 1270 |
+
"eval_runtime": 7.7872,
|
| 1271 |
+
"eval_samples_per_second": 73.068,
|
| 1272 |
+
"eval_steps_per_second": 9.246,
|
| 1273 |
+
"step": 1600
|
| 1274 |
+
},
|
| 1275 |
+
{
|
| 1276 |
+
"epoch": 6.01,
|
| 1277 |
+
"grad_norm": 11.5170316696167,
|
| 1278 |
+
"learning_rate": 4.003731343283582e-05,
|
| 1279 |
+
"loss": 0.3778,
|
| 1280 |
+
"step": 1610
|
| 1281 |
+
},
|
| 1282 |
+
{
|
| 1283 |
+
"epoch": 6.04,
|
| 1284 |
+
"grad_norm": 6.004143238067627,
|
| 1285 |
+
"learning_rate": 3.966417910447761e-05,
|
| 1286 |
+
"loss": 0.3624,
|
| 1287 |
+
"step": 1620
|
| 1288 |
+
},
|
| 1289 |
+
{
|
| 1290 |
+
"epoch": 6.08,
|
| 1291 |
+
"grad_norm": 4.328847885131836,
|
| 1292 |
+
"learning_rate": 3.9291044776119404e-05,
|
| 1293 |
+
"loss": 0.3478,
|
| 1294 |
+
"step": 1630
|
| 1295 |
+
},
|
| 1296 |
+
{
|
| 1297 |
+
"epoch": 6.12,
|
| 1298 |
+
"grad_norm": 3.5757558345794678,
|
| 1299 |
+
"learning_rate": 3.8917910447761195e-05,
|
| 1300 |
+
"loss": 0.2208,
|
| 1301 |
+
"step": 1640
|
| 1302 |
+
},
|
| 1303 |
+
{
|
| 1304 |
+
"epoch": 6.16,
|
| 1305 |
+
"grad_norm": 8.37783432006836,
|
| 1306 |
+
"learning_rate": 3.854477611940299e-05,
|
| 1307 |
+
"loss": 0.3614,
|
| 1308 |
+
"step": 1650
|
| 1309 |
+
},
|
| 1310 |
+
{
|
| 1311 |
+
"epoch": 6.19,
|
| 1312 |
+
"grad_norm": 2.4890713691711426,
|
| 1313 |
+
"learning_rate": 3.817164179104478e-05,
|
| 1314 |
+
"loss": 0.2514,
|
| 1315 |
+
"step": 1660
|
| 1316 |
+
},
|
| 1317 |
+
{
|
| 1318 |
+
"epoch": 6.23,
|
| 1319 |
+
"grad_norm": 8.873276710510254,
|
| 1320 |
+
"learning_rate": 3.7798507462686563e-05,
|
| 1321 |
+
"loss": 0.2233,
|
| 1322 |
+
"step": 1670
|
| 1323 |
+
},
|
| 1324 |
+
{
|
| 1325 |
+
"epoch": 6.27,
|
| 1326 |
+
"grad_norm": 0.29393309354782104,
|
| 1327 |
+
"learning_rate": 3.742537313432836e-05,
|
| 1328 |
+
"loss": 0.2474,
|
| 1329 |
+
"step": 1680
|
| 1330 |
+
},
|
| 1331 |
+
{
|
| 1332 |
+
"epoch": 6.31,
|
| 1333 |
+
"grad_norm": 3.810150384902954,
|
| 1334 |
+
"learning_rate": 3.7052238805970153e-05,
|
| 1335 |
+
"loss": 0.2481,
|
| 1336 |
+
"step": 1690
|
| 1337 |
+
},
|
| 1338 |
+
{
|
| 1339 |
+
"epoch": 6.34,
|
| 1340 |
+
"grad_norm": 1.989057183265686,
|
| 1341 |
+
"learning_rate": 3.6679104477611945e-05,
|
| 1342 |
+
"loss": 0.2626,
|
| 1343 |
+
"step": 1700
|
| 1344 |
+
},
|
| 1345 |
+
{
|
| 1346 |
+
"epoch": 6.34,
|
| 1347 |
+
"eval_accuracy": 0.8488576449912126,
|
| 1348 |
+
"eval_loss": 0.5402765274047852,
|
| 1349 |
+
"eval_runtime": 7.9293,
|
| 1350 |
+
"eval_samples_per_second": 71.759,
|
| 1351 |
+
"eval_steps_per_second": 9.08,
|
| 1352 |
+
"step": 1700
|
| 1353 |
+
},
|
| 1354 |
+
{
|
| 1355 |
+
"epoch": 6.38,
|
| 1356 |
+
"grad_norm": 8.488819122314453,
|
| 1357 |
+
"learning_rate": 3.630597014925373e-05,
|
| 1358 |
+
"loss": 0.2826,
|
| 1359 |
+
"step": 1710
|
| 1360 |
+
},
|
| 1361 |
+
{
|
| 1362 |
+
"epoch": 6.42,
|
| 1363 |
+
"grad_norm": 5.542993068695068,
|
| 1364 |
+
"learning_rate": 3.593283582089552e-05,
|
| 1365 |
+
"loss": 0.3552,
|
| 1366 |
+
"step": 1720
|
| 1367 |
+
},
|
| 1368 |
+
{
|
| 1369 |
+
"epoch": 6.46,
|
| 1370 |
+
"grad_norm": 6.646905422210693,
|
| 1371 |
+
"learning_rate": 3.555970149253732e-05,
|
| 1372 |
+
"loss": 0.4405,
|
| 1373 |
+
"step": 1730
|
| 1374 |
+
},
|
| 1375 |
+
{
|
| 1376 |
+
"epoch": 6.49,
|
| 1377 |
+
"grad_norm": 4.022976398468018,
|
| 1378 |
+
"learning_rate": 3.5186567164179105e-05,
|
| 1379 |
+
"loss": 0.2738,
|
| 1380 |
+
"step": 1740
|
| 1381 |
+
},
|
| 1382 |
+
{
|
| 1383 |
+
"epoch": 6.53,
|
| 1384 |
+
"grad_norm": 3.5472657680511475,
|
| 1385 |
+
"learning_rate": 3.4813432835820896e-05,
|
| 1386 |
+
"loss": 0.2807,
|
| 1387 |
+
"step": 1750
|
| 1388 |
+
},
|
| 1389 |
+
{
|
| 1390 |
+
"epoch": 6.57,
|
| 1391 |
+
"grad_norm": 12.070052146911621,
|
| 1392 |
+
"learning_rate": 3.444029850746269e-05,
|
| 1393 |
+
"loss": 0.3634,
|
| 1394 |
+
"step": 1760
|
| 1395 |
+
},
|
| 1396 |
+
{
|
| 1397 |
+
"epoch": 6.6,
|
| 1398 |
+
"grad_norm": 5.368374347686768,
|
| 1399 |
+
"learning_rate": 3.406716417910448e-05,
|
| 1400 |
+
"loss": 0.3252,
|
| 1401 |
+
"step": 1770
|
| 1402 |
+
},
|
| 1403 |
+
{
|
| 1404 |
+
"epoch": 6.64,
|
| 1405 |
+
"grad_norm": 5.566130638122559,
|
| 1406 |
+
"learning_rate": 3.369402985074627e-05,
|
| 1407 |
+
"loss": 0.3034,
|
| 1408 |
+
"step": 1780
|
| 1409 |
+
},
|
| 1410 |
+
{
|
| 1411 |
+
"epoch": 6.68,
|
| 1412 |
+
"grad_norm": 5.875336170196533,
|
| 1413 |
+
"learning_rate": 3.332089552238806e-05,
|
| 1414 |
+
"loss": 0.3406,
|
| 1415 |
+
"step": 1790
|
| 1416 |
+
},
|
| 1417 |
+
{
|
| 1418 |
+
"epoch": 6.72,
|
| 1419 |
+
"grad_norm": 2.4168920516967773,
|
| 1420 |
+
"learning_rate": 3.2947761194029854e-05,
|
| 1421 |
+
"loss": 0.3041,
|
| 1422 |
+
"step": 1800
|
| 1423 |
+
},
|
| 1424 |
+
{
|
| 1425 |
+
"epoch": 6.72,
|
| 1426 |
+
"eval_accuracy": 0.8488576449912126,
|
| 1427 |
+
"eval_loss": 0.4907586872577667,
|
| 1428 |
+
"eval_runtime": 7.8209,
|
| 1429 |
+
"eval_samples_per_second": 72.754,
|
| 1430 |
+
"eval_steps_per_second": 9.206,
|
| 1431 |
+
"step": 1800
|
| 1432 |
+
},
|
| 1433 |
+
{
|
| 1434 |
+
"epoch": 6.75,
|
| 1435 |
+
"grad_norm": 3.1040282249450684,
|
| 1436 |
+
"learning_rate": 3.2574626865671646e-05,
|
| 1437 |
+
"loss": 0.3167,
|
| 1438 |
+
"step": 1810
|
| 1439 |
+
},
|
| 1440 |
+
{
|
| 1441 |
+
"epoch": 6.79,
|
| 1442 |
+
"grad_norm": 1.8458846807479858,
|
| 1443 |
+
"learning_rate": 3.220149253731343e-05,
|
| 1444 |
+
"loss": 0.2061,
|
| 1445 |
+
"step": 1820
|
| 1446 |
+
},
|
| 1447 |
+
{
|
| 1448 |
+
"epoch": 6.83,
|
| 1449 |
+
"grad_norm": 0.4053177833557129,
|
| 1450 |
+
"learning_rate": 3.182835820895523e-05,
|
| 1451 |
+
"loss": 0.3113,
|
| 1452 |
+
"step": 1830
|
| 1453 |
+
},
|
| 1454 |
+
{
|
| 1455 |
+
"epoch": 6.87,
|
| 1456 |
+
"grad_norm": 0.23064230382442474,
|
| 1457 |
+
"learning_rate": 3.145522388059702e-05,
|
| 1458 |
+
"loss": 0.2368,
|
| 1459 |
+
"step": 1840
|
| 1460 |
+
},
|
| 1461 |
+
{
|
| 1462 |
+
"epoch": 6.9,
|
| 1463 |
+
"grad_norm": 1.006479263305664,
|
| 1464 |
+
"learning_rate": 3.1082089552238805e-05,
|
| 1465 |
+
"loss": 0.2265,
|
| 1466 |
+
"step": 1850
|
| 1467 |
+
},
|
| 1468 |
+
{
|
| 1469 |
+
"epoch": 6.94,
|
| 1470 |
+
"grad_norm": 4.072957992553711,
|
| 1471 |
+
"learning_rate": 3.07089552238806e-05,
|
| 1472 |
+
"loss": 0.2976,
|
| 1473 |
+
"step": 1860
|
| 1474 |
+
},
|
| 1475 |
+
{
|
| 1476 |
+
"epoch": 6.98,
|
| 1477 |
+
"grad_norm": 16.575963973999023,
|
| 1478 |
+
"learning_rate": 3.033582089552239e-05,
|
| 1479 |
+
"loss": 0.1504,
|
| 1480 |
+
"step": 1870
|
| 1481 |
+
},
|
| 1482 |
+
{
|
| 1483 |
+
"epoch": 7.01,
|
| 1484 |
+
"grad_norm": 2.9144656658172607,
|
| 1485 |
+
"learning_rate": 2.9962686567164183e-05,
|
| 1486 |
+
"loss": 0.2156,
|
| 1487 |
+
"step": 1880
|
| 1488 |
+
},
|
| 1489 |
+
{
|
| 1490 |
+
"epoch": 7.05,
|
| 1491 |
+
"grad_norm": 4.547207832336426,
|
| 1492 |
+
"learning_rate": 2.958955223880597e-05,
|
| 1493 |
+
"loss": 0.2693,
|
| 1494 |
+
"step": 1890
|
| 1495 |
+
},
|
| 1496 |
+
{
|
| 1497 |
+
"epoch": 7.09,
|
| 1498 |
+
"grad_norm": 0.5566532611846924,
|
| 1499 |
+
"learning_rate": 2.9216417910447763e-05,
|
| 1500 |
+
"loss": 0.1831,
|
| 1501 |
+
"step": 1900
|
| 1502 |
+
},
|
| 1503 |
+
{
|
| 1504 |
+
"epoch": 7.09,
|
| 1505 |
+
"eval_accuracy": 0.8383128295254832,
|
| 1506 |
+
"eval_loss": 0.5721341967582703,
|
| 1507 |
+
"eval_runtime": 7.7377,
|
| 1508 |
+
"eval_samples_per_second": 73.536,
|
| 1509 |
+
"eval_steps_per_second": 9.305,
|
| 1510 |
+
"step": 1900
|
| 1511 |
+
},
|
| 1512 |
+
{
|
| 1513 |
+
"epoch": 7.13,
|
| 1514 |
+
"grad_norm": 7.9241838455200195,
|
| 1515 |
+
"learning_rate": 2.8843283582089555e-05,
|
| 1516 |
+
"loss": 0.3037,
|
| 1517 |
+
"step": 1910
|
| 1518 |
+
},
|
| 1519 |
+
{
|
| 1520 |
+
"epoch": 7.16,
|
| 1521 |
+
"grad_norm": 4.847833156585693,
|
| 1522 |
+
"learning_rate": 2.8470149253731343e-05,
|
| 1523 |
+
"loss": 0.2744,
|
| 1524 |
+
"step": 1920
|
| 1525 |
+
},
|
| 1526 |
+
{
|
| 1527 |
+
"epoch": 7.2,
|
| 1528 |
+
"grad_norm": 4.368974208831787,
|
| 1529 |
+
"learning_rate": 2.8097014925373134e-05,
|
| 1530 |
+
"loss": 0.1603,
|
| 1531 |
+
"step": 1930
|
| 1532 |
+
},
|
| 1533 |
+
{
|
| 1534 |
+
"epoch": 7.24,
|
| 1535 |
+
"grad_norm": 5.848027229309082,
|
| 1536 |
+
"learning_rate": 2.772388059701493e-05,
|
| 1537 |
+
"loss": 0.3318,
|
| 1538 |
+
"step": 1940
|
| 1539 |
+
},
|
| 1540 |
+
{
|
| 1541 |
+
"epoch": 7.28,
|
| 1542 |
+
"grad_norm": 5.53363037109375,
|
| 1543 |
+
"learning_rate": 2.7350746268656718e-05,
|
| 1544 |
+
"loss": 0.2568,
|
| 1545 |
+
"step": 1950
|
| 1546 |
+
},
|
| 1547 |
+
{
|
| 1548 |
+
"epoch": 7.31,
|
| 1549 |
+
"grad_norm": 1.3791863918304443,
|
| 1550 |
+
"learning_rate": 2.697761194029851e-05,
|
| 1551 |
+
"loss": 0.2186,
|
| 1552 |
+
"step": 1960
|
| 1553 |
+
},
|
| 1554 |
+
{
|
| 1555 |
+
"epoch": 7.35,
|
| 1556 |
+
"grad_norm": 13.533841133117676,
|
| 1557 |
+
"learning_rate": 2.6604477611940297e-05,
|
| 1558 |
+
"loss": 0.2772,
|
| 1559 |
+
"step": 1970
|
| 1560 |
+
},
|
| 1561 |
+
{
|
| 1562 |
+
"epoch": 7.39,
|
| 1563 |
+
"grad_norm": 1.113595962524414,
|
| 1564 |
+
"learning_rate": 2.623134328358209e-05,
|
| 1565 |
+
"loss": 0.3396,
|
| 1566 |
+
"step": 1980
|
| 1567 |
+
},
|
| 1568 |
+
{
|
| 1569 |
+
"epoch": 7.43,
|
| 1570 |
+
"grad_norm": 3.193376064300537,
|
| 1571 |
+
"learning_rate": 2.5858208955223884e-05,
|
| 1572 |
+
"loss": 0.2171,
|
| 1573 |
+
"step": 1990
|
| 1574 |
+
},
|
| 1575 |
+
{
|
| 1576 |
+
"epoch": 7.46,
|
| 1577 |
+
"grad_norm": 2.8687243461608887,
|
| 1578 |
+
"learning_rate": 2.5485074626865672e-05,
|
| 1579 |
+
"loss": 0.2275,
|
| 1580 |
+
"step": 2000
|
| 1581 |
+
},
|
| 1582 |
+
{
|
| 1583 |
+
"epoch": 7.46,
|
| 1584 |
+
"eval_accuracy": 0.8312829525483304,
|
| 1585 |
+
"eval_loss": 0.5349107980728149,
|
| 1586 |
+
"eval_runtime": 8.0113,
|
| 1587 |
+
"eval_samples_per_second": 71.025,
|
| 1588 |
+
"eval_steps_per_second": 8.987,
|
| 1589 |
+
"step": 2000
|
| 1590 |
+
},
|
| 1591 |
+
{
|
| 1592 |
+
"epoch": 7.5,
|
| 1593 |
+
"grad_norm": 6.330258846282959,
|
| 1594 |
+
"learning_rate": 2.5111940298507464e-05,
|
| 1595 |
+
"loss": 0.2165,
|
| 1596 |
+
"step": 2010
|
| 1597 |
+
},
|
| 1598 |
+
{
|
| 1599 |
+
"epoch": 7.54,
|
| 1600 |
+
"grad_norm": 2.457519769668579,
|
| 1601 |
+
"learning_rate": 2.4738805970149252e-05,
|
| 1602 |
+
"loss": 0.3275,
|
| 1603 |
+
"step": 2020
|
| 1604 |
+
},
|
| 1605 |
+
{
|
| 1606 |
+
"epoch": 7.57,
|
| 1607 |
+
"grad_norm": 1.468772053718567,
|
| 1608 |
+
"learning_rate": 2.4365671641791047e-05,
|
| 1609 |
+
"loss": 0.186,
|
| 1610 |
+
"step": 2030
|
| 1611 |
+
},
|
| 1612 |
+
{
|
| 1613 |
+
"epoch": 7.61,
|
| 1614 |
+
"grad_norm": 4.308888912200928,
|
| 1615 |
+
"learning_rate": 2.3992537313432835e-05,
|
| 1616 |
+
"loss": 0.3182,
|
| 1617 |
+
"step": 2040
|
| 1618 |
+
},
|
| 1619 |
+
{
|
| 1620 |
+
"epoch": 7.65,
|
| 1621 |
+
"grad_norm": 1.8849867582321167,
|
| 1622 |
+
"learning_rate": 2.361940298507463e-05,
|
| 1623 |
+
"loss": 0.2631,
|
| 1624 |
+
"step": 2050
|
| 1625 |
+
},
|
| 1626 |
+
{
|
| 1627 |
+
"epoch": 7.69,
|
| 1628 |
+
"grad_norm": 2.6795170307159424,
|
| 1629 |
+
"learning_rate": 2.3246268656716418e-05,
|
| 1630 |
+
"loss": 0.1724,
|
| 1631 |
+
"step": 2060
|
| 1632 |
+
},
|
| 1633 |
+
{
|
| 1634 |
+
"epoch": 7.72,
|
| 1635 |
+
"grad_norm": 0.22702960669994354,
|
| 1636 |
+
"learning_rate": 2.287313432835821e-05,
|
| 1637 |
+
"loss": 0.2542,
|
| 1638 |
+
"step": 2070
|
| 1639 |
+
},
|
| 1640 |
+
{
|
| 1641 |
+
"epoch": 7.76,
|
| 1642 |
+
"grad_norm": 4.6633429527282715,
|
| 1643 |
+
"learning_rate": 2.25e-05,
|
| 1644 |
+
"loss": 0.259,
|
| 1645 |
+
"step": 2080
|
| 1646 |
+
},
|
| 1647 |
+
{
|
| 1648 |
+
"epoch": 7.8,
|
| 1649 |
+
"grad_norm": 6.543178558349609,
|
| 1650 |
+
"learning_rate": 2.2126865671641793e-05,
|
| 1651 |
+
"loss": 0.3752,
|
| 1652 |
+
"step": 2090
|
| 1653 |
+
},
|
| 1654 |
+
{
|
| 1655 |
+
"epoch": 7.84,
|
| 1656 |
+
"grad_norm": 7.109080791473389,
|
| 1657 |
+
"learning_rate": 2.1753731343283585e-05,
|
| 1658 |
+
"loss": 0.1762,
|
| 1659 |
+
"step": 2100
|
| 1660 |
+
},
|
| 1661 |
+
{
|
| 1662 |
+
"epoch": 7.84,
|
| 1663 |
+
"eval_accuracy": 0.8541300527240774,
|
| 1664 |
+
"eval_loss": 0.5203543901443481,
|
| 1665 |
+
"eval_runtime": 7.8922,
|
| 1666 |
+
"eval_samples_per_second": 72.096,
|
| 1667 |
+
"eval_steps_per_second": 9.123,
|
| 1668 |
+
"step": 2100
|
| 1669 |
+
},
|
| 1670 |
+
{
|
| 1671 |
+
"epoch": 7.87,
|
| 1672 |
+
"grad_norm": 3.3965115547180176,
|
| 1673 |
+
"learning_rate": 2.1380597014925373e-05,
|
| 1674 |
+
"loss": 0.1965,
|
| 1675 |
+
"step": 2110
|
| 1676 |
+
},
|
| 1677 |
+
{
|
| 1678 |
+
"epoch": 7.91,
|
| 1679 |
+
"grad_norm": 0.1386798918247223,
|
| 1680 |
+
"learning_rate": 2.1007462686567164e-05,
|
| 1681 |
+
"loss": 0.1448,
|
| 1682 |
+
"step": 2120
|
| 1683 |
+
},
|
| 1684 |
+
{
|
| 1685 |
+
"epoch": 7.95,
|
| 1686 |
+
"grad_norm": 8.268773078918457,
|
| 1687 |
+
"learning_rate": 2.0634328358208956e-05,
|
| 1688 |
+
"loss": 0.2203,
|
| 1689 |
+
"step": 2130
|
| 1690 |
+
},
|
| 1691 |
+
{
|
| 1692 |
+
"epoch": 7.99,
|
| 1693 |
+
"grad_norm": 2.712890625,
|
| 1694 |
+
"learning_rate": 2.0261194029850748e-05,
|
| 1695 |
+
"loss": 0.2104,
|
| 1696 |
+
"step": 2140
|
| 1697 |
+
},
|
| 1698 |
+
{
|
| 1699 |
+
"epoch": 8.02,
|
| 1700 |
+
"grad_norm": 2.0390050411224365,
|
| 1701 |
+
"learning_rate": 1.988805970149254e-05,
|
| 1702 |
+
"loss": 0.2063,
|
| 1703 |
+
"step": 2150
|
| 1704 |
+
},
|
| 1705 |
+
{
|
| 1706 |
+
"epoch": 8.06,
|
| 1707 |
+
"grad_norm": 4.355598449707031,
|
| 1708 |
+
"learning_rate": 1.951492537313433e-05,
|
| 1709 |
+
"loss": 0.1356,
|
| 1710 |
+
"step": 2160
|
| 1711 |
+
},
|
| 1712 |
+
{
|
| 1713 |
+
"epoch": 8.1,
|
| 1714 |
+
"grad_norm": 9.854630470275879,
|
| 1715 |
+
"learning_rate": 1.914179104477612e-05,
|
| 1716 |
+
"loss": 0.1686,
|
| 1717 |
+
"step": 2170
|
| 1718 |
+
},
|
| 1719 |
+
{
|
| 1720 |
+
"epoch": 8.13,
|
| 1721 |
+
"grad_norm": 4.178330421447754,
|
| 1722 |
+
"learning_rate": 1.8768656716417914e-05,
|
| 1723 |
+
"loss": 0.2578,
|
| 1724 |
+
"step": 2180
|
| 1725 |
+
},
|
| 1726 |
+
{
|
| 1727 |
+
"epoch": 8.17,
|
| 1728 |
+
"grad_norm": 5.019784450531006,
|
| 1729 |
+
"learning_rate": 1.8395522388059702e-05,
|
| 1730 |
+
"loss": 0.1923,
|
| 1731 |
+
"step": 2190
|
| 1732 |
+
},
|
| 1733 |
+
{
|
| 1734 |
+
"epoch": 8.21,
|
| 1735 |
+
"grad_norm": 3.8136210441589355,
|
| 1736 |
+
"learning_rate": 1.8022388059701494e-05,
|
| 1737 |
+
"loss": 0.2112,
|
| 1738 |
+
"step": 2200
|
| 1739 |
+
},
|
| 1740 |
+
{
|
| 1741 |
+
"epoch": 8.21,
|
| 1742 |
+
"eval_accuracy": 0.8629173989455184,
|
| 1743 |
+
"eval_loss": 0.5188840627670288,
|
| 1744 |
+
"eval_runtime": 8.1412,
|
| 1745 |
+
"eval_samples_per_second": 69.891,
|
| 1746 |
+
"eval_steps_per_second": 8.844,
|
| 1747 |
+
"step": 2200
|
| 1748 |
+
},
|
| 1749 |
+
{
|
| 1750 |
+
"epoch": 8.25,
|
| 1751 |
+
"grad_norm": 2.7035305500030518,
|
| 1752 |
+
"learning_rate": 1.7649253731343285e-05,
|
| 1753 |
+
"loss": 0.2501,
|
| 1754 |
+
"step": 2210
|
| 1755 |
+
},
|
| 1756 |
+
{
|
| 1757 |
+
"epoch": 8.28,
|
| 1758 |
+
"grad_norm": 6.736306190490723,
|
| 1759 |
+
"learning_rate": 1.7276119402985073e-05,
|
| 1760 |
+
"loss": 0.2213,
|
| 1761 |
+
"step": 2220
|
| 1762 |
+
},
|
| 1763 |
+
{
|
| 1764 |
+
"epoch": 8.32,
|
| 1765 |
+
"grad_norm": 3.0436556339263916,
|
| 1766 |
+
"learning_rate": 1.690298507462687e-05,
|
| 1767 |
+
"loss": 0.1285,
|
| 1768 |
+
"step": 2230
|
| 1769 |
+
},
|
| 1770 |
+
{
|
| 1771 |
+
"epoch": 8.36,
|
| 1772 |
+
"grad_norm": 4.729572772979736,
|
| 1773 |
+
"learning_rate": 1.6529850746268657e-05,
|
| 1774 |
+
"loss": 0.2984,
|
| 1775 |
+
"step": 2240
|
| 1776 |
+
},
|
| 1777 |
+
{
|
| 1778 |
+
"epoch": 8.4,
|
| 1779 |
+
"grad_norm": 3.6665098667144775,
|
| 1780 |
+
"learning_rate": 1.6156716417910448e-05,
|
| 1781 |
+
"loss": 0.1796,
|
| 1782 |
+
"step": 2250
|
| 1783 |
+
},
|
| 1784 |
+
{
|
| 1785 |
+
"epoch": 8.43,
|
| 1786 |
+
"grad_norm": 8.485068321228027,
|
| 1787 |
+
"learning_rate": 1.578358208955224e-05,
|
| 1788 |
+
"loss": 0.2137,
|
| 1789 |
+
"step": 2260
|
| 1790 |
+
},
|
| 1791 |
+
{
|
| 1792 |
+
"epoch": 8.47,
|
| 1793 |
+
"grad_norm": 4.643974304199219,
|
| 1794 |
+
"learning_rate": 1.541044776119403e-05,
|
| 1795 |
+
"loss": 0.3009,
|
| 1796 |
+
"step": 2270
|
| 1797 |
+
},
|
| 1798 |
+
{
|
| 1799 |
+
"epoch": 8.51,
|
| 1800 |
+
"grad_norm": 2.91859769821167,
|
| 1801 |
+
"learning_rate": 1.5037313432835823e-05,
|
| 1802 |
+
"loss": 0.1855,
|
| 1803 |
+
"step": 2280
|
| 1804 |
+
},
|
| 1805 |
+
{
|
| 1806 |
+
"epoch": 8.54,
|
| 1807 |
+
"grad_norm": 9.799684524536133,
|
| 1808 |
+
"learning_rate": 1.4664179104477613e-05,
|
| 1809 |
+
"loss": 0.2186,
|
| 1810 |
+
"step": 2290
|
| 1811 |
+
},
|
| 1812 |
+
{
|
| 1813 |
+
"epoch": 8.58,
|
| 1814 |
+
"grad_norm": 4.92659330368042,
|
| 1815 |
+
"learning_rate": 1.4291044776119403e-05,
|
| 1816 |
+
"loss": 0.1242,
|
| 1817 |
+
"step": 2300
|
| 1818 |
+
},
|
| 1819 |
+
{
|
| 1820 |
+
"epoch": 8.58,
|
| 1821 |
+
"eval_accuracy": 0.8471001757469244,
|
| 1822 |
+
"eval_loss": 0.5376706123352051,
|
| 1823 |
+
"eval_runtime": 7.8653,
|
| 1824 |
+
"eval_samples_per_second": 72.343,
|
| 1825 |
+
"eval_steps_per_second": 9.154,
|
| 1826 |
+
"step": 2300
|
| 1827 |
+
},
|
| 1828 |
+
{
|
| 1829 |
+
"epoch": 8.62,
|
| 1830 |
+
"grad_norm": 0.7728621363639832,
|
| 1831 |
+
"learning_rate": 1.3917910447761196e-05,
|
| 1832 |
+
"loss": 0.2769,
|
| 1833 |
+
"step": 2310
|
| 1834 |
+
},
|
| 1835 |
+
{
|
| 1836 |
+
"epoch": 8.66,
|
| 1837 |
+
"grad_norm": 3.757192373275757,
|
| 1838 |
+
"learning_rate": 1.3544776119402986e-05,
|
| 1839 |
+
"loss": 0.31,
|
| 1840 |
+
"step": 2320
|
| 1841 |
+
},
|
| 1842 |
+
{
|
| 1843 |
+
"epoch": 8.69,
|
| 1844 |
+
"grad_norm": 5.901330471038818,
|
| 1845 |
+
"learning_rate": 1.3171641791044777e-05,
|
| 1846 |
+
"loss": 0.2488,
|
| 1847 |
+
"step": 2330
|
| 1848 |
+
},
|
| 1849 |
+
{
|
| 1850 |
+
"epoch": 8.73,
|
| 1851 |
+
"grad_norm": 0.1360226422548294,
|
| 1852 |
+
"learning_rate": 1.2798507462686567e-05,
|
| 1853 |
+
"loss": 0.2359,
|
| 1854 |
+
"step": 2340
|
| 1855 |
+
},
|
| 1856 |
+
{
|
| 1857 |
+
"epoch": 8.77,
|
| 1858 |
+
"grad_norm": 5.801501750946045,
|
| 1859 |
+
"learning_rate": 1.2425373134328359e-05,
|
| 1860 |
+
"loss": 0.23,
|
| 1861 |
+
"step": 2350
|
| 1862 |
+
},
|
| 1863 |
+
{
|
| 1864 |
+
"epoch": 8.81,
|
| 1865 |
+
"grad_norm": 3.3060359954833984,
|
| 1866 |
+
"learning_rate": 1.2052238805970149e-05,
|
| 1867 |
+
"loss": 0.1114,
|
| 1868 |
+
"step": 2360
|
| 1869 |
+
},
|
| 1870 |
+
{
|
| 1871 |
+
"epoch": 8.84,
|
| 1872 |
+
"grad_norm": 2.0813100337982178,
|
| 1873 |
+
"learning_rate": 1.167910447761194e-05,
|
| 1874 |
+
"loss": 0.1569,
|
| 1875 |
+
"step": 2370
|
| 1876 |
+
},
|
| 1877 |
+
{
|
| 1878 |
+
"epoch": 8.88,
|
| 1879 |
+
"grad_norm": 0.42951256036758423,
|
| 1880 |
+
"learning_rate": 1.1305970149253732e-05,
|
| 1881 |
+
"loss": 0.2636,
|
| 1882 |
+
"step": 2380
|
| 1883 |
+
},
|
| 1884 |
+
{
|
| 1885 |
+
"epoch": 8.92,
|
| 1886 |
+
"grad_norm": 3.2714788913726807,
|
| 1887 |
+
"learning_rate": 1.0932835820895524e-05,
|
| 1888 |
+
"loss": 0.2197,
|
| 1889 |
+
"step": 2390
|
| 1890 |
+
},
|
| 1891 |
+
{
|
| 1892 |
+
"epoch": 8.96,
|
| 1893 |
+
"grad_norm": 4.24855375289917,
|
| 1894 |
+
"learning_rate": 1.0559701492537313e-05,
|
| 1895 |
+
"loss": 0.1207,
|
| 1896 |
+
"step": 2400
|
| 1897 |
+
},
|
| 1898 |
+
{
|
| 1899 |
+
"epoch": 8.96,
|
| 1900 |
+
"eval_accuracy": 0.8558875219683656,
|
| 1901 |
+
"eval_loss": 0.5324714779853821,
|
| 1902 |
+
"eval_runtime": 7.9022,
|
| 1903 |
+
"eval_samples_per_second": 72.006,
|
| 1904 |
+
"eval_steps_per_second": 9.111,
|
| 1905 |
+
"step": 2400
|
| 1906 |
+
},
|
| 1907 |
+
{
|
| 1908 |
+
"epoch": 8.99,
|
| 1909 |
+
"grad_norm": 3.989713430404663,
|
| 1910 |
+
"learning_rate": 1.0186567164179105e-05,
|
| 1911 |
+
"loss": 0.2336,
|
| 1912 |
+
"step": 2410
|
| 1913 |
+
},
|
| 1914 |
+
{
|
| 1915 |
+
"epoch": 9.03,
|
| 1916 |
+
"grad_norm": 5.590869903564453,
|
| 1917 |
+
"learning_rate": 9.813432835820897e-06,
|
| 1918 |
+
"loss": 0.2292,
|
| 1919 |
+
"step": 2420
|
| 1920 |
+
},
|
| 1921 |
+
{
|
| 1922 |
+
"epoch": 9.07,
|
| 1923 |
+
"grad_norm": 3.405966281890869,
|
| 1924 |
+
"learning_rate": 9.440298507462688e-06,
|
| 1925 |
+
"loss": 0.1654,
|
| 1926 |
+
"step": 2430
|
| 1927 |
+
},
|
| 1928 |
+
{
|
| 1929 |
+
"epoch": 9.1,
|
| 1930 |
+
"grad_norm": 3.733381986618042,
|
| 1931 |
+
"learning_rate": 9.067164179104478e-06,
|
| 1932 |
+
"loss": 0.2104,
|
| 1933 |
+
"step": 2440
|
| 1934 |
+
},
|
| 1935 |
+
{
|
| 1936 |
+
"epoch": 9.14,
|
| 1937 |
+
"grad_norm": 0.1994183361530304,
|
| 1938 |
+
"learning_rate": 8.694029850746268e-06,
|
| 1939 |
+
"loss": 0.0789,
|
| 1940 |
+
"step": 2450
|
| 1941 |
+
},
|
| 1942 |
+
{
|
| 1943 |
+
"epoch": 9.18,
|
| 1944 |
+
"grad_norm": 7.948019504547119,
|
| 1945 |
+
"learning_rate": 8.32089552238806e-06,
|
| 1946 |
+
"loss": 0.3335,
|
| 1947 |
+
"step": 2460
|
| 1948 |
+
},
|
| 1949 |
+
{
|
| 1950 |
+
"epoch": 9.22,
|
| 1951 |
+
"grad_norm": 3.020522117614746,
|
| 1952 |
+
"learning_rate": 7.947761194029851e-06,
|
| 1953 |
+
"loss": 0.1838,
|
| 1954 |
+
"step": 2470
|
| 1955 |
+
},
|
| 1956 |
+
{
|
| 1957 |
+
"epoch": 9.25,
|
| 1958 |
+
"grad_norm": 2.4797592163085938,
|
| 1959 |
+
"learning_rate": 7.574626865671643e-06,
|
| 1960 |
+
"loss": 0.1573,
|
| 1961 |
+
"step": 2480
|
| 1962 |
+
},
|
| 1963 |
+
{
|
| 1964 |
+
"epoch": 9.29,
|
| 1965 |
+
"grad_norm": 0.7854322195053101,
|
| 1966 |
+
"learning_rate": 7.201492537313433e-06,
|
| 1967 |
+
"loss": 0.1868,
|
| 1968 |
+
"step": 2490
|
| 1969 |
+
},
|
| 1970 |
+
{
|
| 1971 |
+
"epoch": 9.33,
|
| 1972 |
+
"grad_norm": 8.424530982971191,
|
| 1973 |
+
"learning_rate": 6.828358208955224e-06,
|
| 1974 |
+
"loss": 0.1806,
|
| 1975 |
+
"step": 2500
|
| 1976 |
+
},
|
| 1977 |
+
{
|
| 1978 |
+
"epoch": 9.33,
|
| 1979 |
+
"eval_accuracy": 0.8646748681898067,
|
| 1980 |
+
"eval_loss": 0.5149648785591125,
|
| 1981 |
+
"eval_runtime": 7.8422,
|
| 1982 |
+
"eval_samples_per_second": 72.556,
|
| 1983 |
+
"eval_steps_per_second": 9.181,
|
| 1984 |
+
"step": 2500
|
| 1985 |
+
},
|
| 1986 |
+
{
|
| 1987 |
+
"epoch": 9.37,
|
| 1988 |
+
"grad_norm": 2.9176523685455322,
|
| 1989 |
+
"learning_rate": 6.455223880597015e-06,
|
| 1990 |
+
"loss": 0.1977,
|
| 1991 |
+
"step": 2510
|
| 1992 |
+
},
|
| 1993 |
+
{
|
| 1994 |
+
"epoch": 9.4,
|
| 1995 |
+
"grad_norm": 4.15384578704834,
|
| 1996 |
+
"learning_rate": 6.082089552238806e-06,
|
| 1997 |
+
"loss": 0.2007,
|
| 1998 |
+
"step": 2520
|
| 1999 |
+
},
|
| 2000 |
+
{
|
| 2001 |
+
"epoch": 9.44,
|
| 2002 |
+
"grad_norm": 2.4758641719818115,
|
| 2003 |
+
"learning_rate": 5.708955223880597e-06,
|
| 2004 |
+
"loss": 0.2,
|
| 2005 |
+
"step": 2530
|
| 2006 |
+
},
|
| 2007 |
+
{
|
| 2008 |
+
"epoch": 9.48,
|
| 2009 |
+
"grad_norm": 4.053123950958252,
|
| 2010 |
+
"learning_rate": 5.335820895522389e-06,
|
| 2011 |
+
"loss": 0.2514,
|
| 2012 |
+
"step": 2540
|
| 2013 |
+
},
|
| 2014 |
+
{
|
| 2015 |
+
"epoch": 9.51,
|
| 2016 |
+
"grad_norm": 2.3916337490081787,
|
| 2017 |
+
"learning_rate": 4.9626865671641796e-06,
|
| 2018 |
+
"loss": 0.2104,
|
| 2019 |
+
"step": 2550
|
| 2020 |
+
},
|
| 2021 |
+
{
|
| 2022 |
+
"epoch": 9.55,
|
| 2023 |
+
"grad_norm": 4.113661766052246,
|
| 2024 |
+
"learning_rate": 4.58955223880597e-06,
|
| 2025 |
+
"loss": 0.1998,
|
| 2026 |
+
"step": 2560
|
| 2027 |
+
},
|
| 2028 |
+
{
|
| 2029 |
+
"epoch": 9.59,
|
| 2030 |
+
"grad_norm": 3.558722972869873,
|
| 2031 |
+
"learning_rate": 4.216417910447761e-06,
|
| 2032 |
+
"loss": 0.144,
|
| 2033 |
+
"step": 2570
|
| 2034 |
+
},
|
| 2035 |
+
{
|
| 2036 |
+
"epoch": 9.63,
|
| 2037 |
+
"grad_norm": 2.689765691757202,
|
| 2038 |
+
"learning_rate": 3.843283582089553e-06,
|
| 2039 |
+
"loss": 0.1691,
|
| 2040 |
+
"step": 2580
|
| 2041 |
+
},
|
| 2042 |
+
{
|
| 2043 |
+
"epoch": 9.66,
|
| 2044 |
+
"grad_norm": 4.95484733581543,
|
| 2045 |
+
"learning_rate": 3.4701492537313434e-06,
|
| 2046 |
+
"loss": 0.1875,
|
| 2047 |
+
"step": 2590
|
| 2048 |
+
},
|
| 2049 |
+
{
|
| 2050 |
+
"epoch": 9.7,
|
| 2051 |
+
"grad_norm": 6.025635242462158,
|
| 2052 |
+
"learning_rate": 3.0970149253731345e-06,
|
| 2053 |
+
"loss": 0.1793,
|
| 2054 |
+
"step": 2600
|
| 2055 |
+
},
|
| 2056 |
+
{
|
| 2057 |
+
"epoch": 9.7,
|
| 2058 |
+
"eval_accuracy": 0.8664323374340949,
|
| 2059 |
+
"eval_loss": 0.5153330564498901,
|
| 2060 |
+
"eval_runtime": 7.9144,
|
| 2061 |
+
"eval_samples_per_second": 71.894,
|
| 2062 |
+
"eval_steps_per_second": 9.097,
|
| 2063 |
+
"step": 2600
|
| 2064 |
+
},
|
| 2065 |
+
{
|
| 2066 |
+
"epoch": 9.74,
|
| 2067 |
+
"grad_norm": 0.3092793822288513,
|
| 2068 |
+
"learning_rate": 2.7238805970149257e-06,
|
| 2069 |
+
"loss": 0.1385,
|
| 2070 |
+
"step": 2610
|
| 2071 |
+
},
|
| 2072 |
+
{
|
| 2073 |
+
"epoch": 9.78,
|
| 2074 |
+
"grad_norm": 1.1317028999328613,
|
| 2075 |
+
"learning_rate": 2.3507462686567164e-06,
|
| 2076 |
+
"loss": 0.1628,
|
| 2077 |
+
"step": 2620
|
| 2078 |
+
},
|
| 2079 |
+
{
|
| 2080 |
+
"epoch": 9.81,
|
| 2081 |
+
"grad_norm": 7.642726898193359,
|
| 2082 |
+
"learning_rate": 1.9776119402985076e-06,
|
| 2083 |
+
"loss": 0.2142,
|
| 2084 |
+
"step": 2630
|
| 2085 |
+
},
|
| 2086 |
+
{
|
| 2087 |
+
"epoch": 9.85,
|
| 2088 |
+
"grad_norm": 4.3891191482543945,
|
| 2089 |
+
"learning_rate": 1.6044776119402985e-06,
|
| 2090 |
+
"loss": 0.2115,
|
| 2091 |
+
"step": 2640
|
| 2092 |
+
},
|
| 2093 |
+
{
|
| 2094 |
+
"epoch": 9.89,
|
| 2095 |
+
"grad_norm": 5.876834869384766,
|
| 2096 |
+
"learning_rate": 1.2313432835820897e-06,
|
| 2097 |
+
"loss": 0.2859,
|
| 2098 |
+
"step": 2650
|
| 2099 |
+
},
|
| 2100 |
+
{
|
| 2101 |
+
"epoch": 9.93,
|
| 2102 |
+
"grad_norm": 1.6104581356048584,
|
| 2103 |
+
"learning_rate": 8.582089552238806e-07,
|
| 2104 |
+
"loss": 0.2752,
|
| 2105 |
+
"step": 2660
|
| 2106 |
+
},
|
| 2107 |
+
{
|
| 2108 |
+
"epoch": 9.96,
|
| 2109 |
+
"grad_norm": 5.835386276245117,
|
| 2110 |
+
"learning_rate": 4.850746268656717e-07,
|
| 2111 |
+
"loss": 0.2057,
|
| 2112 |
+
"step": 2670
|
| 2113 |
+
},
|
| 2114 |
+
{
|
| 2115 |
+
"epoch": 10.0,
|
| 2116 |
+
"grad_norm": 7.006475925445557,
|
| 2117 |
+
"learning_rate": 1.119402985074627e-07,
|
| 2118 |
+
"loss": 0.2098,
|
| 2119 |
+
"step": 2680
|
| 2120 |
+
},
|
| 2121 |
{
|
| 2122 |
"epoch": 10.0,
|
| 2123 |
+
"step": 2680,
|
| 2124 |
"total_flos": 3.3230947683690086e+18,
|
| 2125 |
+
"train_loss": 0.45543073504718384,
|
| 2126 |
+
"train_runtime": 1353.2313,
|
| 2127 |
+
"train_samples_per_second": 31.687,
|
| 2128 |
+
"train_steps_per_second": 1.98
|
| 2129 |
}
|
| 2130 |
],
|
| 2131 |
+
"logging_steps": 10,
|
| 2132 |
+
"max_steps": 2680,
|
| 2133 |
"num_input_tokens_seen": 0,
|
| 2134 |
"num_train_epochs": 10,
|
| 2135 |
"save_steps": 100,
|
| 2136 |
"total_flos": 3.3230947683690086e+18,
|
| 2137 |
+
"train_batch_size": 16,
|
| 2138 |
"trial_name": null,
|
| 2139 |
"trial_params": null
|
| 2140 |
}
|