End of training
Browse files
all_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_accuracy": 0.
|
| 4 |
-
"eval_loss": 0.
|
| 5 |
-
"eval_runtime":
|
| 6 |
-
"eval_samples_per_second":
|
| 7 |
-
"eval_steps_per_second":
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 7.12,
|
| 3 |
+
"eval_accuracy": 0.9483870967741935,
|
| 4 |
+
"eval_loss": 0.2060869336128235,
|
| 5 |
+
"eval_runtime": 16.8116,
|
| 6 |
+
"eval_samples_per_second": 9.22,
|
| 7 |
+
"eval_steps_per_second": 2.32
|
| 8 |
}
|
runs/Jan06_19-59-38_44017.example.com/events.out.tfevents.1704568368.44017.example.com
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0ce021c659212b72ec1c4a88ed698724ba8feb748022c03a91157acde50292e0
|
| 3 |
+
size 363
|
test_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_accuracy": 0.
|
| 4 |
-
"eval_loss": 0.
|
| 5 |
-
"eval_runtime":
|
| 6 |
-
"eval_samples_per_second":
|
| 7 |
-
"eval_steps_per_second":
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 7.12,
|
| 3 |
+
"eval_accuracy": 0.9483870967741935,
|
| 4 |
+
"eval_loss": 0.2060869336128235,
|
| 5 |
+
"eval_runtime": 16.8116,
|
| 6 |
+
"eval_samples_per_second": 9.22,
|
| 7 |
+
"eval_steps_per_second": 2.32
|
| 8 |
}
|
trainer_state.json
CHANGED
|
@@ -1,262 +1,480 @@
|
|
| 1 |
{
|
| 2 |
-
"best_metric": 0.
|
| 3 |
-
"best_model_checkpoint": "videomae-base-finetuned-ucf101-subset/checkpoint-
|
| 4 |
-
"epoch":
|
| 5 |
"eval_steps": 500,
|
| 6 |
-
"global_step":
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
-
"epoch": 0.
|
| 13 |
-
"learning_rate":
|
| 14 |
-
"loss": 2.
|
| 15 |
"step": 10
|
| 16 |
},
|
| 17 |
{
|
| 18 |
-
"epoch": 0.
|
| 19 |
-
"learning_rate":
|
| 20 |
-
"loss": 2.
|
| 21 |
"step": 20
|
| 22 |
},
|
| 23 |
{
|
| 24 |
-
"epoch": 0.
|
| 25 |
-
"learning_rate": 5e-05,
|
| 26 |
-
"loss": 2.
|
| 27 |
"step": 30
|
| 28 |
},
|
| 29 |
{
|
| 30 |
-
"epoch": 0.
|
| 31 |
-
"learning_rate":
|
| 32 |
-
"loss": 2.
|
| 33 |
"step": 40
|
| 34 |
},
|
| 35 |
{
|
| 36 |
-
"epoch": 0.
|
| 37 |
-
"learning_rate": 4.
|
| 38 |
-
"loss": 2.
|
| 39 |
"step": 50
|
| 40 |
},
|
| 41 |
{
|
| 42 |
-
"epoch": 0.
|
| 43 |
-
"learning_rate":
|
| 44 |
-
"loss":
|
| 45 |
"step": 60
|
| 46 |
},
|
| 47 |
{
|
| 48 |
-
"epoch": 0.
|
| 49 |
-
"learning_rate": 4.
|
| 50 |
-
"loss": 1.
|
| 51 |
"step": 70
|
| 52 |
},
|
| 53 |
{
|
| 54 |
-
"epoch": 0.
|
| 55 |
-
"eval_accuracy": 0.
|
| 56 |
-
"eval_loss": 1.
|
| 57 |
-
"eval_runtime":
|
| 58 |
-
"eval_samples_per_second":
|
| 59 |
-
"eval_steps_per_second":
|
| 60 |
"step": 75
|
| 61 |
},
|
| 62 |
{
|
| 63 |
-
"epoch": 1.
|
| 64 |
-
"learning_rate": 4.
|
| 65 |
-
"loss": 1.
|
| 66 |
"step": 80
|
| 67 |
},
|
| 68 |
{
|
| 69 |
-
"epoch": 1.
|
| 70 |
-
"learning_rate":
|
| 71 |
-
"loss": 1.
|
| 72 |
"step": 90
|
| 73 |
},
|
| 74 |
{
|
| 75 |
-
"epoch": 1.
|
| 76 |
-
"learning_rate":
|
| 77 |
-
"loss": 1.
|
| 78 |
"step": 100
|
| 79 |
},
|
| 80 |
{
|
| 81 |
-
"epoch": 1.
|
| 82 |
-
"learning_rate":
|
| 83 |
-
"loss": 0.
|
| 84 |
"step": 110
|
| 85 |
},
|
| 86 |
{
|
| 87 |
-
"epoch": 1.
|
| 88 |
-
"learning_rate":
|
| 89 |
-
"loss": 0.
|
| 90 |
"step": 120
|
| 91 |
},
|
| 92 |
{
|
| 93 |
-
"epoch": 1.
|
| 94 |
-
"learning_rate":
|
| 95 |
-
"loss": 0.
|
| 96 |
"step": 130
|
| 97 |
},
|
| 98 |
{
|
| 99 |
-
"epoch": 1.
|
| 100 |
-
"learning_rate":
|
| 101 |
-
"loss": 0.
|
| 102 |
"step": 140
|
| 103 |
},
|
| 104 |
{
|
| 105 |
-
"epoch": 1.
|
| 106 |
-
"learning_rate":
|
| 107 |
-
"loss": 0.
|
| 108 |
"step": 150
|
| 109 |
},
|
| 110 |
{
|
| 111 |
-
"epoch": 1.
|
| 112 |
-
"eval_accuracy": 0.
|
| 113 |
-
"eval_loss": 0.
|
| 114 |
-
"eval_runtime":
|
| 115 |
-
"eval_samples_per_second":
|
| 116 |
-
"eval_steps_per_second":
|
| 117 |
"step": 150
|
| 118 |
},
|
| 119 |
{
|
| 120 |
-
"epoch": 2.
|
| 121 |
-
"learning_rate":
|
| 122 |
-
"loss": 0.
|
| 123 |
"step": 160
|
| 124 |
},
|
| 125 |
{
|
| 126 |
-
"epoch": 2.
|
| 127 |
-
"learning_rate":
|
| 128 |
-
"loss": 0.
|
| 129 |
"step": 170
|
| 130 |
},
|
| 131 |
{
|
| 132 |
-
"epoch": 2.
|
| 133 |
-
"learning_rate":
|
| 134 |
-
"loss": 0.
|
| 135 |
"step": 180
|
| 136 |
},
|
| 137 |
{
|
| 138 |
-
"epoch": 2.
|
| 139 |
-
"learning_rate":
|
| 140 |
-
"loss": 0.
|
| 141 |
"step": 190
|
| 142 |
},
|
| 143 |
{
|
| 144 |
-
"epoch": 2.
|
| 145 |
-
"learning_rate":
|
| 146 |
-
"loss": 0.
|
| 147 |
"step": 200
|
| 148 |
},
|
| 149 |
{
|
| 150 |
-
"epoch": 2.
|
| 151 |
-
"learning_rate":
|
| 152 |
-
"loss": 0.
|
| 153 |
"step": 210
|
| 154 |
},
|
| 155 |
{
|
| 156 |
-
"epoch": 2.
|
| 157 |
-
"learning_rate":
|
| 158 |
-
"loss": 0.
|
| 159 |
"step": 220
|
| 160 |
},
|
| 161 |
{
|
| 162 |
-
"epoch": 2.
|
| 163 |
-
"eval_accuracy": 0.
|
| 164 |
-
"eval_loss": 0.
|
| 165 |
-
"eval_runtime":
|
| 166 |
-
"eval_samples_per_second":
|
| 167 |
-
"eval_steps_per_second":
|
| 168 |
"step": 225
|
| 169 |
},
|
| 170 |
{
|
| 171 |
-
"epoch": 3.
|
| 172 |
-
"learning_rate":
|
| 173 |
-
"loss": 0.
|
| 174 |
"step": 230
|
| 175 |
},
|
| 176 |
{
|
| 177 |
-
"epoch": 3.
|
| 178 |
-
"learning_rate":
|
| 179 |
-
"loss": 0.
|
| 180 |
"step": 240
|
| 181 |
},
|
| 182 |
{
|
| 183 |
-
"epoch": 3.
|
| 184 |
-
"learning_rate":
|
| 185 |
-
"loss": 0.
|
| 186 |
"step": 250
|
| 187 |
},
|
| 188 |
{
|
| 189 |
-
"epoch": 3.
|
| 190 |
-
"learning_rate":
|
| 191 |
-
"loss": 0.
|
| 192 |
"step": 260
|
| 193 |
},
|
| 194 |
{
|
| 195 |
-
"epoch": 3.
|
| 196 |
-
"learning_rate":
|
| 197 |
-
"loss": 0.
|
| 198 |
"step": 270
|
| 199 |
},
|
| 200 |
{
|
| 201 |
-
"epoch": 3.
|
| 202 |
-
"learning_rate":
|
| 203 |
-
"loss": 0.
|
| 204 |
"step": 280
|
| 205 |
},
|
| 206 |
{
|
| 207 |
-
"epoch": 3.
|
| 208 |
-
"learning_rate":
|
| 209 |
-
"loss": 0.
|
| 210 |
"step": 290
|
| 211 |
},
|
| 212 |
{
|
| 213 |
-
"epoch": 3.
|
| 214 |
-
"learning_rate":
|
| 215 |
-
"loss": 0.
|
| 216 |
"step": 300
|
| 217 |
},
|
| 218 |
{
|
| 219 |
-
"epoch": 3.
|
| 220 |
-
"eval_accuracy": 0.
|
| 221 |
-
"eval_loss": 0.
|
| 222 |
-
"eval_runtime":
|
| 223 |
-
"eval_samples_per_second":
|
| 224 |
-
"eval_steps_per_second":
|
| 225 |
"step": 300
|
| 226 |
},
|
| 227 |
{
|
| 228 |
-
"epoch":
|
| 229 |
-
"
|
| 230 |
-
"
|
| 231 |
-
"
|
| 232 |
-
"train_runtime": 866.4529,
|
| 233 |
-
"train_samples_per_second": 1.385,
|
| 234 |
-
"train_steps_per_second": 0.346
|
| 235 |
},
|
| 236 |
{
|
| 237 |
-
"epoch":
|
| 238 |
-
"
|
| 239 |
-
"
|
| 240 |
-
"
|
| 241 |
-
"eval_samples_per_second": 3.353,
|
| 242 |
-
"eval_steps_per_second": 0.844,
|
| 243 |
-
"step": 300
|
| 244 |
},
|
| 245 |
{
|
| 246 |
-
"epoch":
|
| 247 |
-
"
|
| 248 |
-
"
|
| 249 |
-
"
|
| 250 |
-
|
| 251 |
-
|
| 252 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 253 |
}
|
| 254 |
],
|
| 255 |
"logging_steps": 10,
|
| 256 |
-
"max_steps":
|
|
|
|
| 257 |
"num_train_epochs": 9223372036854775807,
|
| 258 |
"save_steps": 500,
|
| 259 |
-
"total_flos":
|
|
|
|
| 260 |
"trial_name": null,
|
| 261 |
"trial_params": null
|
| 262 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_metric": 0.9857142857142858,
|
| 3 |
+
"best_model_checkpoint": "videomae-base-finetuned-ucf101-subset/checkpoint-525",
|
| 4 |
+
"epoch": 7.125,
|
| 5 |
"eval_steps": 500,
|
| 6 |
+
"global_step": 600,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
+
"epoch": 0.02,
|
| 13 |
+
"learning_rate": 8.333333333333334e-06,
|
| 14 |
+
"loss": 2.4415,
|
| 15 |
"step": 10
|
| 16 |
},
|
| 17 |
{
|
| 18 |
+
"epoch": 0.03,
|
| 19 |
+
"learning_rate": 1.6666666666666667e-05,
|
| 20 |
+
"loss": 2.2673,
|
| 21 |
"step": 20
|
| 22 |
},
|
| 23 |
{
|
| 24 |
+
"epoch": 0.05,
|
| 25 |
+
"learning_rate": 2.5e-05,
|
| 26 |
+
"loss": 2.2863,
|
| 27 |
"step": 30
|
| 28 |
},
|
| 29 |
{
|
| 30 |
+
"epoch": 0.07,
|
| 31 |
+
"learning_rate": 3.3333333333333335e-05,
|
| 32 |
+
"loss": 2.1765,
|
| 33 |
"step": 40
|
| 34 |
},
|
| 35 |
{
|
| 36 |
+
"epoch": 0.08,
|
| 37 |
+
"learning_rate": 4.166666666666667e-05,
|
| 38 |
+
"loss": 2.2121,
|
| 39 |
"step": 50
|
| 40 |
},
|
| 41 |
{
|
| 42 |
+
"epoch": 0.1,
|
| 43 |
+
"learning_rate": 5e-05,
|
| 44 |
+
"loss": 2.0465,
|
| 45 |
"step": 60
|
| 46 |
},
|
| 47 |
{
|
| 48 |
+
"epoch": 0.12,
|
| 49 |
+
"learning_rate": 4.9074074074074075e-05,
|
| 50 |
+
"loss": 1.9543,
|
| 51 |
"step": 70
|
| 52 |
},
|
| 53 |
{
|
| 54 |
+
"epoch": 0.12,
|
| 55 |
+
"eval_accuracy": 0.6,
|
| 56 |
+
"eval_loss": 1.5145318508148193,
|
| 57 |
+
"eval_runtime": 7.6774,
|
| 58 |
+
"eval_samples_per_second": 9.118,
|
| 59 |
+
"eval_steps_per_second": 2.345,
|
| 60 |
"step": 75
|
| 61 |
},
|
| 62 |
{
|
| 63 |
+
"epoch": 1.01,
|
| 64 |
+
"learning_rate": 4.814814814814815e-05,
|
| 65 |
+
"loss": 1.4717,
|
| 66 |
"step": 80
|
| 67 |
},
|
| 68 |
{
|
| 69 |
+
"epoch": 1.02,
|
| 70 |
+
"learning_rate": 4.722222222222222e-05,
|
| 71 |
+
"loss": 1.2937,
|
| 72 |
"step": 90
|
| 73 |
},
|
| 74 |
{
|
| 75 |
+
"epoch": 1.04,
|
| 76 |
+
"learning_rate": 4.62962962962963e-05,
|
| 77 |
+
"loss": 1.0696,
|
| 78 |
"step": 100
|
| 79 |
},
|
| 80 |
{
|
| 81 |
+
"epoch": 1.06,
|
| 82 |
+
"learning_rate": 4.5370370370370374e-05,
|
| 83 |
+
"loss": 0.8328,
|
| 84 |
"step": 110
|
| 85 |
},
|
| 86 |
{
|
| 87 |
+
"epoch": 1.07,
|
| 88 |
+
"learning_rate": 4.4444444444444447e-05,
|
| 89 |
+
"loss": 0.7714,
|
| 90 |
"step": 120
|
| 91 |
},
|
| 92 |
{
|
| 93 |
+
"epoch": 1.09,
|
| 94 |
+
"learning_rate": 4.351851851851852e-05,
|
| 95 |
+
"loss": 0.6604,
|
| 96 |
"step": 130
|
| 97 |
},
|
| 98 |
{
|
| 99 |
+
"epoch": 1.11,
|
| 100 |
+
"learning_rate": 4.259259259259259e-05,
|
| 101 |
+
"loss": 0.4873,
|
| 102 |
"step": 140
|
| 103 |
},
|
| 104 |
{
|
| 105 |
+
"epoch": 1.12,
|
| 106 |
+
"learning_rate": 4.166666666666667e-05,
|
| 107 |
+
"loss": 0.7175,
|
| 108 |
"step": 150
|
| 109 |
},
|
| 110 |
{
|
| 111 |
+
"epoch": 1.12,
|
| 112 |
+
"eval_accuracy": 0.6571428571428571,
|
| 113 |
+
"eval_loss": 0.9734774827957153,
|
| 114 |
+
"eval_runtime": 7.7823,
|
| 115 |
+
"eval_samples_per_second": 8.995,
|
| 116 |
+
"eval_steps_per_second": 2.313,
|
| 117 |
"step": 150
|
| 118 |
},
|
| 119 |
{
|
| 120 |
+
"epoch": 2.02,
|
| 121 |
+
"learning_rate": 4.074074074074074e-05,
|
| 122 |
+
"loss": 0.5271,
|
| 123 |
"step": 160
|
| 124 |
},
|
| 125 |
{
|
| 126 |
+
"epoch": 2.03,
|
| 127 |
+
"learning_rate": 3.981481481481482e-05,
|
| 128 |
+
"loss": 0.5743,
|
| 129 |
"step": 170
|
| 130 |
},
|
| 131 |
{
|
| 132 |
+
"epoch": 2.05,
|
| 133 |
+
"learning_rate": 3.888888888888889e-05,
|
| 134 |
+
"loss": 0.3496,
|
| 135 |
"step": 180
|
| 136 |
},
|
| 137 |
{
|
| 138 |
+
"epoch": 2.07,
|
| 139 |
+
"learning_rate": 3.7962962962962964e-05,
|
| 140 |
+
"loss": 0.379,
|
| 141 |
"step": 190
|
| 142 |
},
|
| 143 |
{
|
| 144 |
+
"epoch": 2.08,
|
| 145 |
+
"learning_rate": 3.7037037037037037e-05,
|
| 146 |
+
"loss": 0.4612,
|
| 147 |
"step": 200
|
| 148 |
},
|
| 149 |
{
|
| 150 |
+
"epoch": 2.1,
|
| 151 |
+
"learning_rate": 3.611111111111111e-05,
|
| 152 |
+
"loss": 0.4579,
|
| 153 |
"step": 210
|
| 154 |
},
|
| 155 |
{
|
| 156 |
+
"epoch": 2.12,
|
| 157 |
+
"learning_rate": 3.518518518518519e-05,
|
| 158 |
+
"loss": 0.2914,
|
| 159 |
"step": 220
|
| 160 |
},
|
| 161 |
{
|
| 162 |
+
"epoch": 2.12,
|
| 163 |
+
"eval_accuracy": 0.8,
|
| 164 |
+
"eval_loss": 0.47639983892440796,
|
| 165 |
+
"eval_runtime": 8.028,
|
| 166 |
+
"eval_samples_per_second": 8.719,
|
| 167 |
+
"eval_steps_per_second": 2.242,
|
| 168 |
"step": 225
|
| 169 |
},
|
| 170 |
{
|
| 171 |
+
"epoch": 3.01,
|
| 172 |
+
"learning_rate": 3.425925925925926e-05,
|
| 173 |
+
"loss": 0.2305,
|
| 174 |
"step": 230
|
| 175 |
},
|
| 176 |
{
|
| 177 |
+
"epoch": 3.02,
|
| 178 |
+
"learning_rate": 3.3333333333333335e-05,
|
| 179 |
+
"loss": 0.1102,
|
| 180 |
"step": 240
|
| 181 |
},
|
| 182 |
{
|
| 183 |
+
"epoch": 3.04,
|
| 184 |
+
"learning_rate": 3.240740740740741e-05,
|
| 185 |
+
"loss": 0.0558,
|
| 186 |
"step": 250
|
| 187 |
},
|
| 188 |
{
|
| 189 |
+
"epoch": 3.06,
|
| 190 |
+
"learning_rate": 3.148148148148148e-05,
|
| 191 |
+
"loss": 0.4813,
|
| 192 |
"step": 260
|
| 193 |
},
|
| 194 |
{
|
| 195 |
+
"epoch": 3.08,
|
| 196 |
+
"learning_rate": 3.055555555555556e-05,
|
| 197 |
+
"loss": 0.4584,
|
| 198 |
"step": 270
|
| 199 |
},
|
| 200 |
{
|
| 201 |
+
"epoch": 3.09,
|
| 202 |
+
"learning_rate": 2.962962962962963e-05,
|
| 203 |
+
"loss": 0.288,
|
| 204 |
"step": 280
|
| 205 |
},
|
| 206 |
{
|
| 207 |
+
"epoch": 3.11,
|
| 208 |
+
"learning_rate": 2.8703703703703706e-05,
|
| 209 |
+
"loss": 0.4266,
|
| 210 |
"step": 290
|
| 211 |
},
|
| 212 |
{
|
| 213 |
+
"epoch": 3.12,
|
| 214 |
+
"learning_rate": 2.777777777777778e-05,
|
| 215 |
+
"loss": 0.54,
|
| 216 |
"step": 300
|
| 217 |
},
|
| 218 |
{
|
| 219 |
+
"epoch": 3.12,
|
| 220 |
+
"eval_accuracy": 0.8,
|
| 221 |
+
"eval_loss": 0.6035991907119751,
|
| 222 |
+
"eval_runtime": 7.9026,
|
| 223 |
+
"eval_samples_per_second": 8.858,
|
| 224 |
+
"eval_steps_per_second": 2.278,
|
| 225 |
"step": 300
|
| 226 |
},
|
| 227 |
{
|
| 228 |
+
"epoch": 4.02,
|
| 229 |
+
"learning_rate": 2.6851851851851855e-05,
|
| 230 |
+
"loss": 0.3826,
|
| 231 |
+
"step": 310
|
|
|
|
|
|
|
|
|
|
| 232 |
},
|
| 233 |
{
|
| 234 |
+
"epoch": 4.03,
|
| 235 |
+
"learning_rate": 2.5925925925925925e-05,
|
| 236 |
+
"loss": 0.4745,
|
| 237 |
+
"step": 320
|
|
|
|
|
|
|
|
|
|
| 238 |
},
|
| 239 |
{
|
| 240 |
+
"epoch": 4.05,
|
| 241 |
+
"learning_rate": 2.5e-05,
|
| 242 |
+
"loss": 0.2059,
|
| 243 |
+
"step": 330
|
| 244 |
+
},
|
| 245 |
+
{
|
| 246 |
+
"epoch": 4.07,
|
| 247 |
+
"learning_rate": 2.4074074074074074e-05,
|
| 248 |
+
"loss": 0.1123,
|
| 249 |
+
"step": 340
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 4.08,
|
| 253 |
+
"learning_rate": 2.314814814814815e-05,
|
| 254 |
+
"loss": 0.0443,
|
| 255 |
+
"step": 350
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 4.1,
|
| 259 |
+
"learning_rate": 2.2222222222222223e-05,
|
| 260 |
+
"loss": 0.1167,
|
| 261 |
+
"step": 360
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 4.12,
|
| 265 |
+
"learning_rate": 2.1296296296296296e-05,
|
| 266 |
+
"loss": 0.1074,
|
| 267 |
+
"step": 370
|
| 268 |
+
},
|
| 269 |
+
{
|
| 270 |
+
"epoch": 4.12,
|
| 271 |
+
"eval_accuracy": 0.9285714285714286,
|
| 272 |
+
"eval_loss": 0.19949281215667725,
|
| 273 |
+
"eval_runtime": 7.9455,
|
| 274 |
+
"eval_samples_per_second": 8.81,
|
| 275 |
+
"eval_steps_per_second": 2.265,
|
| 276 |
+
"step": 375
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 5.01,
|
| 280 |
+
"learning_rate": 2.037037037037037e-05,
|
| 281 |
+
"loss": 0.1591,
|
| 282 |
+
"step": 380
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 5.03,
|
| 286 |
+
"learning_rate": 1.9444444444444445e-05,
|
| 287 |
+
"loss": 0.1139,
|
| 288 |
+
"step": 390
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"epoch": 5.04,
|
| 292 |
+
"learning_rate": 1.8518518518518518e-05,
|
| 293 |
+
"loss": 0.028,
|
| 294 |
+
"step": 400
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"epoch": 5.06,
|
| 298 |
+
"learning_rate": 1.7592592592592595e-05,
|
| 299 |
+
"loss": 0.0256,
|
| 300 |
+
"step": 410
|
| 301 |
+
},
|
| 302 |
+
{
|
| 303 |
+
"epoch": 5.08,
|
| 304 |
+
"learning_rate": 1.6666666666666667e-05,
|
| 305 |
+
"loss": 0.0579,
|
| 306 |
+
"step": 420
|
| 307 |
+
},
|
| 308 |
+
{
|
| 309 |
+
"epoch": 5.09,
|
| 310 |
+
"learning_rate": 1.574074074074074e-05,
|
| 311 |
+
"loss": 0.0093,
|
| 312 |
+
"step": 430
|
| 313 |
+
},
|
| 314 |
+
{
|
| 315 |
+
"epoch": 5.11,
|
| 316 |
+
"learning_rate": 1.4814814814814815e-05,
|
| 317 |
+
"loss": 0.0308,
|
| 318 |
+
"step": 440
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 5.12,
|
| 322 |
+
"learning_rate": 1.388888888888889e-05,
|
| 323 |
+
"loss": 0.0087,
|
| 324 |
+
"step": 450
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 5.12,
|
| 328 |
+
"eval_accuracy": 0.9714285714285714,
|
| 329 |
+
"eval_loss": 0.12557251751422882,
|
| 330 |
+
"eval_runtime": 7.9052,
|
| 331 |
+
"eval_samples_per_second": 8.855,
|
| 332 |
+
"eval_steps_per_second": 2.277,
|
| 333 |
+
"step": 450
|
| 334 |
+
},
|
| 335 |
+
{
|
| 336 |
+
"epoch": 6.02,
|
| 337 |
+
"learning_rate": 1.2962962962962962e-05,
|
| 338 |
+
"loss": 0.1549,
|
| 339 |
+
"step": 460
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 6.03,
|
| 343 |
+
"learning_rate": 1.2037037037037037e-05,
|
| 344 |
+
"loss": 0.0059,
|
| 345 |
+
"step": 470
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 6.05,
|
| 349 |
+
"learning_rate": 1.1111111111111112e-05,
|
| 350 |
+
"loss": 0.0332,
|
| 351 |
+
"step": 480
|
| 352 |
+
},
|
| 353 |
+
{
|
| 354 |
+
"epoch": 6.07,
|
| 355 |
+
"learning_rate": 1.0185185185185185e-05,
|
| 356 |
+
"loss": 0.0143,
|
| 357 |
+
"step": 490
|
| 358 |
+
},
|
| 359 |
+
{
|
| 360 |
+
"epoch": 6.08,
|
| 361 |
+
"learning_rate": 9.259259259259259e-06,
|
| 362 |
+
"loss": 0.0056,
|
| 363 |
+
"step": 500
|
| 364 |
+
},
|
| 365 |
+
{
|
| 366 |
+
"epoch": 6.1,
|
| 367 |
+
"learning_rate": 8.333333333333334e-06,
|
| 368 |
+
"loss": 0.0063,
|
| 369 |
+
"step": 510
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 6.12,
|
| 373 |
+
"learning_rate": 7.4074074074074075e-06,
|
| 374 |
+
"loss": 0.0061,
|
| 375 |
+
"step": 520
|
| 376 |
+
},
|
| 377 |
+
{
|
| 378 |
+
"epoch": 6.12,
|
| 379 |
+
"eval_accuracy": 0.9857142857142858,
|
| 380 |
+
"eval_loss": 0.04664895310997963,
|
| 381 |
+
"eval_runtime": 8.079,
|
| 382 |
+
"eval_samples_per_second": 8.664,
|
| 383 |
+
"eval_steps_per_second": 2.228,
|
| 384 |
+
"step": 525
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 7.01,
|
| 388 |
+
"learning_rate": 6.481481481481481e-06,
|
| 389 |
+
"loss": 0.005,
|
| 390 |
+
"step": 530
|
| 391 |
+
},
|
| 392 |
+
{
|
| 393 |
+
"epoch": 7.03,
|
| 394 |
+
"learning_rate": 5.555555555555556e-06,
|
| 395 |
+
"loss": 0.0056,
|
| 396 |
+
"step": 540
|
| 397 |
+
},
|
| 398 |
+
{
|
| 399 |
+
"epoch": 7.04,
|
| 400 |
+
"learning_rate": 4.6296296296296296e-06,
|
| 401 |
+
"loss": 0.0048,
|
| 402 |
+
"step": 550
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 7.06,
|
| 406 |
+
"learning_rate": 3.7037037037037037e-06,
|
| 407 |
+
"loss": 0.0101,
|
| 408 |
+
"step": 560
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 7.08,
|
| 412 |
+
"learning_rate": 2.777777777777778e-06,
|
| 413 |
+
"loss": 0.0046,
|
| 414 |
+
"step": 570
|
| 415 |
+
},
|
| 416 |
+
{
|
| 417 |
+
"epoch": 7.09,
|
| 418 |
+
"learning_rate": 1.8518518518518519e-06,
|
| 419 |
+
"loss": 0.0098,
|
| 420 |
+
"step": 580
|
| 421 |
+
},
|
| 422 |
+
{
|
| 423 |
+
"epoch": 7.11,
|
| 424 |
+
"learning_rate": 9.259259259259259e-07,
|
| 425 |
+
"loss": 0.011,
|
| 426 |
+
"step": 590
|
| 427 |
+
},
|
| 428 |
+
{
|
| 429 |
+
"epoch": 7.12,
|
| 430 |
+
"learning_rate": 0.0,
|
| 431 |
+
"loss": 0.0082,
|
| 432 |
+
"step": 600
|
| 433 |
+
},
|
| 434 |
+
{
|
| 435 |
+
"epoch": 7.12,
|
| 436 |
+
"eval_accuracy": 0.9857142857142858,
|
| 437 |
+
"eval_loss": 0.03631773963570595,
|
| 438 |
+
"eval_runtime": 7.8675,
|
| 439 |
+
"eval_samples_per_second": 8.897,
|
| 440 |
+
"eval_steps_per_second": 2.288,
|
| 441 |
+
"step": 600
|
| 442 |
+
},
|
| 443 |
+
{
|
| 444 |
+
"epoch": 7.12,
|
| 445 |
+
"step": 600,
|
| 446 |
+
"total_flos": 2.990768376250368e+18,
|
| 447 |
+
"train_loss": 0.5080375221868356,
|
| 448 |
+
"train_runtime": 676.6462,
|
| 449 |
+
"train_samples_per_second": 3.547,
|
| 450 |
+
"train_steps_per_second": 0.887
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 7.12,
|
| 454 |
+
"eval_accuracy": 0.9483870967741935,
|
| 455 |
+
"eval_loss": 0.2060869336128235,
|
| 456 |
+
"eval_runtime": 16.6655,
|
| 457 |
+
"eval_samples_per_second": 9.301,
|
| 458 |
+
"eval_steps_per_second": 2.34,
|
| 459 |
+
"step": 600
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 7.12,
|
| 463 |
+
"eval_accuracy": 0.9483870967741935,
|
| 464 |
+
"eval_loss": 0.2060869336128235,
|
| 465 |
+
"eval_runtime": 16.8116,
|
| 466 |
+
"eval_samples_per_second": 9.22,
|
| 467 |
+
"eval_steps_per_second": 2.32,
|
| 468 |
+
"step": 600
|
| 469 |
}
|
| 470 |
],
|
| 471 |
"logging_steps": 10,
|
| 472 |
+
"max_steps": 600,
|
| 473 |
+
"num_input_tokens_seen": 0,
|
| 474 |
"num_train_epochs": 9223372036854775807,
|
| 475 |
"save_steps": 500,
|
| 476 |
+
"total_flos": 2.990768376250368e+18,
|
| 477 |
+
"train_batch_size": 4,
|
| 478 |
"trial_name": null,
|
| 479 |
"trial_params": null
|
| 480 |
}
|