Commit
•
4463110
1
Parent(s):
6e09e0d
End of training
Browse files- all_results.json +12 -0
- eval_results.json +8 -0
- train_results.json +7 -0
- trainer_state.json +262 -0
all_results.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 5.0,
|
| 3 |
+
"eval_accuracy": 0.9699248120300752,
|
| 4 |
+
"eval_loss": 0.13279759883880615,
|
| 5 |
+
"eval_runtime": 1.71,
|
| 6 |
+
"eval_samples_per_second": 77.778,
|
| 7 |
+
"eval_steps_per_second": 5.263,
|
| 8 |
+
"train_loss": 0.3078647000973041,
|
| 9 |
+
"train_runtime": 120.5221,
|
| 10 |
+
"train_samples_per_second": 42.897,
|
| 11 |
+
"train_steps_per_second": 2.697
|
| 12 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 5.0,
|
| 3 |
+
"eval_accuracy": 0.9699248120300752,
|
| 4 |
+
"eval_loss": 0.13279759883880615,
|
| 5 |
+
"eval_runtime": 1.71,
|
| 6 |
+
"eval_samples_per_second": 77.778,
|
| 7 |
+
"eval_steps_per_second": 5.263
|
| 8 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 5.0,
|
| 3 |
+
"train_loss": 0.3078647000973041,
|
| 4 |
+
"train_runtime": 120.5221,
|
| 5 |
+
"train_samples_per_second": 42.897,
|
| 6 |
+
"train_steps_per_second": 2.697
|
| 7 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,262 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.13279759883880615,
|
| 3 |
+
"best_model_checkpoint": "./beans_outputs/checkpoint-325",
|
| 4 |
+
"epoch": 5.0,
|
| 5 |
+
"global_step": 325,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.15,
|
| 12 |
+
"learning_rate": 1.9384615384615386e-05,
|
| 13 |
+
"loss": 1.0095,
|
| 14 |
+
"step": 10
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.31,
|
| 18 |
+
"learning_rate": 1.876923076923077e-05,
|
| 19 |
+
"loss": 0.8982,
|
| 20 |
+
"step": 20
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.46,
|
| 24 |
+
"learning_rate": 1.8153846153846155e-05,
|
| 25 |
+
"loss": 0.7749,
|
| 26 |
+
"step": 30
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 0.62,
|
| 30 |
+
"learning_rate": 1.753846153846154e-05,
|
| 31 |
+
"loss": 0.6896,
|
| 32 |
+
"step": 40
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.77,
|
| 36 |
+
"learning_rate": 1.6923076923076924e-05,
|
| 37 |
+
"loss": 0.5892,
|
| 38 |
+
"step": 50
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.92,
|
| 42 |
+
"learning_rate": 1.630769230769231e-05,
|
| 43 |
+
"loss": 0.49,
|
| 44 |
+
"step": 60
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 1.0,
|
| 48 |
+
"eval_accuracy": 0.9624060150375939,
|
| 49 |
+
"eval_loss": 0.405004620552063,
|
| 50 |
+
"eval_runtime": 1.5758,
|
| 51 |
+
"eval_samples_per_second": 84.401,
|
| 52 |
+
"eval_steps_per_second": 5.711,
|
| 53 |
+
"step": 65
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 1.08,
|
| 57 |
+
"learning_rate": 1.5692307692307693e-05,
|
| 58 |
+
"loss": 0.4349,
|
| 59 |
+
"step": 70
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 1.23,
|
| 63 |
+
"learning_rate": 1.5076923076923078e-05,
|
| 64 |
+
"loss": 0.3709,
|
| 65 |
+
"step": 80
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 1.38,
|
| 69 |
+
"learning_rate": 1.4461538461538462e-05,
|
| 70 |
+
"loss": 0.3631,
|
| 71 |
+
"step": 90
|
| 72 |
+
},
|
| 73 |
+
{
|
| 74 |
+
"epoch": 1.54,
|
| 75 |
+
"learning_rate": 1.3846153846153847e-05,
|
| 76 |
+
"loss": 0.2787,
|
| 77 |
+
"step": 100
|
| 78 |
+
},
|
| 79 |
+
{
|
| 80 |
+
"epoch": 1.69,
|
| 81 |
+
"learning_rate": 1.3230769230769231e-05,
|
| 82 |
+
"loss": 0.2455,
|
| 83 |
+
"step": 110
|
| 84 |
+
},
|
| 85 |
+
{
|
| 86 |
+
"epoch": 1.85,
|
| 87 |
+
"learning_rate": 1.2615384615384616e-05,
|
| 88 |
+
"loss": 0.2542,
|
| 89 |
+
"step": 120
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 2.0,
|
| 93 |
+
"learning_rate": 1.2e-05,
|
| 94 |
+
"loss": 0.2769,
|
| 95 |
+
"step": 130
|
| 96 |
+
},
|
| 97 |
+
{
|
| 98 |
+
"epoch": 2.0,
|
| 99 |
+
"eval_accuracy": 0.9849624060150376,
|
| 100 |
+
"eval_loss": 0.1862308382987976,
|
| 101 |
+
"eval_runtime": 1.554,
|
| 102 |
+
"eval_samples_per_second": 85.586,
|
| 103 |
+
"eval_steps_per_second": 5.792,
|
| 104 |
+
"step": 130
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 2.15,
|
| 108 |
+
"learning_rate": 1.1384615384615385e-05,
|
| 109 |
+
"loss": 0.2338,
|
| 110 |
+
"step": 140
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 2.31,
|
| 114 |
+
"learning_rate": 1.076923076923077e-05,
|
| 115 |
+
"loss": 0.2129,
|
| 116 |
+
"step": 150
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 2.46,
|
| 120 |
+
"learning_rate": 1.0153846153846154e-05,
|
| 121 |
+
"loss": 0.1908,
|
| 122 |
+
"step": 160
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 2.62,
|
| 126 |
+
"learning_rate": 9.53846153846154e-06,
|
| 127 |
+
"loss": 0.1731,
|
| 128 |
+
"step": 170
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 2.77,
|
| 132 |
+
"learning_rate": 8.923076923076925e-06,
|
| 133 |
+
"loss": 0.1826,
|
| 134 |
+
"step": 180
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 2.92,
|
| 138 |
+
"learning_rate": 8.307692307692309e-06,
|
| 139 |
+
"loss": 0.1441,
|
| 140 |
+
"step": 190
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 3.0,
|
| 144 |
+
"eval_accuracy": 0.9774436090225563,
|
| 145 |
+
"eval_loss": 0.15537278354167938,
|
| 146 |
+
"eval_runtime": 1.5401,
|
| 147 |
+
"eval_samples_per_second": 86.36,
|
| 148 |
+
"eval_steps_per_second": 5.844,
|
| 149 |
+
"step": 195
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 3.08,
|
| 153 |
+
"learning_rate": 7.692307692307694e-06,
|
| 154 |
+
"loss": 0.1676,
|
| 155 |
+
"step": 200
|
| 156 |
+
},
|
| 157 |
+
{
|
| 158 |
+
"epoch": 3.23,
|
| 159 |
+
"learning_rate": 7.076923076923078e-06,
|
| 160 |
+
"loss": 0.1733,
|
| 161 |
+
"step": 210
|
| 162 |
+
},
|
| 163 |
+
{
|
| 164 |
+
"epoch": 3.38,
|
| 165 |
+
"learning_rate": 6.461538461538463e-06,
|
| 166 |
+
"loss": 0.1785,
|
| 167 |
+
"step": 220
|
| 168 |
+
},
|
| 169 |
+
{
|
| 170 |
+
"epoch": 3.54,
|
| 171 |
+
"learning_rate": 5.846153846153847e-06,
|
| 172 |
+
"loss": 0.1697,
|
| 173 |
+
"step": 230
|
| 174 |
+
},
|
| 175 |
+
{
|
| 176 |
+
"epoch": 3.69,
|
| 177 |
+
"learning_rate": 5.230769230769232e-06,
|
| 178 |
+
"loss": 0.1536,
|
| 179 |
+
"step": 240
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 3.85,
|
| 183 |
+
"learning_rate": 4.615384615384616e-06,
|
| 184 |
+
"loss": 0.1862,
|
| 185 |
+
"step": 250
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 4.0,
|
| 189 |
+
"learning_rate": 4.000000000000001e-06,
|
| 190 |
+
"loss": 0.1661,
|
| 191 |
+
"step": 260
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 4.0,
|
| 195 |
+
"eval_accuracy": 0.9774436090225563,
|
| 196 |
+
"eval_loss": 0.13332900404930115,
|
| 197 |
+
"eval_runtime": 1.5476,
|
| 198 |
+
"eval_samples_per_second": 85.942,
|
| 199 |
+
"eval_steps_per_second": 5.816,
|
| 200 |
+
"step": 260
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 4.15,
|
| 204 |
+
"learning_rate": 3.384615384615385e-06,
|
| 205 |
+
"loss": 0.1189,
|
| 206 |
+
"step": 270
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 4.31,
|
| 210 |
+
"learning_rate": 2.7692307692307697e-06,
|
| 211 |
+
"loss": 0.1368,
|
| 212 |
+
"step": 280
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 4.46,
|
| 216 |
+
"learning_rate": 2.153846153846154e-06,
|
| 217 |
+
"loss": 0.1742,
|
| 218 |
+
"step": 290
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 4.62,
|
| 222 |
+
"learning_rate": 1.5384615384615387e-06,
|
| 223 |
+
"loss": 0.1238,
|
| 224 |
+
"step": 300
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 4.77,
|
| 228 |
+
"learning_rate": 9.230769230769232e-07,
|
| 229 |
+
"loss": 0.1637,
|
| 230 |
+
"step": 310
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 4.92,
|
| 234 |
+
"learning_rate": 3.0769230769230774e-07,
|
| 235 |
+
"loss": 0.1754,
|
| 236 |
+
"step": 320
|
| 237 |
+
},
|
| 238 |
+
{
|
| 239 |
+
"epoch": 5.0,
|
| 240 |
+
"eval_accuracy": 0.9699248120300752,
|
| 241 |
+
"eval_loss": 0.13279759883880615,
|
| 242 |
+
"eval_runtime": 1.5399,
|
| 243 |
+
"eval_samples_per_second": 86.369,
|
| 244 |
+
"eval_steps_per_second": 5.845,
|
| 245 |
+
"step": 325
|
| 246 |
+
},
|
| 247 |
+
{
|
| 248 |
+
"epoch": 5.0,
|
| 249 |
+
"step": 325,
|
| 250 |
+
"total_flos": 4.006371770595533e+17,
|
| 251 |
+
"train_loss": 0.3078647000973041,
|
| 252 |
+
"train_runtime": 120.5221,
|
| 253 |
+
"train_samples_per_second": 42.897,
|
| 254 |
+
"train_steps_per_second": 2.697
|
| 255 |
+
}
|
| 256 |
+
],
|
| 257 |
+
"max_steps": 325,
|
| 258 |
+
"num_train_epochs": 5,
|
| 259 |
+
"total_flos": 4.006371770595533e+17,
|
| 260 |
+
"trial_name": null,
|
| 261 |
+
"trial_params": null
|
| 262 |
+
}
|