Commit
•
f1432e4
1
Parent(s):
de7c025
End of training
Browse files- all_results.json +8 -0
- test_results.json +8 -0
- trainer_state.json +276 -0
all_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.06,
|
| 3 |
+
"eval_accuracy": 0.6829268292682927,
|
| 4 |
+
"eval_loss": 0.6236140727996826,
|
| 5 |
+
"eval_runtime": 8.1467,
|
| 6 |
+
"eval_samples_per_second": 5.033,
|
| 7 |
+
"eval_steps_per_second": 1.35
|
| 8 |
+
}
|
test_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.06,
|
| 3 |
+
"eval_accuracy": 0.6829268292682927,
|
| 4 |
+
"eval_loss": 0.6236140727996826,
|
| 5 |
+
"eval_runtime": 8.1467,
|
| 6 |
+
"eval_samples_per_second": 5.033,
|
| 7 |
+
"eval_steps_per_second": 1.35
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,276 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.6829268292682927,
|
| 3 |
+
"best_model_checkpoint": "MAE-CT-CPC-Dicotomized-v4-early-stop/checkpoint-81",
|
| 4 |
+
"epoch": 3.05625,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 324,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.01,
|
| 13 |
+
"learning_rate": 6.944444444444446e-07,
|
| 14 |
+
"loss": 0.6989,
|
| 15 |
+
"step": 10
|
| 16 |
+
},
|
| 17 |
+
{
|
| 18 |
+
"epoch": 0.01,
|
| 19 |
+
"learning_rate": 1.3888888888888892e-06,
|
| 20 |
+
"loss": 0.6577,
|
| 21 |
+
"step": 20
|
| 22 |
+
},
|
| 23 |
+
{
|
| 24 |
+
"epoch": 0.02,
|
| 25 |
+
"learning_rate": 2.0833333333333334e-06,
|
| 26 |
+
"loss": 0.6586,
|
| 27 |
+
"step": 30
|
| 28 |
+
},
|
| 29 |
+
{
|
| 30 |
+
"epoch": 0.03,
|
| 31 |
+
"learning_rate": 2.7777777777777783e-06,
|
| 32 |
+
"loss": 0.5975,
|
| 33 |
+
"step": 40
|
| 34 |
+
},
|
| 35 |
+
{
|
| 36 |
+
"epoch": 0.03,
|
| 37 |
+
"learning_rate": 3.4722222222222224e-06,
|
| 38 |
+
"loss": 0.6342,
|
| 39 |
+
"step": 50
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 0.04,
|
| 43 |
+
"learning_rate": 4.166666666666667e-06,
|
| 44 |
+
"loss": 0.685,
|
| 45 |
+
"step": 60
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.05,
|
| 49 |
+
"learning_rate": 4.861111111111111e-06,
|
| 50 |
+
"loss": 0.5336,
|
| 51 |
+
"step": 70
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.06,
|
| 55 |
+
"learning_rate": 5.555555555555557e-06,
|
| 56 |
+
"loss": 0.5496,
|
| 57 |
+
"step": 80
|
| 58 |
+
},
|
| 59 |
+
{
|
| 60 |
+
"epoch": 0.06,
|
| 61 |
+
"eval_accuracy": 0.6829268292682927,
|
| 62 |
+
"eval_loss": 0.6272094249725342,
|
| 63 |
+
"eval_runtime": 8.9631,
|
| 64 |
+
"eval_samples_per_second": 4.574,
|
| 65 |
+
"eval_steps_per_second": 1.227,
|
| 66 |
+
"step": 81
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 1.01,
|
| 70 |
+
"learning_rate": 6.25e-06,
|
| 71 |
+
"loss": 0.4225,
|
| 72 |
+
"step": 90
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 1.01,
|
| 76 |
+
"learning_rate": 6.944444444444445e-06,
|
| 77 |
+
"loss": 0.8021,
|
| 78 |
+
"step": 100
|
| 79 |
+
},
|
| 80 |
+
{
|
| 81 |
+
"epoch": 1.02,
|
| 82 |
+
"learning_rate": 7.638888888888888e-06,
|
| 83 |
+
"loss": 0.5772,
|
| 84 |
+
"step": 110
|
| 85 |
+
},
|
| 86 |
+
{
|
| 87 |
+
"epoch": 1.03,
|
| 88 |
+
"learning_rate": 8.333333333333334e-06,
|
| 89 |
+
"loss": 0.667,
|
| 90 |
+
"step": 120
|
| 91 |
+
},
|
| 92 |
+
{
|
| 93 |
+
"epoch": 1.03,
|
| 94 |
+
"learning_rate": 9.027777777777779e-06,
|
| 95 |
+
"loss": 0.7481,
|
| 96 |
+
"step": 130
|
| 97 |
+
},
|
| 98 |
+
{
|
| 99 |
+
"epoch": 1.04,
|
| 100 |
+
"learning_rate": 9.722222222222223e-06,
|
| 101 |
+
"loss": 0.5948,
|
| 102 |
+
"step": 140
|
| 103 |
+
},
|
| 104 |
+
{
|
| 105 |
+
"epoch": 1.05,
|
| 106 |
+
"learning_rate": 9.953703703703704e-06,
|
| 107 |
+
"loss": 0.6092,
|
| 108 |
+
"step": 150
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 1.05,
|
| 112 |
+
"learning_rate": 9.876543209876543e-06,
|
| 113 |
+
"loss": 0.6089,
|
| 114 |
+
"step": 160
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 1.06,
|
| 118 |
+
"eval_accuracy": 0.6829268292682927,
|
| 119 |
+
"eval_loss": 0.596904993057251,
|
| 120 |
+
"eval_runtime": 8.0298,
|
| 121 |
+
"eval_samples_per_second": 5.106,
|
| 122 |
+
"eval_steps_per_second": 1.37,
|
| 123 |
+
"step": 162
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"epoch": 2.01,
|
| 127 |
+
"learning_rate": 9.799382716049384e-06,
|
| 128 |
+
"loss": 0.5514,
|
| 129 |
+
"step": 170
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 2.01,
|
| 133 |
+
"learning_rate": 9.722222222222223e-06,
|
| 134 |
+
"loss": 0.5377,
|
| 135 |
+
"step": 180
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 2.02,
|
| 139 |
+
"learning_rate": 9.645061728395062e-06,
|
| 140 |
+
"loss": 0.5121,
|
| 141 |
+
"step": 190
|
| 142 |
+
},
|
| 143 |
+
{
|
| 144 |
+
"epoch": 2.03,
|
| 145 |
+
"learning_rate": 9.567901234567902e-06,
|
| 146 |
+
"loss": 0.8356,
|
| 147 |
+
"step": 200
|
| 148 |
+
},
|
| 149 |
+
{
|
| 150 |
+
"epoch": 2.03,
|
| 151 |
+
"learning_rate": 9.490740740740741e-06,
|
| 152 |
+
"loss": 0.591,
|
| 153 |
+
"step": 210
|
| 154 |
+
},
|
| 155 |
+
{
|
| 156 |
+
"epoch": 2.04,
|
| 157 |
+
"learning_rate": 9.413580246913581e-06,
|
| 158 |
+
"loss": 0.4467,
|
| 159 |
+
"step": 220
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 2.05,
|
| 163 |
+
"learning_rate": 9.33641975308642e-06,
|
| 164 |
+
"loss": 0.5282,
|
| 165 |
+
"step": 230
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 2.05,
|
| 169 |
+
"learning_rate": 9.25925925925926e-06,
|
| 170 |
+
"loss": 0.715,
|
| 171 |
+
"step": 240
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 2.06,
|
| 175 |
+
"eval_accuracy": 0.6585365853658537,
|
| 176 |
+
"eval_loss": 0.6061528921127319,
|
| 177 |
+
"eval_runtime": 7.8186,
|
| 178 |
+
"eval_samples_per_second": 5.244,
|
| 179 |
+
"eval_steps_per_second": 1.407,
|
| 180 |
+
"step": 243
|
| 181 |
+
},
|
| 182 |
+
{
|
| 183 |
+
"epoch": 3.0,
|
| 184 |
+
"learning_rate": 9.1820987654321e-06,
|
| 185 |
+
"loss": 0.6485,
|
| 186 |
+
"step": 250
|
| 187 |
+
},
|
| 188 |
+
{
|
| 189 |
+
"epoch": 3.01,
|
| 190 |
+
"learning_rate": 9.10493827160494e-06,
|
| 191 |
+
"loss": 0.7628,
|
| 192 |
+
"step": 260
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 3.02,
|
| 196 |
+
"learning_rate": 9.027777777777779e-06,
|
| 197 |
+
"loss": 0.5779,
|
| 198 |
+
"step": 270
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 3.03,
|
| 202 |
+
"learning_rate": 8.950617283950618e-06,
|
| 203 |
+
"loss": 0.5771,
|
| 204 |
+
"step": 280
|
| 205 |
+
},
|
| 206 |
+
{
|
| 207 |
+
"epoch": 3.03,
|
| 208 |
+
"learning_rate": 8.873456790123458e-06,
|
| 209 |
+
"loss": 0.7393,
|
| 210 |
+
"step": 290
|
| 211 |
+
},
|
| 212 |
+
{
|
| 213 |
+
"epoch": 3.04,
|
| 214 |
+
"learning_rate": 8.796296296296297e-06,
|
| 215 |
+
"loss": 0.6242,
|
| 216 |
+
"step": 300
|
| 217 |
+
},
|
| 218 |
+
{
|
| 219 |
+
"epoch": 3.05,
|
| 220 |
+
"learning_rate": 8.719135802469136e-06,
|
| 221 |
+
"loss": 0.5647,
|
| 222 |
+
"step": 310
|
| 223 |
+
},
|
| 224 |
+
{
|
| 225 |
+
"epoch": 3.05,
|
| 226 |
+
"learning_rate": 8.641975308641975e-06,
|
| 227 |
+
"loss": 0.474,
|
| 228 |
+
"step": 320
|
| 229 |
+
},
|
| 230 |
+
{
|
| 231 |
+
"epoch": 3.06,
|
| 232 |
+
"eval_accuracy": 0.6829268292682927,
|
| 233 |
+
"eval_loss": 0.6187382936477661,
|
| 234 |
+
"eval_runtime": 7.9589,
|
| 235 |
+
"eval_samples_per_second": 5.151,
|
| 236 |
+
"eval_steps_per_second": 1.382,
|
| 237 |
+
"step": 324
|
| 238 |
+
},
|
| 239 |
+
{
|
| 240 |
+
"epoch": 3.06,
|
| 241 |
+
"step": 324,
|
| 242 |
+
"total_flos": 5.655641894102237e+18,
|
| 243 |
+
"train_loss": 0.6119254886368175,
|
| 244 |
+
"train_runtime": 569.9574,
|
| 245 |
+
"train_samples_per_second": 10.106,
|
| 246 |
+
"train_steps_per_second": 2.527
|
| 247 |
+
},
|
| 248 |
+
{
|
| 249 |
+
"epoch": 3.06,
|
| 250 |
+
"eval_accuracy": 0.6829268292682927,
|
| 251 |
+
"eval_loss": 0.6236140727996826,
|
| 252 |
+
"eval_runtime": 8.7888,
|
| 253 |
+
"eval_samples_per_second": 4.665,
|
| 254 |
+
"eval_steps_per_second": 1.252,
|
| 255 |
+
"step": 324
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 3.06,
|
| 259 |
+
"eval_accuracy": 0.6829268292682927,
|
| 260 |
+
"eval_loss": 0.6236140727996826,
|
| 261 |
+
"eval_runtime": 8.1467,
|
| 262 |
+
"eval_samples_per_second": 5.033,
|
| 263 |
+
"eval_steps_per_second": 1.35,
|
| 264 |
+
"step": 324
|
| 265 |
+
}
|
| 266 |
+
],
|
| 267 |
+
"logging_steps": 10,
|
| 268 |
+
"max_steps": 1440,
|
| 269 |
+
"num_input_tokens_seen": 0,
|
| 270 |
+
"num_train_epochs": 9223372036854775807,
|
| 271 |
+
"save_steps": 500,
|
| 272 |
+
"total_flos": 5.655641894102237e+18,
|
| 273 |
+
"train_batch_size": 4,
|
| 274 |
+
"trial_name": null,
|
| 275 |
+
"trial_params": null
|
| 276 |
+
}
|