End of training
Browse files- all_results.json +20 -0
- eval_results.json +9 -0
- generated_predictions.txt +0 -0
- predict_results.json +8 -0
- train_results.json +8 -0
- trainer_state.json +250 -0
all_results.json
ADDED
|
@@ -0,0 +1,20 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 428.5,
|
| 3 |
+
"eval_exact_match": 0.8050089445438283,
|
| 4 |
+
"eval_loss": 0.08161530643701553,
|
| 5 |
+
"eval_runtime": 18.4667,
|
| 6 |
+
"eval_samples": 559,
|
| 7 |
+
"eval_samples_per_second": 30.271,
|
| 8 |
+
"eval_steps_per_second": 1.895,
|
| 9 |
+
"predict_exact_match": 0.8003427592116538,
|
| 10 |
+
"predict_loss": 0.07206200808286667,
|
| 11 |
+
"predict_runtime": 38.0721,
|
| 12 |
+
"predict_samples": 1167,
|
| 13 |
+
"predict_samples_per_second": 30.652,
|
| 14 |
+
"predict_steps_per_second": 1.917,
|
| 15 |
+
"train_loss": 0.16957045908768972,
|
| 16 |
+
"train_runtime": 14253.2821,
|
| 17 |
+
"train_samples": 4077,
|
| 18 |
+
"train_samples_per_second": 107.765,
|
| 19 |
+
"train_steps_per_second": 0.21
|
| 20 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 428.5,
|
| 3 |
+
"eval_exact_match": 0.8050089445438283,
|
| 4 |
+
"eval_loss": 0.08161530643701553,
|
| 5 |
+
"eval_runtime": 18.4667,
|
| 6 |
+
"eval_samples": 559,
|
| 7 |
+
"eval_samples_per_second": 30.271,
|
| 8 |
+
"eval_steps_per_second": 1.895
|
| 9 |
+
}
|
generated_predictions.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
predict_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"predict_exact_match": 0.8003427592116538,
|
| 3 |
+
"predict_loss": 0.07206200808286667,
|
| 4 |
+
"predict_runtime": 38.0721,
|
| 5 |
+
"predict_samples": 1167,
|
| 6 |
+
"predict_samples_per_second": 30.652,
|
| 7 |
+
"predict_steps_per_second": 1.917
|
| 8 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 428.5,
|
| 3 |
+
"train_loss": 0.16957045908768972,
|
| 4 |
+
"train_runtime": 14253.2821,
|
| 5 |
+
"train_samples": 4077,
|
| 6 |
+
"train_samples_per_second": 107.765,
|
| 7 |
+
"train_steps_per_second": 0.21
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,250 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.08161530643701553,
|
| 3 |
+
"best_model_checkpoint": "/data/wheld3/mt5-small-pointer-cstop_artificial/checkpoint-800",
|
| 4 |
+
"epoch": 428.5019607843137,
|
| 5 |
+
"global_step": 3000,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 28.5,
|
| 12 |
+
"learning_rate": 0.0009333333333333333,
|
| 13 |
+
"loss": 2.08,
|
| 14 |
+
"step": 200
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 28.5,
|
| 18 |
+
"eval_exact_match": 0.03756708407871199,
|
| 19 |
+
"eval_loss": 0.331952303647995,
|
| 20 |
+
"eval_runtime": 10.6308,
|
| 21 |
+
"eval_samples_per_second": 52.583,
|
| 22 |
+
"eval_steps_per_second": 3.292,
|
| 23 |
+
"step": 200
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 57.13,
|
| 27 |
+
"learning_rate": 0.0008666666666666667,
|
| 28 |
+
"loss": 0.272,
|
| 29 |
+
"step": 400
|
| 30 |
+
},
|
| 31 |
+
{
|
| 32 |
+
"epoch": 57.13,
|
| 33 |
+
"eval_exact_match": 0.2629695885509839,
|
| 34 |
+
"eval_loss": 0.10836012661457062,
|
| 35 |
+
"eval_runtime": 10.9563,
|
| 36 |
+
"eval_samples_per_second": 51.021,
|
| 37 |
+
"eval_steps_per_second": 3.194,
|
| 38 |
+
"step": 400
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 85.63,
|
| 42 |
+
"learning_rate": 0.0008,
|
| 43 |
+
"loss": 0.0789,
|
| 44 |
+
"step": 600
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 85.63,
|
| 48 |
+
"eval_exact_match": 0.3184257602862254,
|
| 49 |
+
"eval_loss": 0.08299708366394043,
|
| 50 |
+
"eval_runtime": 11.1629,
|
| 51 |
+
"eval_samples_per_second": 50.077,
|
| 52 |
+
"eval_steps_per_second": 3.135,
|
| 53 |
+
"step": 600
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 114.25,
|
| 57 |
+
"learning_rate": 0.0007333333333333333,
|
| 58 |
+
"loss": 0.0355,
|
| 59 |
+
"step": 800
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 114.25,
|
| 63 |
+
"eval_exact_match": 0.3363148479427549,
|
| 64 |
+
"eval_loss": 0.08161530643701553,
|
| 65 |
+
"eval_runtime": 11.0139,
|
| 66 |
+
"eval_samples_per_second": 50.754,
|
| 67 |
+
"eval_steps_per_second": 3.178,
|
| 68 |
+
"step": 800
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 142.75,
|
| 72 |
+
"learning_rate": 0.0006666666666666666,
|
| 73 |
+
"loss": 0.0207,
|
| 74 |
+
"step": 1000
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 142.75,
|
| 78 |
+
"eval_exact_match": 0.3291592128801431,
|
| 79 |
+
"eval_loss": 0.08680771291255951,
|
| 80 |
+
"eval_runtime": 11.1063,
|
| 81 |
+
"eval_samples_per_second": 50.332,
|
| 82 |
+
"eval_steps_per_second": 3.151,
|
| 83 |
+
"step": 1000
|
| 84 |
+
},
|
| 85 |
+
{
|
| 86 |
+
"epoch": 171.38,
|
| 87 |
+
"learning_rate": 0.0006,
|
| 88 |
+
"loss": 0.014,
|
| 89 |
+
"step": 1200
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 171.38,
|
| 93 |
+
"eval_exact_match": 0.33989266547406083,
|
| 94 |
+
"eval_loss": 0.09518682211637497,
|
| 95 |
+
"eval_runtime": 10.9222,
|
| 96 |
+
"eval_samples_per_second": 51.18,
|
| 97 |
+
"eval_steps_per_second": 3.204,
|
| 98 |
+
"step": 1200
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 199.88,
|
| 102 |
+
"learning_rate": 0.0005333333333333334,
|
| 103 |
+
"loss": 0.0099,
|
| 104 |
+
"step": 1400
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 199.88,
|
| 108 |
+
"eval_exact_match": 0.33810375670840787,
|
| 109 |
+
"eval_loss": 0.1089245155453682,
|
| 110 |
+
"eval_runtime": 10.85,
|
| 111 |
+
"eval_samples_per_second": 51.521,
|
| 112 |
+
"eval_steps_per_second": 3.226,
|
| 113 |
+
"step": 1400
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"epoch": 228.5,
|
| 117 |
+
"learning_rate": 0.00046666666666666666,
|
| 118 |
+
"loss": 0.0076,
|
| 119 |
+
"step": 1600
|
| 120 |
+
},
|
| 121 |
+
{
|
| 122 |
+
"epoch": 228.5,
|
| 123 |
+
"eval_exact_match": 0.33810375670840787,
|
| 124 |
+
"eval_loss": 0.11035683006048203,
|
| 125 |
+
"eval_runtime": 11.2414,
|
| 126 |
+
"eval_samples_per_second": 49.727,
|
| 127 |
+
"eval_steps_per_second": 3.113,
|
| 128 |
+
"step": 1600
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 257.13,
|
| 132 |
+
"learning_rate": 0.0004,
|
| 133 |
+
"loss": 0.0057,
|
| 134 |
+
"step": 1800
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 257.13,
|
| 138 |
+
"eval_exact_match": 0.3291592128801431,
|
| 139 |
+
"eval_loss": 0.11526591330766678,
|
| 140 |
+
"eval_runtime": 10.9866,
|
| 141 |
+
"eval_samples_per_second": 50.88,
|
| 142 |
+
"eval_steps_per_second": 3.186,
|
| 143 |
+
"step": 1800
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 285.63,
|
| 147 |
+
"learning_rate": 0.0003333333333333333,
|
| 148 |
+
"loss": 0.0048,
|
| 149 |
+
"step": 2000
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 285.63,
|
| 153 |
+
"eval_exact_match": 0.33273703041144903,
|
| 154 |
+
"eval_loss": 0.11532965302467346,
|
| 155 |
+
"eval_runtime": 11.0349,
|
| 156 |
+
"eval_samples_per_second": 50.658,
|
| 157 |
+
"eval_steps_per_second": 3.172,
|
| 158 |
+
"step": 2000
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 314.25,
|
| 162 |
+
"learning_rate": 0.0002666666666666667,
|
| 163 |
+
"loss": 0.004,
|
| 164 |
+
"step": 2200
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 314.25,
|
| 168 |
+
"eval_exact_match": 0.3363148479427549,
|
| 169 |
+
"eval_loss": 0.1206129789352417,
|
| 170 |
+
"eval_runtime": 10.904,
|
| 171 |
+
"eval_samples_per_second": 51.266,
|
| 172 |
+
"eval_steps_per_second": 3.21,
|
| 173 |
+
"step": 2200
|
| 174 |
+
},
|
| 175 |
+
{
|
| 176 |
+
"epoch": 342.75,
|
| 177 |
+
"learning_rate": 0.0002,
|
| 178 |
+
"loss": 0.0032,
|
| 179 |
+
"step": 2400
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 342.75,
|
| 183 |
+
"eval_exact_match": 0.3363148479427549,
|
| 184 |
+
"eval_loss": 0.1229204535484314,
|
| 185 |
+
"eval_runtime": 10.9966,
|
| 186 |
+
"eval_samples_per_second": 50.834,
|
| 187 |
+
"eval_steps_per_second": 3.183,
|
| 188 |
+
"step": 2400
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 371.38,
|
| 192 |
+
"learning_rate": 0.00013333333333333334,
|
| 193 |
+
"loss": 0.0028,
|
| 194 |
+
"step": 2600
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 371.38,
|
| 198 |
+
"eval_exact_match": 0.33810375670840787,
|
| 199 |
+
"eval_loss": 0.12684644758701324,
|
| 200 |
+
"eval_runtime": 10.9769,
|
| 201 |
+
"eval_samples_per_second": 50.925,
|
| 202 |
+
"eval_steps_per_second": 3.189,
|
| 203 |
+
"step": 2600
|
| 204 |
+
},
|
| 205 |
+
{
|
| 206 |
+
"epoch": 399.88,
|
| 207 |
+
"learning_rate": 6.666666666666667e-05,
|
| 208 |
+
"loss": 0.0023,
|
| 209 |
+
"step": 2800
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 399.88,
|
| 213 |
+
"eval_exact_match": 0.33989266547406083,
|
| 214 |
+
"eval_loss": 0.12882176041603088,
|
| 215 |
+
"eval_runtime": 10.9618,
|
| 216 |
+
"eval_samples_per_second": 50.995,
|
| 217 |
+
"eval_steps_per_second": 3.193,
|
| 218 |
+
"step": 2800
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 428.5,
|
| 222 |
+
"learning_rate": 0.0,
|
| 223 |
+
"loss": 0.002,
|
| 224 |
+
"step": 3000
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 428.5,
|
| 228 |
+
"eval_exact_match": 0.33989266547406083,
|
| 229 |
+
"eval_loss": 0.12921586632728577,
|
| 230 |
+
"eval_runtime": 11.4552,
|
| 231 |
+
"eval_samples_per_second": 48.799,
|
| 232 |
+
"eval_steps_per_second": 3.055,
|
| 233 |
+
"step": 3000
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 428.5,
|
| 237 |
+
"step": 3000,
|
| 238 |
+
"total_flos": 3.4731986474409984e+16,
|
| 239 |
+
"train_loss": 0.16957045908768972,
|
| 240 |
+
"train_runtime": 14253.2821,
|
| 241 |
+
"train_samples_per_second": 107.765,
|
| 242 |
+
"train_steps_per_second": 0.21
|
| 243 |
+
}
|
| 244 |
+
],
|
| 245 |
+
"max_steps": 3000,
|
| 246 |
+
"num_train_epochs": 429,
|
| 247 |
+
"total_flos": 3.4731986474409984e+16,
|
| 248 |
+
"trial_name": null,
|
| 249 |
+
"trial_params": null
|
| 250 |
+
}
|