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": 0.774597495527728,
|
| 4 |
+
"eval_loss": 0.07756359130144119,
|
| 5 |
+
"eval_runtime": 81.3105,
|
| 6 |
+
"eval_samples": 559,
|
| 7 |
+
"eval_samples_per_second": 6.875,
|
| 8 |
+
"eval_steps_per_second": 0.861,
|
| 9 |
+
"predict_exact_match": 0.7754927163667523,
|
| 10 |
+
"predict_loss": 0.06615994870662689,
|
| 11 |
+
"predict_runtime": 125.5106,
|
| 12 |
+
"predict_samples": 1167,
|
| 13 |
+
"predict_samples_per_second": 9.298,
|
| 14 |
+
"predict_steps_per_second": 1.163,
|
| 15 |
+
"train_loss": 0.12990600236753622,
|
| 16 |
+
"train_runtime": 34726.3796,
|
| 17 |
+
"train_samples": 4077,
|
| 18 |
+
"train_samples_per_second": 44.232,
|
| 19 |
+
"train_steps_per_second": 0.086
|
| 20 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 428.5,
|
| 3 |
+
"eval_exact_match": 0.774597495527728,
|
| 4 |
+
"eval_loss": 0.07756359130144119,
|
| 5 |
+
"eval_runtime": 81.3105,
|
| 6 |
+
"eval_samples": 559,
|
| 7 |
+
"eval_samples_per_second": 6.875,
|
| 8 |
+
"eval_steps_per_second": 0.861
|
| 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": 0.7754927163667523,
|
| 3 |
+
"predict_loss": 0.06615994870662689,
|
| 4 |
+
"predict_runtime": 125.5106,
|
| 5 |
+
"predict_samples": 1167,
|
| 6 |
+
"predict_samples_per_second": 9.298,
|
| 7 |
+
"predict_steps_per_second": 1.163
|
| 8 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 428.5,
|
| 3 |
+
"train_loss": 0.12990600236753622,
|
| 4 |
+
"train_runtime": 34726.3796,
|
| 5 |
+
"train_samples": 4077,
|
| 6 |
+
"train_samples_per_second": 44.232,
|
| 7 |
+
"train_steps_per_second": 0.086
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,250 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.07756359130144119,
|
| 3 |
+
"best_model_checkpoint": "/data/wheld3/mt5-base-pointer-cstop_artificial/checkpoint-400",
|
| 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": 1.7482,
|
| 14 |
+
"step": 200
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 28.5,
|
| 18 |
+
"eval_exact_match": 0.10196779964221825,
|
| 19 |
+
"eval_loss": 0.2504979968070984,
|
| 20 |
+
"eval_runtime": 32.6293,
|
| 21 |
+
"eval_samples_per_second": 17.132,
|
| 22 |
+
"eval_steps_per_second": 2.145,
|
| 23 |
+
"step": 200
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 57.13,
|
| 27 |
+
"learning_rate": 0.0008666666666666667,
|
| 28 |
+
"loss": 0.1366,
|
| 29 |
+
"step": 400
|
| 30 |
+
},
|
| 31 |
+
{
|
| 32 |
+
"epoch": 57.13,
|
| 33 |
+
"eval_exact_match": 0.32379248658318427,
|
| 34 |
+
"eval_loss": 0.07756359130144119,
|
| 35 |
+
"eval_runtime": 32.4871,
|
| 36 |
+
"eval_samples_per_second": 17.207,
|
| 37 |
+
"eval_steps_per_second": 2.155,
|
| 38 |
+
"step": 400
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 85.63,
|
| 42 |
+
"learning_rate": 0.0008,
|
| 43 |
+
"loss": 0.0275,
|
| 44 |
+
"step": 600
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 85.63,
|
| 48 |
+
"eval_exact_match": 0.33810375670840787,
|
| 49 |
+
"eval_loss": 0.0881282389163971,
|
| 50 |
+
"eval_runtime": 32.966,
|
| 51 |
+
"eval_samples_per_second": 16.957,
|
| 52 |
+
"eval_steps_per_second": 2.123,
|
| 53 |
+
"step": 600
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 114.25,
|
| 57 |
+
"learning_rate": 0.0007333333333333333,
|
| 58 |
+
"loss": 0.0114,
|
| 59 |
+
"step": 800
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 114.25,
|
| 63 |
+
"eval_exact_match": 0.33989266547406083,
|
| 64 |
+
"eval_loss": 0.0990389883518219,
|
| 65 |
+
"eval_runtime": 33.8803,
|
| 66 |
+
"eval_samples_per_second": 16.499,
|
| 67 |
+
"eval_steps_per_second": 2.066,
|
| 68 |
+
"step": 800
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 142.75,
|
| 72 |
+
"learning_rate": 0.0006666666666666666,
|
| 73 |
+
"loss": 0.0064,
|
| 74 |
+
"step": 1000
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 142.75,
|
| 78 |
+
"eval_exact_match": 0.3416815742397138,
|
| 79 |
+
"eval_loss": 0.11197585612535477,
|
| 80 |
+
"eval_runtime": 32.7339,
|
| 81 |
+
"eval_samples_per_second": 17.077,
|
| 82 |
+
"eval_steps_per_second": 2.138,
|
| 83 |
+
"step": 1000
|
| 84 |
+
},
|
| 85 |
+
{
|
| 86 |
+
"epoch": 171.38,
|
| 87 |
+
"learning_rate": 0.0006,
|
| 88 |
+
"loss": 0.0045,
|
| 89 |
+
"step": 1200
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 171.38,
|
| 93 |
+
"eval_exact_match": 0.3434704830053667,
|
| 94 |
+
"eval_loss": 0.10809481889009476,
|
| 95 |
+
"eval_runtime": 32.4483,
|
| 96 |
+
"eval_samples_per_second": 17.227,
|
| 97 |
+
"eval_steps_per_second": 2.157,
|
| 98 |
+
"step": 1200
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 199.88,
|
| 102 |
+
"learning_rate": 0.0005333333333333334,
|
| 103 |
+
"loss": 0.0036,
|
| 104 |
+
"step": 1400
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 199.88,
|
| 108 |
+
"eval_exact_match": 0.3434704830053667,
|
| 109 |
+
"eval_loss": 0.12303601205348969,
|
| 110 |
+
"eval_runtime": 33.761,
|
| 111 |
+
"eval_samples_per_second": 16.558,
|
| 112 |
+
"eval_steps_per_second": 2.073,
|
| 113 |
+
"step": 1400
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"epoch": 228.5,
|
| 117 |
+
"learning_rate": 0.00046666666666666666,
|
| 118 |
+
"loss": 0.0025,
|
| 119 |
+
"step": 1600
|
| 120 |
+
},
|
| 121 |
+
{
|
| 122 |
+
"epoch": 228.5,
|
| 123 |
+
"eval_exact_match": 0.33989266547406083,
|
| 124 |
+
"eval_loss": 0.12110502272844315,
|
| 125 |
+
"eval_runtime": 39.4879,
|
| 126 |
+
"eval_samples_per_second": 14.156,
|
| 127 |
+
"eval_steps_per_second": 1.773,
|
| 128 |
+
"step": 1600
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 257.13,
|
| 132 |
+
"learning_rate": 0.0004,
|
| 133 |
+
"loss": 0.002,
|
| 134 |
+
"step": 1800
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 257.13,
|
| 138 |
+
"eval_exact_match": 0.33989266547406083,
|
| 139 |
+
"eval_loss": 0.13667185604572296,
|
| 140 |
+
"eval_runtime": 33.6552,
|
| 141 |
+
"eval_samples_per_second": 16.61,
|
| 142 |
+
"eval_steps_per_second": 2.08,
|
| 143 |
+
"step": 1800
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 285.63,
|
| 147 |
+
"learning_rate": 0.0003333333333333333,
|
| 148 |
+
"loss": 0.0016,
|
| 149 |
+
"step": 2000
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 285.63,
|
| 153 |
+
"eval_exact_match": 0.3434704830053667,
|
| 154 |
+
"eval_loss": 0.13241663575172424,
|
| 155 |
+
"eval_runtime": 33.4579,
|
| 156 |
+
"eval_samples_per_second": 16.708,
|
| 157 |
+
"eval_steps_per_second": 2.092,
|
| 158 |
+
"step": 2000
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 314.25,
|
| 162 |
+
"learning_rate": 0.0002666666666666667,
|
| 163 |
+
"loss": 0.0013,
|
| 164 |
+
"step": 2200
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 314.25,
|
| 168 |
+
"eval_exact_match": 0.3470483005366726,
|
| 169 |
+
"eval_loss": 0.13395807147026062,
|
| 170 |
+
"eval_runtime": 34.1281,
|
| 171 |
+
"eval_samples_per_second": 16.379,
|
| 172 |
+
"eval_steps_per_second": 2.051,
|
| 173 |
+
"step": 2200
|
| 174 |
+
},
|
| 175 |
+
{
|
| 176 |
+
"epoch": 342.75,
|
| 177 |
+
"learning_rate": 0.0002,
|
| 178 |
+
"loss": 0.001,
|
| 179 |
+
"step": 2400
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 342.75,
|
| 183 |
+
"eval_exact_match": 0.3434704830053667,
|
| 184 |
+
"eval_loss": 0.13736827671527863,
|
| 185 |
+
"eval_runtime": 33.0567,
|
| 186 |
+
"eval_samples_per_second": 16.91,
|
| 187 |
+
"eval_steps_per_second": 2.118,
|
| 188 |
+
"step": 2400
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 371.38,
|
| 192 |
+
"learning_rate": 0.00013333333333333334,
|
| 193 |
+
"loss": 0.0009,
|
| 194 |
+
"step": 2600
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 371.38,
|
| 198 |
+
"eval_exact_match": 0.3416815742397138,
|
| 199 |
+
"eval_loss": 0.13843561708927155,
|
| 200 |
+
"eval_runtime": 33.1328,
|
| 201 |
+
"eval_samples_per_second": 16.871,
|
| 202 |
+
"eval_steps_per_second": 2.113,
|
| 203 |
+
"step": 2600
|
| 204 |
+
},
|
| 205 |
+
{
|
| 206 |
+
"epoch": 399.88,
|
| 207 |
+
"learning_rate": 6.666666666666667e-05,
|
| 208 |
+
"loss": 0.0007,
|
| 209 |
+
"step": 2800
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 399.88,
|
| 213 |
+
"eval_exact_match": 0.3434704830053667,
|
| 214 |
+
"eval_loss": 0.1421527862548828,
|
| 215 |
+
"eval_runtime": 32.5079,
|
| 216 |
+
"eval_samples_per_second": 17.196,
|
| 217 |
+
"eval_steps_per_second": 2.153,
|
| 218 |
+
"step": 2800
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 428.5,
|
| 222 |
+
"learning_rate": 0.0,
|
| 223 |
+
"loss": 0.0006,
|
| 224 |
+
"step": 3000
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 428.5,
|
| 228 |
+
"eval_exact_match": 0.3416815742397138,
|
| 229 |
+
"eval_loss": 0.1451684683561325,
|
| 230 |
+
"eval_runtime": 32.385,
|
| 231 |
+
"eval_samples_per_second": 17.261,
|
| 232 |
+
"eval_steps_per_second": 2.161,
|
| 233 |
+
"step": 3000
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 428.5,
|
| 237 |
+
"step": 3000,
|
| 238 |
+
"total_flos": 7.330479471426355e+16,
|
| 239 |
+
"train_loss": 0.12990600236753622,
|
| 240 |
+
"train_runtime": 34726.3796,
|
| 241 |
+
"train_samples_per_second": 44.232,
|
| 242 |
+
"train_steps_per_second": 0.086
|
| 243 |
+
}
|
| 244 |
+
],
|
| 245 |
+
"max_steps": 3000,
|
| 246 |
+
"num_train_epochs": 429,
|
| 247 |
+
"total_flos": 7.330479471426355e+16,
|
| 248 |
+
"trial_name": null,
|
| 249 |
+
"trial_params": null
|
| 250 |
+
}
|