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": 99.98,
|
| 3 |
+
"eval_exact_match": 0.719910514541387,
|
| 4 |
+
"eval_loss": 0.11308582127094269,
|
| 5 |
+
"eval_runtime": 220.7134,
|
| 6 |
+
"eval_samples": 2235,
|
| 7 |
+
"eval_samples_per_second": 10.126,
|
| 8 |
+
"eval_steps_per_second": 1.269,
|
| 9 |
+
"predict_exact_match": 0.7193342453260374,
|
| 10 |
+
"predict_loss": 0.09834922105073929,
|
| 11 |
+
"predict_runtime": 438.0967,
|
| 12 |
+
"predict_samples": 4386,
|
| 13 |
+
"predict_samples_per_second": 10.011,
|
| 14 |
+
"predict_steps_per_second": 1.253,
|
| 15 |
+
"train_loss": 0.2523062037229538,
|
| 16 |
+
"train_runtime": 33274.6395,
|
| 17 |
+
"train_samples": 15667,
|
| 18 |
+
"train_samples_per_second": 46.161,
|
| 19 |
+
"train_steps_per_second": 0.09
|
| 20 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 99.98,
|
| 3 |
+
"eval_exact_match": 0.719910514541387,
|
| 4 |
+
"eval_loss": 0.11308582127094269,
|
| 5 |
+
"eval_runtime": 220.7134,
|
| 6 |
+
"eval_samples": 2235,
|
| 7 |
+
"eval_samples_per_second": 10.126,
|
| 8 |
+
"eval_steps_per_second": 1.269
|
| 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": 0.7193342453260374,
|
| 3 |
+
"predict_loss": 0.09834922105073929,
|
| 4 |
+
"predict_runtime": 438.0967,
|
| 5 |
+
"predict_samples": 4386,
|
| 6 |
+
"predict_samples_per_second": 10.011,
|
| 7 |
+
"predict_steps_per_second": 1.253
|
| 8 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 99.98,
|
| 3 |
+
"train_loss": 0.2523062037229538,
|
| 4 |
+
"train_runtime": 33274.6395,
|
| 5 |
+
"train_samples": 15667,
|
| 6 |
+
"train_samples_per_second": 46.161,
|
| 7 |
+
"train_steps_per_second": 0.09
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,250 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.11308582127094269,
|
| 3 |
+
"best_model_checkpoint": "/data/wheld3/mt5-base-pointer-mtop/checkpoint-1200",
|
| 4 |
+
"epoch": 99.9800918836141,
|
| 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": 6.65,
|
| 12 |
+
"learning_rate": 0.0009333333333333333,
|
| 13 |
+
"loss": 1.7749,
|
| 14 |
+
"step": 200
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 6.65,
|
| 18 |
+
"eval_exact_match": 0.003131991051454139,
|
| 19 |
+
"eval_loss": 0.5892038345336914,
|
| 20 |
+
"eval_runtime": 135.5401,
|
| 21 |
+
"eval_samples_per_second": 16.49,
|
| 22 |
+
"eval_steps_per_second": 2.066,
|
| 23 |
+
"step": 200
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 13.33,
|
| 27 |
+
"learning_rate": 0.0008666666666666667,
|
| 28 |
+
"loss": 0.6021,
|
| 29 |
+
"step": 400
|
| 30 |
+
},
|
| 31 |
+
{
|
| 32 |
+
"epoch": 13.33,
|
| 33 |
+
"eval_exact_match": 0.013870246085011185,
|
| 34 |
+
"eval_loss": 0.5159956216812134,
|
| 35 |
+
"eval_runtime": 130.613,
|
| 36 |
+
"eval_samples_per_second": 17.112,
|
| 37 |
+
"eval_steps_per_second": 2.144,
|
| 38 |
+
"step": 400
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 19.98,
|
| 42 |
+
"learning_rate": 0.0008,
|
| 43 |
+
"loss": 0.6044,
|
| 44 |
+
"step": 600
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 19.98,
|
| 48 |
+
"eval_exact_match": 0.053243847874720356,
|
| 49 |
+
"eval_loss": 0.40801870822906494,
|
| 50 |
+
"eval_runtime": 132.1423,
|
| 51 |
+
"eval_samples_per_second": 16.914,
|
| 52 |
+
"eval_steps_per_second": 2.119,
|
| 53 |
+
"step": 600
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 26.65,
|
| 57 |
+
"learning_rate": 0.0007333333333333333,
|
| 58 |
+
"loss": 0.3302,
|
| 59 |
+
"step": 800
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 26.65,
|
| 63 |
+
"eval_exact_match": 0.36196868008948546,
|
| 64 |
+
"eval_loss": 0.18654391169548035,
|
| 65 |
+
"eval_runtime": 134.276,
|
| 66 |
+
"eval_samples_per_second": 16.645,
|
| 67 |
+
"eval_steps_per_second": 2.085,
|
| 68 |
+
"step": 800
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 33.33,
|
| 72 |
+
"learning_rate": 0.0006666666666666666,
|
| 73 |
+
"loss": 0.1483,
|
| 74 |
+
"step": 1000
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 33.33,
|
| 78 |
+
"eval_exact_match": 0.5105145413870246,
|
| 79 |
+
"eval_loss": 0.1267053484916687,
|
| 80 |
+
"eval_runtime": 133.0912,
|
| 81 |
+
"eval_samples_per_second": 16.793,
|
| 82 |
+
"eval_steps_per_second": 2.104,
|
| 83 |
+
"step": 1000
|
| 84 |
+
},
|
| 85 |
+
{
|
| 86 |
+
"epoch": 39.98,
|
| 87 |
+
"learning_rate": 0.0006,
|
| 88 |
+
"loss": 0.0768,
|
| 89 |
+
"step": 1200
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 39.98,
|
| 93 |
+
"eval_exact_match": 0.5297539149888143,
|
| 94 |
+
"eval_loss": 0.11308582127094269,
|
| 95 |
+
"eval_runtime": 139.8147,
|
| 96 |
+
"eval_samples_per_second": 15.985,
|
| 97 |
+
"eval_steps_per_second": 2.003,
|
| 98 |
+
"step": 1200
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 46.65,
|
| 102 |
+
"learning_rate": 0.0005333333333333334,
|
| 103 |
+
"loss": 0.0525,
|
| 104 |
+
"step": 1400
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 46.65,
|
| 108 |
+
"eval_exact_match": 0.5413870246085011,
|
| 109 |
+
"eval_loss": 0.12185565382242203,
|
| 110 |
+
"eval_runtime": 132.0637,
|
| 111 |
+
"eval_samples_per_second": 16.924,
|
| 112 |
+
"eval_steps_per_second": 2.12,
|
| 113 |
+
"step": 1400
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"epoch": 53.33,
|
| 117 |
+
"learning_rate": 0.00046666666666666666,
|
| 118 |
+
"loss": 0.0801,
|
| 119 |
+
"step": 1600
|
| 120 |
+
},
|
| 121 |
+
{
|
| 122 |
+
"epoch": 53.33,
|
| 123 |
+
"eval_exact_match": 0.5275167785234899,
|
| 124 |
+
"eval_loss": 0.11860152333974838,
|
| 125 |
+
"eval_runtime": 132.256,
|
| 126 |
+
"eval_samples_per_second": 16.899,
|
| 127 |
+
"eval_steps_per_second": 2.117,
|
| 128 |
+
"step": 1600
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 59.98,
|
| 132 |
+
"learning_rate": 0.0004,
|
| 133 |
+
"loss": 0.0331,
|
| 134 |
+
"step": 1800
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 59.98,
|
| 138 |
+
"eval_exact_match": 0.5422818791946309,
|
| 139 |
+
"eval_loss": 0.13056021928787231,
|
| 140 |
+
"eval_runtime": 132.9523,
|
| 141 |
+
"eval_samples_per_second": 16.811,
|
| 142 |
+
"eval_steps_per_second": 2.106,
|
| 143 |
+
"step": 1800
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 66.65,
|
| 147 |
+
"learning_rate": 0.0003333333333333333,
|
| 148 |
+
"loss": 0.0254,
|
| 149 |
+
"step": 2000
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 66.65,
|
| 153 |
+
"eval_exact_match": 0.5395973154362416,
|
| 154 |
+
"eval_loss": 0.13960428535938263,
|
| 155 |
+
"eval_runtime": 135.7759,
|
| 156 |
+
"eval_samples_per_second": 16.461,
|
| 157 |
+
"eval_steps_per_second": 2.062,
|
| 158 |
+
"step": 2000
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 73.33,
|
| 162 |
+
"learning_rate": 0.0002666666666666667,
|
| 163 |
+
"loss": 0.0168,
|
| 164 |
+
"step": 2200
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 73.33,
|
| 168 |
+
"eval_exact_match": 0.5436241610738255,
|
| 169 |
+
"eval_loss": 0.15595464408397675,
|
| 170 |
+
"eval_runtime": 134.0818,
|
| 171 |
+
"eval_samples_per_second": 16.669,
|
| 172 |
+
"eval_steps_per_second": 2.088,
|
| 173 |
+
"step": 2200
|
| 174 |
+
},
|
| 175 |
+
{
|
| 176 |
+
"epoch": 79.98,
|
| 177 |
+
"learning_rate": 0.0002,
|
| 178 |
+
"loss": 0.0129,
|
| 179 |
+
"step": 2400
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 79.98,
|
| 183 |
+
"eval_exact_match": 0.5494407158836689,
|
| 184 |
+
"eval_loss": 0.16592496633529663,
|
| 185 |
+
"eval_runtime": 133.0433,
|
| 186 |
+
"eval_samples_per_second": 16.799,
|
| 187 |
+
"eval_steps_per_second": 2.105,
|
| 188 |
+
"step": 2400
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 86.65,
|
| 192 |
+
"learning_rate": 0.00013333333333333334,
|
| 193 |
+
"loss": 0.0105,
|
| 194 |
+
"step": 2600
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 86.65,
|
| 198 |
+
"eval_exact_match": 0.5422818791946309,
|
| 199 |
+
"eval_loss": 0.16985595226287842,
|
| 200 |
+
"eval_runtime": 137.112,
|
| 201 |
+
"eval_samples_per_second": 16.301,
|
| 202 |
+
"eval_steps_per_second": 2.042,
|
| 203 |
+
"step": 2600
|
| 204 |
+
},
|
| 205 |
+
{
|
| 206 |
+
"epoch": 93.33,
|
| 207 |
+
"learning_rate": 6.666666666666667e-05,
|
| 208 |
+
"loss": 0.0088,
|
| 209 |
+
"step": 2800
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 93.33,
|
| 213 |
+
"eval_exact_match": 0.5472035794183445,
|
| 214 |
+
"eval_loss": 0.17423103749752045,
|
| 215 |
+
"eval_runtime": 132.2768,
|
| 216 |
+
"eval_samples_per_second": 16.896,
|
| 217 |
+
"eval_steps_per_second": 2.117,
|
| 218 |
+
"step": 2800
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 99.98,
|
| 222 |
+
"learning_rate": 0.0,
|
| 223 |
+
"loss": 0.0077,
|
| 224 |
+
"step": 3000
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 99.98,
|
| 228 |
+
"eval_exact_match": 0.5467561521252796,
|
| 229 |
+
"eval_loss": 0.17754317820072174,
|
| 230 |
+
"eval_runtime": 132.953,
|
| 231 |
+
"eval_samples_per_second": 16.81,
|
| 232 |
+
"eval_steps_per_second": 2.106,
|
| 233 |
+
"step": 3000
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 99.98,
|
| 237 |
+
"step": 3000,
|
| 238 |
+
"total_flos": 7.72413437221586e+16,
|
| 239 |
+
"train_loss": 0.2523062037229538,
|
| 240 |
+
"train_runtime": 33274.6395,
|
| 241 |
+
"train_samples_per_second": 46.161,
|
| 242 |
+
"train_steps_per_second": 0.09
|
| 243 |
+
}
|
| 244 |
+
],
|
| 245 |
+
"max_steps": 3000,
|
| 246 |
+
"num_train_epochs": 100,
|
| 247 |
+
"total_flos": 7.72413437221586e+16,
|
| 248 |
+
"trial_name": null,
|
| 249 |
+
"trial_params": null
|
| 250 |
+
}
|