End of training
Browse files
all_results.json
ADDED
|
@@ -0,0 +1,15 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"eval_exact_match": 29.28125,
|
| 4 |
+
"eval_f1": 41.842087703189854,
|
| 5 |
+
"eval_runtime": 31.9631,
|
| 6 |
+
"eval_samples": 3572,
|
| 7 |
+
"eval_samples_per_second": 111.754,
|
| 8 |
+
"eval_steps_per_second": 4.005,
|
| 9 |
+
"total_flos": 1.651520510954496e+16,
|
| 10 |
+
"train_loss": 2.5174192206804142,
|
| 11 |
+
"train_runtime": 2180.484,
|
| 12 |
+
"train_samples": 16854,
|
| 13 |
+
"train_samples_per_second": 77.295,
|
| 14 |
+
"train_steps_per_second": 2.761
|
| 15 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"eval_exact_match": 29.28125,
|
| 4 |
+
"eval_f1": 41.842087703189854,
|
| 5 |
+
"eval_runtime": 31.9631,
|
| 6 |
+
"eval_samples": 3572,
|
| 7 |
+
"eval_samples_per_second": 111.754,
|
| 8 |
+
"eval_steps_per_second": 4.005
|
| 9 |
+
}
|
runs/May12_06-53-46_ml-machine/events.out.tfevents.1715499103.ml-machine.17470.1
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:106a2b024a8fc2365bd577e6098c18898cf7b4d1982136797fd45b4204bee6c3
|
| 3 |
+
size 412
|
train_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"total_flos": 1.651520510954496e+16,
|
| 4 |
+
"train_loss": 2.5174192206804142,
|
| 5 |
+
"train_runtime": 2180.484,
|
| 6 |
+
"train_samples": 16854,
|
| 7 |
+
"train_samples_per_second": 77.295,
|
| 8 |
+
"train_steps_per_second": 2.761
|
| 9 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,280 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 41.842087703189854,
|
| 3 |
+
"best_model_checkpoint": "/root/turkic_qa/en_kaz_models/orig_kaz_roberta_base_model/checkpoint-6020",
|
| 4 |
+
"epoch": 10.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 6020,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 1.0,
|
| 13 |
+
"step": 602,
|
| 14 |
+
"train_exact_match": 4.595404595404595,
|
| 15 |
+
"train_f1": 11.363671882650538,
|
| 16 |
+
"train_runtime": 9.9586,
|
| 17 |
+
"train_samples_per_second": 112.967,
|
| 18 |
+
"train_steps_per_second": 4.117
|
| 19 |
+
},
|
| 20 |
+
{
|
| 21 |
+
"epoch": 1.0,
|
| 22 |
+
"grad_norm": 7.829487323760986,
|
| 23 |
+
"learning_rate": 5e-06,
|
| 24 |
+
"loss": 4.7559,
|
| 25 |
+
"step": 602
|
| 26 |
+
},
|
| 27 |
+
{
|
| 28 |
+
"epoch": 1.0,
|
| 29 |
+
"eval_exact_match": 4.0625,
|
| 30 |
+
"eval_f1": 10.65255335870508,
|
| 31 |
+
"eval_runtime": 31.5322,
|
| 32 |
+
"eval_samples_per_second": 113.281,
|
| 33 |
+
"eval_steps_per_second": 4.059,
|
| 34 |
+
"step": 602
|
| 35 |
+
},
|
| 36 |
+
{
|
| 37 |
+
"epoch": 2.0,
|
| 38 |
+
"step": 1204,
|
| 39 |
+
"train_exact_match": 16.083916083916083,
|
| 40 |
+
"train_f1": 24.66111862744578,
|
| 41 |
+
"train_runtime": 10.0503,
|
| 42 |
+
"train_samples_per_second": 113.131,
|
| 43 |
+
"train_steps_per_second": 4.079
|
| 44 |
+
},
|
| 45 |
+
{
|
| 46 |
+
"epoch": 2.0,
|
| 47 |
+
"grad_norm": 9.797904014587402,
|
| 48 |
+
"learning_rate": 1e-05,
|
| 49 |
+
"loss": 3.6465,
|
| 50 |
+
"step": 1204
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 2.0,
|
| 54 |
+
"eval_exact_match": 12.375,
|
| 55 |
+
"eval_f1": 19.998235228782818,
|
| 56 |
+
"eval_runtime": 31.6899,
|
| 57 |
+
"eval_samples_per_second": 112.717,
|
| 58 |
+
"eval_steps_per_second": 4.039,
|
| 59 |
+
"step": 1204
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 3.0,
|
| 63 |
+
"step": 1806,
|
| 64 |
+
"train_exact_match": 28.571428571428573,
|
| 65 |
+
"train_f1": 39.799779936991705,
|
| 66 |
+
"train_runtime": 9.9917,
|
| 67 |
+
"train_samples_per_second": 112.694,
|
| 68 |
+
"train_steps_per_second": 4.103
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 3.0,
|
| 72 |
+
"grad_norm": 15.731840133666992,
|
| 73 |
+
"learning_rate": 8.750000000000001e-06,
|
| 74 |
+
"loss": 3.0904,
|
| 75 |
+
"step": 1806
|
| 76 |
+
},
|
| 77 |
+
{
|
| 78 |
+
"epoch": 3.0,
|
| 79 |
+
"eval_exact_match": 20.8125,
|
| 80 |
+
"eval_f1": 31.83521011700413,
|
| 81 |
+
"eval_runtime": 31.6833,
|
| 82 |
+
"eval_samples_per_second": 112.741,
|
| 83 |
+
"eval_steps_per_second": 4.04,
|
| 84 |
+
"step": 1806
|
| 85 |
+
},
|
| 86 |
+
{
|
| 87 |
+
"epoch": 4.0,
|
| 88 |
+
"step": 2408,
|
| 89 |
+
"train_exact_match": 37.76223776223776,
|
| 90 |
+
"train_f1": 50.512835216096065,
|
| 91 |
+
"train_runtime": 10.2559,
|
| 92 |
+
"train_samples_per_second": 113.009,
|
| 93 |
+
"train_steps_per_second": 4.095
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 4.0,
|
| 97 |
+
"grad_norm": 15.223044395446777,
|
| 98 |
+
"learning_rate": 7.500000000000001e-06,
|
| 99 |
+
"loss": 2.5695,
|
| 100 |
+
"step": 2408
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 4.0,
|
| 104 |
+
"eval_exact_match": 26.34375,
|
| 105 |
+
"eval_f1": 37.68196751270749,
|
| 106 |
+
"eval_runtime": 31.6681,
|
| 107 |
+
"eval_samples_per_second": 112.795,
|
| 108 |
+
"eval_steps_per_second": 4.042,
|
| 109 |
+
"step": 2408
|
| 110 |
+
},
|
| 111 |
+
{
|
| 112 |
+
"epoch": 5.0,
|
| 113 |
+
"step": 3010,
|
| 114 |
+
"train_exact_match": 46.553446553446555,
|
| 115 |
+
"train_f1": 59.64402432927812,
|
| 116 |
+
"train_runtime": 9.9179,
|
| 117 |
+
"train_samples_per_second": 112.827,
|
| 118 |
+
"train_steps_per_second": 4.033
|
| 119 |
+
},
|
| 120 |
+
{
|
| 121 |
+
"epoch": 5.0,
|
| 122 |
+
"grad_norm": 19.780818939208984,
|
| 123 |
+
"learning_rate": 6.25e-06,
|
| 124 |
+
"loss": 2.2508,
|
| 125 |
+
"step": 3010
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"epoch": 5.0,
|
| 129 |
+
"eval_exact_match": 27.5625,
|
| 130 |
+
"eval_f1": 39.44221624780652,
|
| 131 |
+
"eval_runtime": 31.714,
|
| 132 |
+
"eval_samples_per_second": 112.632,
|
| 133 |
+
"eval_steps_per_second": 4.036,
|
| 134 |
+
"step": 3010
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 6.0,
|
| 138 |
+
"step": 3612,
|
| 139 |
+
"train_exact_match": 50.94905094905095,
|
| 140 |
+
"train_f1": 63.29384394642245,
|
| 141 |
+
"train_runtime": 9.8492,
|
| 142 |
+
"train_samples_per_second": 111.278,
|
| 143 |
+
"train_steps_per_second": 4.061
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 6.0,
|
| 147 |
+
"grad_norm": 19.713895797729492,
|
| 148 |
+
"learning_rate": 5e-06,
|
| 149 |
+
"loss": 2.0297,
|
| 150 |
+
"step": 3612
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 6.0,
|
| 154 |
+
"eval_exact_match": 28.75,
|
| 155 |
+
"eval_f1": 41.0194110809582,
|
| 156 |
+
"eval_runtime": 31.7278,
|
| 157 |
+
"eval_samples_per_second": 112.583,
|
| 158 |
+
"eval_steps_per_second": 4.034,
|
| 159 |
+
"step": 3612
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 7.0,
|
| 163 |
+
"step": 4214,
|
| 164 |
+
"train_exact_match": 53.74625374625375,
|
| 165 |
+
"train_f1": 65.89175767102272,
|
| 166 |
+
"train_runtime": 9.8541,
|
| 167 |
+
"train_samples_per_second": 113.658,
|
| 168 |
+
"train_steps_per_second": 4.059
|
| 169 |
+
},
|
| 170 |
+
{
|
| 171 |
+
"epoch": 7.0,
|
| 172 |
+
"grad_norm": 18.399381637573242,
|
| 173 |
+
"learning_rate": 3.7500000000000005e-06,
|
| 174 |
+
"loss": 1.8692,
|
| 175 |
+
"step": 4214
|
| 176 |
+
},
|
| 177 |
+
{
|
| 178 |
+
"epoch": 7.0,
|
| 179 |
+
"eval_exact_match": 29.0625,
|
| 180 |
+
"eval_f1": 41.37431545961761,
|
| 181 |
+
"eval_runtime": 31.718,
|
| 182 |
+
"eval_samples_per_second": 112.617,
|
| 183 |
+
"eval_steps_per_second": 4.036,
|
| 184 |
+
"step": 4214
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 8.0,
|
| 188 |
+
"step": 4816,
|
| 189 |
+
"train_exact_match": 57.04295704295704,
|
| 190 |
+
"train_f1": 69.23809998725315,
|
| 191 |
+
"train_runtime": 9.8266,
|
| 192 |
+
"train_samples_per_second": 113.06,
|
| 193 |
+
"train_steps_per_second": 4.071
|
| 194 |
+
},
|
| 195 |
+
{
|
| 196 |
+
"epoch": 8.0,
|
| 197 |
+
"grad_norm": 19.948137283325195,
|
| 198 |
+
"learning_rate": 2.5e-06,
|
| 199 |
+
"loss": 1.7415,
|
| 200 |
+
"step": 4816
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 8.0,
|
| 204 |
+
"eval_exact_match": 29.09375,
|
| 205 |
+
"eval_f1": 41.58572921874224,
|
| 206 |
+
"eval_runtime": 31.6379,
|
| 207 |
+
"eval_samples_per_second": 112.903,
|
| 208 |
+
"eval_steps_per_second": 4.046,
|
| 209 |
+
"step": 4816
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 9.0,
|
| 213 |
+
"step": 5418,
|
| 214 |
+
"train_exact_match": 59.14085914085914,
|
| 215 |
+
"train_f1": 70.72972336346831,
|
| 216 |
+
"train_runtime": 10.1,
|
| 217 |
+
"train_samples_per_second": 111.485,
|
| 218 |
+
"train_steps_per_second": 4.059
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 9.0,
|
| 222 |
+
"grad_norm": 18.038774490356445,
|
| 223 |
+
"learning_rate": 1.25e-06,
|
| 224 |
+
"loss": 1.6377,
|
| 225 |
+
"step": 5418
|
| 226 |
+
},
|
| 227 |
+
{
|
| 228 |
+
"epoch": 9.0,
|
| 229 |
+
"eval_exact_match": 29.21875,
|
| 230 |
+
"eval_f1": 41.68346527288109,
|
| 231 |
+
"eval_runtime": 31.8369,
|
| 232 |
+
"eval_samples_per_second": 112.197,
|
| 233 |
+
"eval_steps_per_second": 4.02,
|
| 234 |
+
"step": 5418
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 10.0,
|
| 238 |
+
"step": 6020,
|
| 239 |
+
"train_exact_match": 59.94005994005994,
|
| 240 |
+
"train_f1": 72.84928684458019,
|
| 241 |
+
"train_runtime": 9.7454,
|
| 242 |
+
"train_samples_per_second": 112.976,
|
| 243 |
+
"train_steps_per_second": 4.104
|
| 244 |
+
},
|
| 245 |
+
{
|
| 246 |
+
"epoch": 10.0,
|
| 247 |
+
"grad_norm": 21.71363639831543,
|
| 248 |
+
"learning_rate": 0.0,
|
| 249 |
+
"loss": 1.583,
|
| 250 |
+
"step": 6020
|
| 251 |
+
},
|
| 252 |
+
{
|
| 253 |
+
"epoch": 10.0,
|
| 254 |
+
"eval_exact_match": 29.28125,
|
| 255 |
+
"eval_f1": 41.842087703189854,
|
| 256 |
+
"eval_runtime": 31.896,
|
| 257 |
+
"eval_samples_per_second": 111.989,
|
| 258 |
+
"eval_steps_per_second": 4.013,
|
| 259 |
+
"step": 6020
|
| 260 |
+
},
|
| 261 |
+
{
|
| 262 |
+
"epoch": 10.0,
|
| 263 |
+
"step": 6020,
|
| 264 |
+
"total_flos": 1.651520510954496e+16,
|
| 265 |
+
"train_loss": 2.5174192206804142,
|
| 266 |
+
"train_runtime": 2180.484,
|
| 267 |
+
"train_samples_per_second": 77.295,
|
| 268 |
+
"train_steps_per_second": 2.761
|
| 269 |
+
}
|
| 270 |
+
],
|
| 271 |
+
"logging_steps": 500,
|
| 272 |
+
"max_steps": 6020,
|
| 273 |
+
"num_input_tokens_seen": 0,
|
| 274 |
+
"num_train_epochs": 10,
|
| 275 |
+
"save_steps": 500,
|
| 276 |
+
"total_flos": 1.651520510954496e+16,
|
| 277 |
+
"train_batch_size": 28,
|
| 278 |
+
"trial_name": null,
|
| 279 |
+
"trial_params": null
|
| 280 |
+
}
|