End of training
Browse files- all_results.json +15 -0
- eval_results.json +9 -0
- runs/May12_06-13-44_ml-machine/events.out.tfevents.1715496479.ml-machine.1179.1 +3 -0
- train_results.json +9 -0
- trainer_state.json +280 -0
all_results.json
ADDED
|
@@ -0,0 +1,15 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"eval_exact_match": 29.1875,
|
| 4 |
+
"eval_f1": 44.04121981988171,
|
| 5 |
+
"eval_runtime": 29.1159,
|
| 6 |
+
"eval_samples": 3261,
|
| 7 |
+
"eval_samples_per_second": 112.0,
|
| 8 |
+
"eval_steps_per_second": 4.018,
|
| 9 |
+
"total_flos": 1.487387554039296e+16,
|
| 10 |
+
"train_loss": 2.7391692272207355,
|
| 11 |
+
"train_runtime": 1969.257,
|
| 12 |
+
"train_samples": 15179,
|
| 13 |
+
"train_samples_per_second": 77.08,
|
| 14 |
+
"train_steps_per_second": 2.757
|
| 15 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"eval_exact_match": 29.1875,
|
| 4 |
+
"eval_f1": 44.04121981988171,
|
| 5 |
+
"eval_runtime": 29.1159,
|
| 6 |
+
"eval_samples": 3261,
|
| 7 |
+
"eval_samples_per_second": 112.0,
|
| 8 |
+
"eval_steps_per_second": 4.018
|
| 9 |
+
}
|
runs/May12_06-13-44_ml-machine/events.out.tfevents.1715496479.ml-machine.1179.1
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d3b96b4315f4abe6c595126a8048b709df63e6ba744d494839f19c81f4744997
|
| 3 |
+
size 412
|
train_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"total_flos": 1.487387554039296e+16,
|
| 4 |
+
"train_loss": 2.7391692272207355,
|
| 5 |
+
"train_runtime": 1969.257,
|
| 6 |
+
"train_samples": 15179,
|
| 7 |
+
"train_samples_per_second": 77.08,
|
| 8 |
+
"train_steps_per_second": 2.757
|
| 9 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,280 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 44.04121981988171,
|
| 3 |
+
"best_model_checkpoint": "/root/turkic_qa/ru_uzn_models/orig_uzn_roberta_base_model/checkpoint-5430",
|
| 4 |
+
"epoch": 10.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 5430,
|
| 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": 543,
|
| 14 |
+
"train_exact_match": 6.993006993006993,
|
| 15 |
+
"train_f1": 14.72956145700345,
|
| 16 |
+
"train_runtime": 8.9919,
|
| 17 |
+
"train_samples_per_second": 114.102,
|
| 18 |
+
"train_steps_per_second": 4.115
|
| 19 |
+
},
|
| 20 |
+
{
|
| 21 |
+
"epoch": 1.0,
|
| 22 |
+
"grad_norm": 22.987104415893555,
|
| 23 |
+
"learning_rate": 5e-06,
|
| 24 |
+
"loss": 4.7716,
|
| 25 |
+
"step": 543
|
| 26 |
+
},
|
| 27 |
+
{
|
| 28 |
+
"epoch": 1.0,
|
| 29 |
+
"eval_exact_match": 6.65625,
|
| 30 |
+
"eval_f1": 13.460548189578436,
|
| 31 |
+
"eval_runtime": 28.875,
|
| 32 |
+
"eval_samples_per_second": 112.935,
|
| 33 |
+
"eval_steps_per_second": 4.052,
|
| 34 |
+
"step": 543
|
| 35 |
+
},
|
| 36 |
+
{
|
| 37 |
+
"epoch": 2.0,
|
| 38 |
+
"step": 1086,
|
| 39 |
+
"train_exact_match": 12.987012987012987,
|
| 40 |
+
"train_f1": 21.035135926338754,
|
| 41 |
+
"train_runtime": 9.0115,
|
| 42 |
+
"train_samples_per_second": 112.523,
|
| 43 |
+
"train_steps_per_second": 4.106
|
| 44 |
+
},
|
| 45 |
+
{
|
| 46 |
+
"epoch": 2.0,
|
| 47 |
+
"grad_norm": 27.96239471435547,
|
| 48 |
+
"learning_rate": 1e-05,
|
| 49 |
+
"loss": 3.7695,
|
| 50 |
+
"step": 1086
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 2.0,
|
| 54 |
+
"eval_exact_match": 10.71875,
|
| 55 |
+
"eval_f1": 17.87426936207211,
|
| 56 |
+
"eval_runtime": 28.7945,
|
| 57 |
+
"eval_samples_per_second": 113.251,
|
| 58 |
+
"eval_steps_per_second": 4.063,
|
| 59 |
+
"step": 1086
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 3.0,
|
| 63 |
+
"step": 1629,
|
| 64 |
+
"train_exact_match": 25.174825174825173,
|
| 65 |
+
"train_f1": 34.8258899586999,
|
| 66 |
+
"train_runtime": 9.0183,
|
| 67 |
+
"train_samples_per_second": 113.214,
|
| 68 |
+
"train_steps_per_second": 4.103
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 3.0,
|
| 72 |
+
"grad_norm": 38.839942932128906,
|
| 73 |
+
"learning_rate": 8.750000000000001e-06,
|
| 74 |
+
"loss": 3.4391,
|
| 75 |
+
"step": 1629
|
| 76 |
+
},
|
| 77 |
+
{
|
| 78 |
+
"epoch": 3.0,
|
| 79 |
+
"eval_exact_match": 15.84375,
|
| 80 |
+
"eval_f1": 25.41741224618572,
|
| 81 |
+
"eval_runtime": 29.0024,
|
| 82 |
+
"eval_samples_per_second": 112.439,
|
| 83 |
+
"eval_steps_per_second": 4.034,
|
| 84 |
+
"step": 1629
|
| 85 |
+
},
|
| 86 |
+
{
|
| 87 |
+
"epoch": 4.0,
|
| 88 |
+
"step": 2172,
|
| 89 |
+
"train_exact_match": 29.97002997002997,
|
| 90 |
+
"train_f1": 41.494806937045226,
|
| 91 |
+
"train_runtime": 9.0706,
|
| 92 |
+
"train_samples_per_second": 112.671,
|
| 93 |
+
"train_steps_per_second": 4.079
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 4.0,
|
| 97 |
+
"grad_norm": 35.55900955200195,
|
| 98 |
+
"learning_rate": 7.500000000000001e-06,
|
| 99 |
+
"loss": 2.9975,
|
| 100 |
+
"step": 2172
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 4.0,
|
| 104 |
+
"eval_exact_match": 20.4375,
|
| 105 |
+
"eval_f1": 31.800313776639523,
|
| 106 |
+
"eval_runtime": 28.9344,
|
| 107 |
+
"eval_samples_per_second": 112.703,
|
| 108 |
+
"eval_steps_per_second": 4.044,
|
| 109 |
+
"step": 2172
|
| 110 |
+
},
|
| 111 |
+
{
|
| 112 |
+
"epoch": 5.0,
|
| 113 |
+
"step": 2715,
|
| 114 |
+
"train_exact_match": 38.86113886113886,
|
| 115 |
+
"train_f1": 51.45493720149751,
|
| 116 |
+
"train_runtime": 9.0591,
|
| 117 |
+
"train_samples_per_second": 112.483,
|
| 118 |
+
"train_steps_per_second": 4.084
|
| 119 |
+
},
|
| 120 |
+
{
|
| 121 |
+
"epoch": 5.0,
|
| 122 |
+
"grad_norm": 60.02823257446289,
|
| 123 |
+
"learning_rate": 6.25e-06,
|
| 124 |
+
"loss": 2.6357,
|
| 125 |
+
"step": 2715
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"epoch": 5.0,
|
| 129 |
+
"eval_exact_match": 23.65625,
|
| 130 |
+
"eval_f1": 36.81284091086545,
|
| 131 |
+
"eval_runtime": 28.8941,
|
| 132 |
+
"eval_samples_per_second": 112.861,
|
| 133 |
+
"eval_steps_per_second": 4.049,
|
| 134 |
+
"step": 2715
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 6.0,
|
| 138 |
+
"step": 3258,
|
| 139 |
+
"train_exact_match": 47.752247752247754,
|
| 140 |
+
"train_f1": 60.68401873806664,
|
| 141 |
+
"train_runtime": 8.9792,
|
| 142 |
+
"train_samples_per_second": 113.373,
|
| 143 |
+
"train_steps_per_second": 4.121
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 6.0,
|
| 147 |
+
"grad_norm": 37.850128173828125,
|
| 148 |
+
"learning_rate": 5e-06,
|
| 149 |
+
"loss": 2.3155,
|
| 150 |
+
"step": 3258
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 6.0,
|
| 154 |
+
"eval_exact_match": 26.125,
|
| 155 |
+
"eval_f1": 40.40349142174044,
|
| 156 |
+
"eval_runtime": 29.0794,
|
| 157 |
+
"eval_samples_per_second": 112.141,
|
| 158 |
+
"eval_steps_per_second": 4.023,
|
| 159 |
+
"step": 3258
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 7.0,
|
| 163 |
+
"step": 3801,
|
| 164 |
+
"train_exact_match": 50.44955044955045,
|
| 165 |
+
"train_f1": 64.76949286642217,
|
| 166 |
+
"train_runtime": 8.8774,
|
| 167 |
+
"train_samples_per_second": 113.547,
|
| 168 |
+
"train_steps_per_second": 4.055
|
| 169 |
+
},
|
| 170 |
+
{
|
| 171 |
+
"epoch": 7.0,
|
| 172 |
+
"grad_norm": 51.20785140991211,
|
| 173 |
+
"learning_rate": 3.7500000000000005e-06,
|
| 174 |
+
"loss": 2.0695,
|
| 175 |
+
"step": 3801
|
| 176 |
+
},
|
| 177 |
+
{
|
| 178 |
+
"epoch": 7.0,
|
| 179 |
+
"eval_exact_match": 27.8125,
|
| 180 |
+
"eval_f1": 42.61264490274304,
|
| 181 |
+
"eval_runtime": 29.0203,
|
| 182 |
+
"eval_samples_per_second": 112.37,
|
| 183 |
+
"eval_steps_per_second": 4.032,
|
| 184 |
+
"step": 3801
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 8.0,
|
| 188 |
+
"step": 4344,
|
| 189 |
+
"train_exact_match": 51.34865134865135,
|
| 190 |
+
"train_f1": 65.11930431582645,
|
| 191 |
+
"train_runtime": 8.9923,
|
| 192 |
+
"train_samples_per_second": 113.208,
|
| 193 |
+
"train_steps_per_second": 4.115
|
| 194 |
+
},
|
| 195 |
+
{
|
| 196 |
+
"epoch": 8.0,
|
| 197 |
+
"grad_norm": 56.128509521484375,
|
| 198 |
+
"learning_rate": 2.5e-06,
|
| 199 |
+
"loss": 1.8975,
|
| 200 |
+
"step": 4344
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 8.0,
|
| 204 |
+
"eval_exact_match": 29.125,
|
| 205 |
+
"eval_f1": 43.50290837664036,
|
| 206 |
+
"eval_runtime": 28.9786,
|
| 207 |
+
"eval_samples_per_second": 112.531,
|
| 208 |
+
"eval_steps_per_second": 4.037,
|
| 209 |
+
"step": 4344
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 9.0,
|
| 213 |
+
"step": 4887,
|
| 214 |
+
"train_exact_match": 57.24275724275724,
|
| 215 |
+
"train_f1": 69.40786106942967,
|
| 216 |
+
"train_runtime": 9.0219,
|
| 217 |
+
"train_samples_per_second": 112.615,
|
| 218 |
+
"train_steps_per_second": 4.101
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 9.0,
|
| 222 |
+
"grad_norm": 60.7649040222168,
|
| 223 |
+
"learning_rate": 1.25e-06,
|
| 224 |
+
"loss": 1.7833,
|
| 225 |
+
"step": 4887
|
| 226 |
+
},
|
| 227 |
+
{
|
| 228 |
+
"epoch": 9.0,
|
| 229 |
+
"eval_exact_match": 29.1875,
|
| 230 |
+
"eval_f1": 43.93033202135494,
|
| 231 |
+
"eval_runtime": 29.1076,
|
| 232 |
+
"eval_samples_per_second": 112.033,
|
| 233 |
+
"eval_steps_per_second": 4.02,
|
| 234 |
+
"step": 4887
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 10.0,
|
| 238 |
+
"step": 5430,
|
| 239 |
+
"train_exact_match": 58.04195804195804,
|
| 240 |
+
"train_f1": 70.9609187667352,
|
| 241 |
+
"train_runtime": 9.0761,
|
| 242 |
+
"train_samples_per_second": 112.383,
|
| 243 |
+
"train_steps_per_second": 4.077
|
| 244 |
+
},
|
| 245 |
+
{
|
| 246 |
+
"epoch": 10.0,
|
| 247 |
+
"grad_norm": 42.582576751708984,
|
| 248 |
+
"learning_rate": 0.0,
|
| 249 |
+
"loss": 1.7125,
|
| 250 |
+
"step": 5430
|
| 251 |
+
},
|
| 252 |
+
{
|
| 253 |
+
"epoch": 10.0,
|
| 254 |
+
"eval_exact_match": 29.1875,
|
| 255 |
+
"eval_f1": 44.04121981988171,
|
| 256 |
+
"eval_runtime": 28.9802,
|
| 257 |
+
"eval_samples_per_second": 112.525,
|
| 258 |
+
"eval_steps_per_second": 4.037,
|
| 259 |
+
"step": 5430
|
| 260 |
+
},
|
| 261 |
+
{
|
| 262 |
+
"epoch": 10.0,
|
| 263 |
+
"step": 5430,
|
| 264 |
+
"total_flos": 1.487387554039296e+16,
|
| 265 |
+
"train_loss": 2.7391692272207355,
|
| 266 |
+
"train_runtime": 1969.257,
|
| 267 |
+
"train_samples_per_second": 77.08,
|
| 268 |
+
"train_steps_per_second": 2.757
|
| 269 |
+
}
|
| 270 |
+
],
|
| 271 |
+
"logging_steps": 500,
|
| 272 |
+
"max_steps": 5430,
|
| 273 |
+
"num_input_tokens_seen": 0,
|
| 274 |
+
"num_train_epochs": 10,
|
| 275 |
+
"save_steps": 500,
|
| 276 |
+
"total_flos": 1.487387554039296e+16,
|
| 277 |
+
"train_batch_size": 28,
|
| 278 |
+
"trial_name": null,
|
| 279 |
+
"trial_params": null
|
| 280 |
+
}
|