End of training
Browse files
all_results.json
ADDED
|
@@ -0,0 +1,15 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"eval_exact_match": 63.28125,
|
| 4 |
+
"eval_f1": 78.83324934908322,
|
| 5 |
+
"eval_runtime": 55.0168,
|
| 6 |
+
"eval_samples": 4618,
|
| 7 |
+
"eval_samples_per_second": 83.938,
|
| 8 |
+
"eval_steps_per_second": 2.999,
|
| 9 |
+
"total_flos": 4.108760851295232e+16,
|
| 10 |
+
"train_loss": 1.1428948192316317,
|
| 11 |
+
"train_runtime": 4968.485,
|
| 12 |
+
"train_samples": 20966,
|
| 13 |
+
"train_samples_per_second": 42.198,
|
| 14 |
+
"train_steps_per_second": 1.508
|
| 15 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"eval_exact_match": 63.28125,
|
| 4 |
+
"eval_f1": 78.83324934908322,
|
| 5 |
+
"eval_runtime": 55.0168,
|
| 6 |
+
"eval_samples": 4618,
|
| 7 |
+
"eval_samples_per_second": 83.938,
|
| 8 |
+
"eval_steps_per_second": 2.999
|
| 9 |
+
}
|
runs/May10_20-23-32_ml-machine/events.out.tfevents.1715377758.ml-machine.119659.1
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4709f0c90511310bb0212ac49564b635774c62877c52fe1bf46191ca0e5218cd
|
| 3 |
+
size 412
|
train_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"total_flos": 4.108760851295232e+16,
|
| 4 |
+
"train_loss": 1.1428948192316317,
|
| 5 |
+
"train_runtime": 4968.485,
|
| 6 |
+
"train_samples": 20966,
|
| 7 |
+
"train_samples_per_second": 42.198,
|
| 8 |
+
"train_steps_per_second": 1.508
|
| 9 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,280 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 78.83324934908322,
|
| 3 |
+
"best_model_checkpoint": "/root/turkic_qa/tr_kaz_models/tr_kaz_xlm_roberta_base_model/checkpoint-7490",
|
| 4 |
+
"epoch": 10.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 7490,
|
| 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": 749,
|
| 14 |
+
"train_exact_match": 24.275724275724276,
|
| 15 |
+
"train_f1": 43.117743763484455,
|
| 16 |
+
"train_runtime": 19.0896,
|
| 17 |
+
"train_samples_per_second": 79.624,
|
| 18 |
+
"train_steps_per_second": 2.881
|
| 19 |
+
},
|
| 20 |
+
{
|
| 21 |
+
"epoch": 1.0,
|
| 22 |
+
"grad_norm": 45.69759750366211,
|
| 23 |
+
"learning_rate": 5e-06,
|
| 24 |
+
"loss": 4.4853,
|
| 25 |
+
"step": 749
|
| 26 |
+
},
|
| 27 |
+
{
|
| 28 |
+
"epoch": 1.0,
|
| 29 |
+
"eval_exact_match": 25.875,
|
| 30 |
+
"eval_f1": 44.604620754612284,
|
| 31 |
+
"eval_runtime": 56.7402,
|
| 32 |
+
"eval_samples_per_second": 81.389,
|
| 33 |
+
"eval_steps_per_second": 2.908,
|
| 34 |
+
"step": 749
|
| 35 |
+
},
|
| 36 |
+
{
|
| 37 |
+
"epoch": 2.0,
|
| 38 |
+
"step": 1498,
|
| 39 |
+
"train_exact_match": 55.744255744255746,
|
| 40 |
+
"train_f1": 73.08153342630565,
|
| 41 |
+
"train_runtime": 17.3477,
|
| 42 |
+
"train_samples_per_second": 85.429,
|
| 43 |
+
"train_steps_per_second": 3.055
|
| 44 |
+
},
|
| 45 |
+
{
|
| 46 |
+
"epoch": 2.0,
|
| 47 |
+
"grad_norm": 27.808273315429688,
|
| 48 |
+
"learning_rate": 1e-05,
|
| 49 |
+
"loss": 1.7458,
|
| 50 |
+
"step": 1498
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 2.0,
|
| 54 |
+
"eval_exact_match": 52.15625,
|
| 55 |
+
"eval_f1": 70.19574840484874,
|
| 56 |
+
"eval_runtime": 53.2765,
|
| 57 |
+
"eval_samples_per_second": 86.68,
|
| 58 |
+
"eval_steps_per_second": 3.097,
|
| 59 |
+
"step": 1498
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 3.0,
|
| 63 |
+
"step": 2247,
|
| 64 |
+
"train_exact_match": 65.83416583416583,
|
| 65 |
+
"train_f1": 81.64944541515409,
|
| 66 |
+
"train_runtime": 17.576,
|
| 67 |
+
"train_samples_per_second": 85.173,
|
| 68 |
+
"train_steps_per_second": 3.072
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 3.0,
|
| 72 |
+
"grad_norm": 26.646291732788086,
|
| 73 |
+
"learning_rate": 8.750000000000001e-06,
|
| 74 |
+
"loss": 1.1281,
|
| 75 |
+
"step": 2247
|
| 76 |
+
},
|
| 77 |
+
{
|
| 78 |
+
"epoch": 3.0,
|
| 79 |
+
"eval_exact_match": 58.0625,
|
| 80 |
+
"eval_f1": 75.09391733824289,
|
| 81 |
+
"eval_runtime": 56.1114,
|
| 82 |
+
"eval_samples_per_second": 82.301,
|
| 83 |
+
"eval_steps_per_second": 2.941,
|
| 84 |
+
"step": 2247
|
| 85 |
+
},
|
| 86 |
+
{
|
| 87 |
+
"epoch": 4.0,
|
| 88 |
+
"step": 2996,
|
| 89 |
+
"train_exact_match": 69.43056943056943,
|
| 90 |
+
"train_f1": 84.4692905630636,
|
| 91 |
+
"train_runtime": 19.0377,
|
| 92 |
+
"train_samples_per_second": 79.106,
|
| 93 |
+
"train_steps_per_second": 2.836
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 4.0,
|
| 97 |
+
"grad_norm": 22.081819534301758,
|
| 98 |
+
"learning_rate": 7.500000000000001e-06,
|
| 99 |
+
"loss": 0.8784,
|
| 100 |
+
"step": 2996
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 4.0,
|
| 104 |
+
"eval_exact_match": 60.25,
|
| 105 |
+
"eval_f1": 76.8410082950037,
|
| 106 |
+
"eval_runtime": 57.4346,
|
| 107 |
+
"eval_samples_per_second": 80.405,
|
| 108 |
+
"eval_steps_per_second": 2.873,
|
| 109 |
+
"step": 2996
|
| 110 |
+
},
|
| 111 |
+
{
|
| 112 |
+
"epoch": 5.0,
|
| 113 |
+
"step": 3745,
|
| 114 |
+
"train_exact_match": 74.12587412587412,
|
| 115 |
+
"train_f1": 87.57278612666818,
|
| 116 |
+
"train_runtime": 18.9437,
|
| 117 |
+
"train_samples_per_second": 78.971,
|
| 118 |
+
"train_steps_per_second": 2.851
|
| 119 |
+
},
|
| 120 |
+
{
|
| 121 |
+
"epoch": 5.0,
|
| 122 |
+
"grad_norm": 27.527299880981445,
|
| 123 |
+
"learning_rate": 6.25e-06,
|
| 124 |
+
"loss": 0.7365,
|
| 125 |
+
"step": 3745
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"epoch": 5.0,
|
| 129 |
+
"eval_exact_match": 60.96875,
|
| 130 |
+
"eval_f1": 77.33365402370875,
|
| 131 |
+
"eval_runtime": 57.2578,
|
| 132 |
+
"eval_samples_per_second": 80.653,
|
| 133 |
+
"eval_steps_per_second": 2.882,
|
| 134 |
+
"step": 3745
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 6.0,
|
| 138 |
+
"step": 4494,
|
| 139 |
+
"train_exact_match": 78.42157842157842,
|
| 140 |
+
"train_f1": 89.99797070222537,
|
| 141 |
+
"train_runtime": 19.0634,
|
| 142 |
+
"train_samples_per_second": 78.265,
|
| 143 |
+
"train_steps_per_second": 2.833
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 6.0,
|
| 147 |
+
"grad_norm": 26.452974319458008,
|
| 148 |
+
"learning_rate": 5e-06,
|
| 149 |
+
"loss": 0.6194,
|
| 150 |
+
"step": 4494
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 6.0,
|
| 154 |
+
"eval_exact_match": 61.96875,
|
| 155 |
+
"eval_f1": 77.91252095789356,
|
| 156 |
+
"eval_runtime": 57.9105,
|
| 157 |
+
"eval_samples_per_second": 79.744,
|
| 158 |
+
"eval_steps_per_second": 2.849,
|
| 159 |
+
"step": 4494
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 7.0,
|
| 163 |
+
"step": 5243,
|
| 164 |
+
"train_exact_match": 82.81718281718281,
|
| 165 |
+
"train_f1": 92.72768225182865,
|
| 166 |
+
"train_runtime": 18.2446,
|
| 167 |
+
"train_samples_per_second": 80.956,
|
| 168 |
+
"train_steps_per_second": 2.905
|
| 169 |
+
},
|
| 170 |
+
{
|
| 171 |
+
"epoch": 7.0,
|
| 172 |
+
"grad_norm": 18.254165649414062,
|
| 173 |
+
"learning_rate": 3.7500000000000005e-06,
|
| 174 |
+
"loss": 0.534,
|
| 175 |
+
"step": 5243
|
| 176 |
+
},
|
| 177 |
+
{
|
| 178 |
+
"epoch": 7.0,
|
| 179 |
+
"eval_exact_match": 62.21875,
|
| 180 |
+
"eval_f1": 78.17772368782505,
|
| 181 |
+
"eval_runtime": 55.9458,
|
| 182 |
+
"eval_samples_per_second": 82.544,
|
| 183 |
+
"eval_steps_per_second": 2.949,
|
| 184 |
+
"step": 5243
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 8.0,
|
| 188 |
+
"step": 5992,
|
| 189 |
+
"train_exact_match": 80.41958041958041,
|
| 190 |
+
"train_f1": 91.49491612552417,
|
| 191 |
+
"train_runtime": 18.2728,
|
| 192 |
+
"train_samples_per_second": 84.059,
|
| 193 |
+
"train_steps_per_second": 3.01
|
| 194 |
+
},
|
| 195 |
+
{
|
| 196 |
+
"epoch": 8.0,
|
| 197 |
+
"grad_norm": 35.931488037109375,
|
| 198 |
+
"learning_rate": 2.5e-06,
|
| 199 |
+
"loss": 0.4747,
|
| 200 |
+
"step": 5992
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 8.0,
|
| 204 |
+
"eval_exact_match": 62.53125,
|
| 205 |
+
"eval_f1": 78.2110139694925,
|
| 206 |
+
"eval_runtime": 56.5716,
|
| 207 |
+
"eval_samples_per_second": 81.631,
|
| 208 |
+
"eval_steps_per_second": 2.917,
|
| 209 |
+
"step": 5992
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 9.0,
|
| 213 |
+
"step": 6741,
|
| 214 |
+
"train_exact_match": 82.61738261738262,
|
| 215 |
+
"train_f1": 92.86226576706642,
|
| 216 |
+
"train_runtime": 18.569,
|
| 217 |
+
"train_samples_per_second": 79.218,
|
| 218 |
+
"train_steps_per_second": 2.854
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 9.0,
|
| 222 |
+
"grad_norm": 60.296566009521484,
|
| 223 |
+
"learning_rate": 1.25e-06,
|
| 224 |
+
"loss": 0.4304,
|
| 225 |
+
"step": 6741
|
| 226 |
+
},
|
| 227 |
+
{
|
| 228 |
+
"epoch": 9.0,
|
| 229 |
+
"eval_exact_match": 63.0625,
|
| 230 |
+
"eval_f1": 78.5170445431728,
|
| 231 |
+
"eval_runtime": 57.3991,
|
| 232 |
+
"eval_samples_per_second": 80.454,
|
| 233 |
+
"eval_steps_per_second": 2.875,
|
| 234 |
+
"step": 6741
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 10.0,
|
| 238 |
+
"step": 7490,
|
| 239 |
+
"train_exact_match": 84.91508491508492,
|
| 240 |
+
"train_f1": 93.37423644646913,
|
| 241 |
+
"train_runtime": 18.377,
|
| 242 |
+
"train_samples_per_second": 81.297,
|
| 243 |
+
"train_steps_per_second": 2.938
|
| 244 |
+
},
|
| 245 |
+
{
|
| 246 |
+
"epoch": 10.0,
|
| 247 |
+
"grad_norm": 23.199573516845703,
|
| 248 |
+
"learning_rate": 0.0,
|
| 249 |
+
"loss": 0.3964,
|
| 250 |
+
"step": 7490
|
| 251 |
+
},
|
| 252 |
+
{
|
| 253 |
+
"epoch": 10.0,
|
| 254 |
+
"eval_exact_match": 63.28125,
|
| 255 |
+
"eval_f1": 78.83324934908322,
|
| 256 |
+
"eval_runtime": 55.9599,
|
| 257 |
+
"eval_samples_per_second": 82.523,
|
| 258 |
+
"eval_steps_per_second": 2.949,
|
| 259 |
+
"step": 7490
|
| 260 |
+
},
|
| 261 |
+
{
|
| 262 |
+
"epoch": 10.0,
|
| 263 |
+
"step": 7490,
|
| 264 |
+
"total_flos": 4.108760851295232e+16,
|
| 265 |
+
"train_loss": 1.1428948192316317,
|
| 266 |
+
"train_runtime": 4968.485,
|
| 267 |
+
"train_samples_per_second": 42.198,
|
| 268 |
+
"train_steps_per_second": 1.508
|
| 269 |
+
}
|
| 270 |
+
],
|
| 271 |
+
"logging_steps": 500,
|
| 272 |
+
"max_steps": 7490,
|
| 273 |
+
"num_input_tokens_seen": 0,
|
| 274 |
+
"num_train_epochs": 10,
|
| 275 |
+
"save_steps": 500,
|
| 276 |
+
"total_flos": 4.108760851295232e+16,
|
| 277 |
+
"train_batch_size": 28,
|
| 278 |
+
"trial_name": null,
|
| 279 |
+
"trial_params": null
|
| 280 |
+
}
|