End of training
Browse files
all_results.json
ADDED
|
@@ -0,0 +1,15 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"eval_exact_match": 61.0625,
|
| 4 |
+
"eval_f1": 78.05994502395909,
|
| 5 |
+
"eval_runtime": 36.2485,
|
| 6 |
+
"eval_samples": 3281,
|
| 7 |
+
"eval_samples_per_second": 90.514,
|
| 8 |
+
"eval_steps_per_second": 3.255,
|
| 9 |
+
"total_flos": 3.025228525300224e+16,
|
| 10 |
+
"train_loss": 0.8031932664954144,
|
| 11 |
+
"train_runtime": 3531.0877,
|
| 12 |
+
"train_samples": 15437,
|
| 13 |
+
"train_samples_per_second": 43.717,
|
| 14 |
+
"train_steps_per_second": 1.563
|
| 15 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"eval_exact_match": 61.0625,
|
| 4 |
+
"eval_f1": 78.05994502395909,
|
| 5 |
+
"eval_runtime": 36.2485,
|
| 6 |
+
"eval_samples": 3281,
|
| 7 |
+
"eval_samples_per_second": 90.514,
|
| 8 |
+
"eval_steps_per_second": 3.255
|
| 9 |
+
}
|
runs/May10_12-29-40_ml-machine/events.out.tfevents.1715347847.ml-machine.53989.1
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f43524d79fd2379793dbbe965b92f583958eecfe88a96f58d01a18fd4c8c29e2
|
| 3 |
+
size 412
|
train_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"total_flos": 3.025228525300224e+16,
|
| 4 |
+
"train_loss": 0.8031932664954144,
|
| 5 |
+
"train_runtime": 3531.0877,
|
| 6 |
+
"train_samples": 15437,
|
| 7 |
+
"train_samples_per_second": 43.717,
|
| 8 |
+
"train_steps_per_second": 1.563
|
| 9 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,280 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 78.05994502395909,
|
| 3 |
+
"best_model_checkpoint": "/root/turkic_qa/ru_kaz_models/ru_kaz_xlm_roberta_base_squad_model/checkpoint-5520",
|
| 4 |
+
"epoch": 10.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 5520,
|
| 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": 552,
|
| 14 |
+
"train_exact_match": 56.34365634365634,
|
| 15 |
+
"train_f1": 75.2715442869406,
|
| 16 |
+
"train_runtime": 11.3355,
|
| 17 |
+
"train_samples_per_second": 89.63,
|
| 18 |
+
"train_steps_per_second": 3.264
|
| 19 |
+
},
|
| 20 |
+
{
|
| 21 |
+
"epoch": 1.0,
|
| 22 |
+
"grad_norm": 50.90898132324219,
|
| 23 |
+
"learning_rate": 5e-06,
|
| 24 |
+
"loss": 1.5648,
|
| 25 |
+
"step": 552
|
| 26 |
+
},
|
| 27 |
+
{
|
| 28 |
+
"epoch": 1.0,
|
| 29 |
+
"eval_exact_match": 55.0625,
|
| 30 |
+
"eval_f1": 73.73760370127802,
|
| 31 |
+
"eval_runtime": 36.5495,
|
| 32 |
+
"eval_samples_per_second": 89.769,
|
| 33 |
+
"eval_steps_per_second": 3.229,
|
| 34 |
+
"step": 552
|
| 35 |
+
},
|
| 36 |
+
{
|
| 37 |
+
"epoch": 2.0,
|
| 38 |
+
"step": 1104,
|
| 39 |
+
"train_exact_match": 66.93306693306694,
|
| 40 |
+
"train_f1": 82.49314886706897,
|
| 41 |
+
"train_runtime": 11.3179,
|
| 42 |
+
"train_samples_per_second": 90.83,
|
| 43 |
+
"train_steps_per_second": 3.269
|
| 44 |
+
},
|
| 45 |
+
{
|
| 46 |
+
"epoch": 2.0,
|
| 47 |
+
"grad_norm": 55.97765350341797,
|
| 48 |
+
"learning_rate": 1e-05,
|
| 49 |
+
"loss": 1.2158,
|
| 50 |
+
"step": 1104
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 2.0,
|
| 54 |
+
"eval_exact_match": 58.84375,
|
| 55 |
+
"eval_f1": 76.49071665255673,
|
| 56 |
+
"eval_runtime": 36.2092,
|
| 57 |
+
"eval_samples_per_second": 90.612,
|
| 58 |
+
"eval_steps_per_second": 3.259,
|
| 59 |
+
"step": 1104
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 3.0,
|
| 63 |
+
"step": 1656,
|
| 64 |
+
"train_exact_match": 67.53246753246754,
|
| 65 |
+
"train_f1": 85.19091955779871,
|
| 66 |
+
"train_runtime": 11.374,
|
| 67 |
+
"train_samples_per_second": 89.415,
|
| 68 |
+
"train_steps_per_second": 3.253
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 3.0,
|
| 72 |
+
"grad_norm": 57.926448822021484,
|
| 73 |
+
"learning_rate": 8.750000000000001e-06,
|
| 74 |
+
"loss": 1.028,
|
| 75 |
+
"step": 1656
|
| 76 |
+
},
|
| 77 |
+
{
|
| 78 |
+
"epoch": 3.0,
|
| 79 |
+
"eval_exact_match": 60.0,
|
| 80 |
+
"eval_f1": 77.43391533868831,
|
| 81 |
+
"eval_runtime": 36.533,
|
| 82 |
+
"eval_samples_per_second": 89.809,
|
| 83 |
+
"eval_steps_per_second": 3.23,
|
| 84 |
+
"step": 1656
|
| 85 |
+
},
|
| 86 |
+
{
|
| 87 |
+
"epoch": 4.0,
|
| 88 |
+
"step": 2208,
|
| 89 |
+
"train_exact_match": 74.22577422577423,
|
| 90 |
+
"train_f1": 89.15395273332221,
|
| 91 |
+
"train_runtime": 11.3619,
|
| 92 |
+
"train_samples_per_second": 89.246,
|
| 93 |
+
"train_steps_per_second": 3.256
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 4.0,
|
| 97 |
+
"grad_norm": 38.46842956542969,
|
| 98 |
+
"learning_rate": 7.500000000000001e-06,
|
| 99 |
+
"loss": 0.8582,
|
| 100 |
+
"step": 2208
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 4.0,
|
| 104 |
+
"eval_exact_match": 60.625,
|
| 105 |
+
"eval_f1": 77.72152196665945,
|
| 106 |
+
"eval_runtime": 36.465,
|
| 107 |
+
"eval_samples_per_second": 89.977,
|
| 108 |
+
"eval_steps_per_second": 3.236,
|
| 109 |
+
"step": 2208
|
| 110 |
+
},
|
| 111 |
+
{
|
| 112 |
+
"epoch": 5.0,
|
| 113 |
+
"step": 2760,
|
| 114 |
+
"train_exact_match": 77.12287712287713,
|
| 115 |
+
"train_f1": 90.14502448136521,
|
| 116 |
+
"train_runtime": 11.5016,
|
| 117 |
+
"train_samples_per_second": 88.857,
|
| 118 |
+
"train_steps_per_second": 3.217
|
| 119 |
+
},
|
| 120 |
+
{
|
| 121 |
+
"epoch": 5.0,
|
| 122 |
+
"grad_norm": 14.753046989440918,
|
| 123 |
+
"learning_rate": 6.25e-06,
|
| 124 |
+
"loss": 0.7281,
|
| 125 |
+
"step": 2760
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"epoch": 5.0,
|
| 129 |
+
"eval_exact_match": 60.8125,
|
| 130 |
+
"eval_f1": 77.62510042626374,
|
| 131 |
+
"eval_runtime": 36.4524,
|
| 132 |
+
"eval_samples_per_second": 90.008,
|
| 133 |
+
"eval_steps_per_second": 3.237,
|
| 134 |
+
"step": 2760
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 6.0,
|
| 138 |
+
"step": 3312,
|
| 139 |
+
"train_exact_match": 80.21978021978022,
|
| 140 |
+
"train_f1": 92.23780908752109,
|
| 141 |
+
"train_runtime": 11.7336,
|
| 142 |
+
"train_samples_per_second": 87.101,
|
| 143 |
+
"train_steps_per_second": 3.153
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 6.0,
|
| 147 |
+
"grad_norm": 23.36362075805664,
|
| 148 |
+
"learning_rate": 5e-06,
|
| 149 |
+
"loss": 0.6422,
|
| 150 |
+
"step": 3312
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 6.0,
|
| 154 |
+
"eval_exact_match": 61.28125,
|
| 155 |
+
"eval_f1": 77.91507488796293,
|
| 156 |
+
"eval_runtime": 37.4038,
|
| 157 |
+
"eval_samples_per_second": 87.718,
|
| 158 |
+
"eval_steps_per_second": 3.155,
|
| 159 |
+
"step": 3312
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 7.0,
|
| 163 |
+
"step": 3864,
|
| 164 |
+
"train_exact_match": 81.81818181818181,
|
| 165 |
+
"train_f1": 93.62881852209286,
|
| 166 |
+
"train_runtime": 11.6143,
|
| 167 |
+
"train_samples_per_second": 88.512,
|
| 168 |
+
"train_steps_per_second": 3.186
|
| 169 |
+
},
|
| 170 |
+
{
|
| 171 |
+
"epoch": 7.0,
|
| 172 |
+
"grad_norm": 21.251754760742188,
|
| 173 |
+
"learning_rate": 3.7500000000000005e-06,
|
| 174 |
+
"loss": 0.5646,
|
| 175 |
+
"step": 3864
|
| 176 |
+
},
|
| 177 |
+
{
|
| 178 |
+
"epoch": 7.0,
|
| 179 |
+
"eval_exact_match": 61.1875,
|
| 180 |
+
"eval_f1": 77.88113274187097,
|
| 181 |
+
"eval_runtime": 36.8966,
|
| 182 |
+
"eval_samples_per_second": 88.924,
|
| 183 |
+
"eval_steps_per_second": 3.198,
|
| 184 |
+
"step": 3864
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 8.0,
|
| 188 |
+
"step": 4416,
|
| 189 |
+
"train_exact_match": 84.31568431568432,
|
| 190 |
+
"train_f1": 93.90553661277687,
|
| 191 |
+
"train_runtime": 11.4894,
|
| 192 |
+
"train_samples_per_second": 89.126,
|
| 193 |
+
"train_steps_per_second": 3.22
|
| 194 |
+
},
|
| 195 |
+
{
|
| 196 |
+
"epoch": 8.0,
|
| 197 |
+
"grad_norm": 31.64153289794922,
|
| 198 |
+
"learning_rate": 2.5e-06,
|
| 199 |
+
"loss": 0.5156,
|
| 200 |
+
"step": 4416
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 8.0,
|
| 204 |
+
"eval_exact_match": 60.71875,
|
| 205 |
+
"eval_f1": 77.56837719428701,
|
| 206 |
+
"eval_runtime": 36.6849,
|
| 207 |
+
"eval_samples_per_second": 89.437,
|
| 208 |
+
"eval_steps_per_second": 3.217,
|
| 209 |
+
"step": 4416
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 9.0,
|
| 213 |
+
"step": 4968,
|
| 214 |
+
"train_exact_match": 85.11488511488511,
|
| 215 |
+
"train_f1": 94.79917066124965,
|
| 216 |
+
"train_runtime": 11.5536,
|
| 217 |
+
"train_samples_per_second": 89.496,
|
| 218 |
+
"train_steps_per_second": 3.202
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 9.0,
|
| 222 |
+
"grad_norm": 46.01136779785156,
|
| 223 |
+
"learning_rate": 1.25e-06,
|
| 224 |
+
"loss": 0.4658,
|
| 225 |
+
"step": 4968
|
| 226 |
+
},
|
| 227 |
+
{
|
| 228 |
+
"epoch": 9.0,
|
| 229 |
+
"eval_exact_match": 61.125,
|
| 230 |
+
"eval_f1": 77.88301584766332,
|
| 231 |
+
"eval_runtime": 36.5869,
|
| 232 |
+
"eval_samples_per_second": 89.677,
|
| 233 |
+
"eval_steps_per_second": 3.225,
|
| 234 |
+
"step": 4968
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 10.0,
|
| 238 |
+
"step": 5520,
|
| 239 |
+
"train_exact_match": 84.51548451548452,
|
| 240 |
+
"train_f1": 94.34363228489812,
|
| 241 |
+
"train_runtime": 11.395,
|
| 242 |
+
"train_samples_per_second": 90.479,
|
| 243 |
+
"train_steps_per_second": 3.247
|
| 244 |
+
},
|
| 245 |
+
{
|
| 246 |
+
"epoch": 10.0,
|
| 247 |
+
"grad_norm": 84.91465759277344,
|
| 248 |
+
"learning_rate": 0.0,
|
| 249 |
+
"loss": 0.4488,
|
| 250 |
+
"step": 5520
|
| 251 |
+
},
|
| 252 |
+
{
|
| 253 |
+
"epoch": 10.0,
|
| 254 |
+
"eval_exact_match": 61.0625,
|
| 255 |
+
"eval_f1": 78.05994502395909,
|
| 256 |
+
"eval_runtime": 36.4836,
|
| 257 |
+
"eval_samples_per_second": 89.931,
|
| 258 |
+
"eval_steps_per_second": 3.234,
|
| 259 |
+
"step": 5520
|
| 260 |
+
},
|
| 261 |
+
{
|
| 262 |
+
"epoch": 10.0,
|
| 263 |
+
"step": 5520,
|
| 264 |
+
"total_flos": 3.025228525300224e+16,
|
| 265 |
+
"train_loss": 0.8031932664954144,
|
| 266 |
+
"train_runtime": 3531.0877,
|
| 267 |
+
"train_samples_per_second": 43.717,
|
| 268 |
+
"train_steps_per_second": 1.563
|
| 269 |
+
}
|
| 270 |
+
],
|
| 271 |
+
"logging_steps": 500,
|
| 272 |
+
"max_steps": 5520,
|
| 273 |
+
"num_input_tokens_seen": 0,
|
| 274 |
+
"num_train_epochs": 10,
|
| 275 |
+
"save_steps": 500,
|
| 276 |
+
"total_flos": 3.025228525300224e+16,
|
| 277 |
+
"train_batch_size": 28,
|
| 278 |
+
"trial_name": null,
|
| 279 |
+
"trial_params": null
|
| 280 |
+
}
|