End of training
Browse files
all_results.json
ADDED
|
@@ -0,0 +1,15 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"eval_exact_match": 24.9375,
|
| 4 |
+
"eval_f1": 36.31328557707662,
|
| 5 |
+
"eval_runtime": 40.5237,
|
| 6 |
+
"eval_samples": 4583,
|
| 7 |
+
"eval_samples_per_second": 113.094,
|
| 8 |
+
"eval_steps_per_second": 4.047,
|
| 9 |
+
"total_flos": 2.004675840340992e+16,
|
| 10 |
+
"train_loss": 2.4619675162715886,
|
| 11 |
+
"train_runtime": 2688.9862,
|
| 12 |
+
"train_samples": 20458,
|
| 13 |
+
"train_samples_per_second": 76.081,
|
| 14 |
+
"train_steps_per_second": 2.718
|
| 15 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"eval_exact_match": 24.9375,
|
| 4 |
+
"eval_f1": 36.31328557707662,
|
| 5 |
+
"eval_runtime": 40.5237,
|
| 6 |
+
"eval_samples": 4583,
|
| 7 |
+
"eval_samples_per_second": 113.094,
|
| 8 |
+
"eval_steps_per_second": 4.047
|
| 9 |
+
}
|
runs/May11_19-45-25_ml-machine/events.out.tfevents.1715459525.ml-machine.20503.1
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d73b4bfcdea5c7e9801055858a7423a5030a317fa92fc903a0c7e3889d8eff9a
|
| 3 |
+
size 412
|
train_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"total_flos": 2.004675840340992e+16,
|
| 4 |
+
"train_loss": 2.4619675162715886,
|
| 5 |
+
"train_runtime": 2688.9862,
|
| 6 |
+
"train_samples": 20458,
|
| 7 |
+
"train_samples_per_second": 76.081,
|
| 8 |
+
"train_steps_per_second": 2.718
|
| 9 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,280 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 36.31328557707662,
|
| 3 |
+
"best_model_checkpoint": "/root/turkic_qa/tr_uzn_models/orig_uzn_roberta_base_model/checkpoint-7310",
|
| 4 |
+
"epoch": 10.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 7310,
|
| 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": 731,
|
| 14 |
+
"train_exact_match": 6.893106893106893,
|
| 15 |
+
"train_f1": 13.147750410374615,
|
| 16 |
+
"train_runtime": 13.1312,
|
| 17 |
+
"train_samples_per_second": 113.851,
|
| 18 |
+
"train_steps_per_second": 4.112
|
| 19 |
+
},
|
| 20 |
+
{
|
| 21 |
+
"epoch": 1.0,
|
| 22 |
+
"grad_norm": 12.239034652709961,
|
| 23 |
+
"learning_rate": 5e-06,
|
| 24 |
+
"loss": 4.418,
|
| 25 |
+
"step": 731
|
| 26 |
+
},
|
| 27 |
+
{
|
| 28 |
+
"epoch": 1.0,
|
| 29 |
+
"eval_exact_match": 4.3125,
|
| 30 |
+
"eval_f1": 11.281225208823463,
|
| 31 |
+
"eval_runtime": 40.2248,
|
| 32 |
+
"eval_samples_per_second": 113.935,
|
| 33 |
+
"eval_steps_per_second": 4.077,
|
| 34 |
+
"step": 731
|
| 35 |
+
},
|
| 36 |
+
{
|
| 37 |
+
"epoch": 2.0,
|
| 38 |
+
"step": 1462,
|
| 39 |
+
"train_exact_match": 15.484515484515484,
|
| 40 |
+
"train_f1": 23.00008673777926,
|
| 41 |
+
"train_runtime": 12.8965,
|
| 42 |
+
"train_samples_per_second": 113.364,
|
| 43 |
+
"train_steps_per_second": 4.11
|
| 44 |
+
},
|
| 45 |
+
{
|
| 46 |
+
"epoch": 2.0,
|
| 47 |
+
"grad_norm": 13.470973014831543,
|
| 48 |
+
"learning_rate": 1e-05,
|
| 49 |
+
"loss": 3.3027,
|
| 50 |
+
"step": 1462
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 2.0,
|
| 54 |
+
"eval_exact_match": 10.5,
|
| 55 |
+
"eval_f1": 18.60249294088304,
|
| 56 |
+
"eval_runtime": 40.2237,
|
| 57 |
+
"eval_samples_per_second": 113.938,
|
| 58 |
+
"eval_steps_per_second": 4.077,
|
| 59 |
+
"step": 1462
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 3.0,
|
| 63 |
+
"step": 2193,
|
| 64 |
+
"train_exact_match": 24.475524475524477,
|
| 65 |
+
"train_f1": 33.75749140151479,
|
| 66 |
+
"train_runtime": 12.4836,
|
| 67 |
+
"train_samples_per_second": 113.83,
|
| 68 |
+
"train_steps_per_second": 4.085
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 3.0,
|
| 72 |
+
"grad_norm": 11.94182014465332,
|
| 73 |
+
"learning_rate": 8.750000000000001e-06,
|
| 74 |
+
"loss": 2.8501,
|
| 75 |
+
"step": 2193
|
| 76 |
+
},
|
| 77 |
+
{
|
| 78 |
+
"epoch": 3.0,
|
| 79 |
+
"eval_exact_match": 14.96875,
|
| 80 |
+
"eval_f1": 23.95818931657813,
|
| 81 |
+
"eval_runtime": 40.3346,
|
| 82 |
+
"eval_samples_per_second": 113.625,
|
| 83 |
+
"eval_steps_per_second": 4.066,
|
| 84 |
+
"step": 2193
|
| 85 |
+
},
|
| 86 |
+
{
|
| 87 |
+
"epoch": 4.0,
|
| 88 |
+
"step": 2924,
|
| 89 |
+
"train_exact_match": 32.56743256743257,
|
| 90 |
+
"train_f1": 41.56263168632173,
|
| 91 |
+
"train_runtime": 12.6655,
|
| 92 |
+
"train_samples_per_second": 112.826,
|
| 93 |
+
"train_steps_per_second": 4.106
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 4.0,
|
| 97 |
+
"grad_norm": 13.756119728088379,
|
| 98 |
+
"learning_rate": 7.500000000000001e-06,
|
| 99 |
+
"loss": 2.5536,
|
| 100 |
+
"step": 2924
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 4.0,
|
| 104 |
+
"eval_exact_match": 17.875,
|
| 105 |
+
"eval_f1": 27.247899577169083,
|
| 106 |
+
"eval_runtime": 40.1853,
|
| 107 |
+
"eval_samples_per_second": 114.047,
|
| 108 |
+
"eval_steps_per_second": 4.081,
|
| 109 |
+
"step": 2924
|
| 110 |
+
},
|
| 111 |
+
{
|
| 112 |
+
"epoch": 5.0,
|
| 113 |
+
"step": 3655,
|
| 114 |
+
"train_exact_match": 36.76323676323676,
|
| 115 |
+
"train_f1": 46.68974245865671,
|
| 116 |
+
"train_runtime": 12.7455,
|
| 117 |
+
"train_samples_per_second": 112.981,
|
| 118 |
+
"train_steps_per_second": 4.08
|
| 119 |
+
},
|
| 120 |
+
{
|
| 121 |
+
"epoch": 5.0,
|
| 122 |
+
"grad_norm": 15.806905746459961,
|
| 123 |
+
"learning_rate": 6.25e-06,
|
| 124 |
+
"loss": 2.3273,
|
| 125 |
+
"step": 3655
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"epoch": 5.0,
|
| 129 |
+
"eval_exact_match": 19.75,
|
| 130 |
+
"eval_f1": 29.976243167371322,
|
| 131 |
+
"eval_runtime": 40.5485,
|
| 132 |
+
"eval_samples_per_second": 113.025,
|
| 133 |
+
"eval_steps_per_second": 4.045,
|
| 134 |
+
"step": 3655
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 6.0,
|
| 138 |
+
"step": 4386,
|
| 139 |
+
"train_exact_match": 45.45454545454545,
|
| 140 |
+
"train_f1": 55.6255817809433,
|
| 141 |
+
"train_runtime": 13.0466,
|
| 142 |
+
"train_samples_per_second": 113.517,
|
| 143 |
+
"train_steps_per_second": 4.062
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 6.0,
|
| 147 |
+
"grad_norm": 22.134902954101562,
|
| 148 |
+
"learning_rate": 5e-06,
|
| 149 |
+
"loss": 2.111,
|
| 150 |
+
"step": 4386
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 6.0,
|
| 154 |
+
"eval_exact_match": 21.875,
|
| 155 |
+
"eval_f1": 32.46367810578293,
|
| 156 |
+
"eval_runtime": 40.5576,
|
| 157 |
+
"eval_samples_per_second": 113.0,
|
| 158 |
+
"eval_steps_per_second": 4.044,
|
| 159 |
+
"step": 4386
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 7.0,
|
| 163 |
+
"step": 5117,
|
| 164 |
+
"train_exact_match": 48.451548451548454,
|
| 165 |
+
"train_f1": 59.11155707778988,
|
| 166 |
+
"train_runtime": 12.7339,
|
| 167 |
+
"train_samples_per_second": 112.456,
|
| 168 |
+
"train_steps_per_second": 4.084
|
| 169 |
+
},
|
| 170 |
+
{
|
| 171 |
+
"epoch": 7.0,
|
| 172 |
+
"grad_norm": 18.71537971496582,
|
| 173 |
+
"learning_rate": 3.7500000000000005e-06,
|
| 174 |
+
"loss": 1.9313,
|
| 175 |
+
"step": 5117
|
| 176 |
+
},
|
| 177 |
+
{
|
| 178 |
+
"epoch": 7.0,
|
| 179 |
+
"eval_exact_match": 23.34375,
|
| 180 |
+
"eval_f1": 34.34464862509141,
|
| 181 |
+
"eval_runtime": 40.5488,
|
| 182 |
+
"eval_samples_per_second": 113.024,
|
| 183 |
+
"eval_steps_per_second": 4.045,
|
| 184 |
+
"step": 5117
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 8.0,
|
| 188 |
+
"step": 5848,
|
| 189 |
+
"train_exact_match": 52.34765234765235,
|
| 190 |
+
"train_f1": 61.4699737458471,
|
| 191 |
+
"train_runtime": 12.2502,
|
| 192 |
+
"train_samples_per_second": 112.978,
|
| 193 |
+
"train_steps_per_second": 4.082
|
| 194 |
+
},
|
| 195 |
+
{
|
| 196 |
+
"epoch": 8.0,
|
| 197 |
+
"grad_norm": 21.410564422607422,
|
| 198 |
+
"learning_rate": 2.5e-06,
|
| 199 |
+
"loss": 1.7907,
|
| 200 |
+
"step": 5848
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 8.0,
|
| 204 |
+
"eval_exact_match": 24.65625,
|
| 205 |
+
"eval_f1": 35.76160110698192,
|
| 206 |
+
"eval_runtime": 40.4412,
|
| 207 |
+
"eval_samples_per_second": 113.325,
|
| 208 |
+
"eval_steps_per_second": 4.055,
|
| 209 |
+
"step": 5848
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 9.0,
|
| 213 |
+
"step": 6579,
|
| 214 |
+
"train_exact_match": 56.34365634365634,
|
| 215 |
+
"train_f1": 65.84711342159365,
|
| 216 |
+
"train_runtime": 12.9755,
|
| 217 |
+
"train_samples_per_second": 113.213,
|
| 218 |
+
"train_steps_per_second": 4.085
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 9.0,
|
| 222 |
+
"grad_norm": 18.834157943725586,
|
| 223 |
+
"learning_rate": 1.25e-06,
|
| 224 |
+
"loss": 1.6986,
|
| 225 |
+
"step": 6579
|
| 226 |
+
},
|
| 227 |
+
{
|
| 228 |
+
"epoch": 9.0,
|
| 229 |
+
"eval_exact_match": 24.8125,
|
| 230 |
+
"eval_f1": 35.999416160714,
|
| 231 |
+
"eval_runtime": 40.5359,
|
| 232 |
+
"eval_samples_per_second": 113.06,
|
| 233 |
+
"eval_steps_per_second": 4.046,
|
| 234 |
+
"step": 6579
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 10.0,
|
| 238 |
+
"step": 7310,
|
| 239 |
+
"train_exact_match": 56.64335664335665,
|
| 240 |
+
"train_f1": 66.49918737685148,
|
| 241 |
+
"train_runtime": 12.1264,
|
| 242 |
+
"train_samples_per_second": 113.224,
|
| 243 |
+
"train_steps_per_second": 4.123
|
| 244 |
+
},
|
| 245 |
+
{
|
| 246 |
+
"epoch": 10.0,
|
| 247 |
+
"grad_norm": 26.555368423461914,
|
| 248 |
+
"learning_rate": 0.0,
|
| 249 |
+
"loss": 1.6365,
|
| 250 |
+
"step": 7310
|
| 251 |
+
},
|
| 252 |
+
{
|
| 253 |
+
"epoch": 10.0,
|
| 254 |
+
"eval_exact_match": 24.9375,
|
| 255 |
+
"eval_f1": 36.31328557707662,
|
| 256 |
+
"eval_runtime": 40.4168,
|
| 257 |
+
"eval_samples_per_second": 113.394,
|
| 258 |
+
"eval_steps_per_second": 4.058,
|
| 259 |
+
"step": 7310
|
| 260 |
+
},
|
| 261 |
+
{
|
| 262 |
+
"epoch": 10.0,
|
| 263 |
+
"step": 7310,
|
| 264 |
+
"total_flos": 2.004675840340992e+16,
|
| 265 |
+
"train_loss": 2.4619675162715886,
|
| 266 |
+
"train_runtime": 2688.9862,
|
| 267 |
+
"train_samples_per_second": 76.081,
|
| 268 |
+
"train_steps_per_second": 2.718
|
| 269 |
+
}
|
| 270 |
+
],
|
| 271 |
+
"logging_steps": 500,
|
| 272 |
+
"max_steps": 7310,
|
| 273 |
+
"num_input_tokens_seen": 0,
|
| 274 |
+
"num_train_epochs": 10,
|
| 275 |
+
"save_steps": 500,
|
| 276 |
+
"total_flos": 2.004675840340992e+16,
|
| 277 |
+
"train_batch_size": 28,
|
| 278 |
+
"trial_name": null,
|
| 279 |
+
"trial_params": null
|
| 280 |
+
}
|