Commit ·
ed08ff8
1
Parent(s): 8e15b9f
Upload trainer_state.json with huggingface_hub
Browse files- trainer_state.json +292 -0
trainer_state.json
ADDED
|
@@ -0,0 +1,292 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.00041871442226693034,
|
| 3 |
+
"best_model_checkpoint": "./models/results_the_naughtyformer_977/checkpoint-225",
|
| 4 |
+
"epoch": 0.20821283979178715,
|
| 5 |
+
"global_step": 225,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.01,
|
| 12 |
+
"learning_rate": 4.000000000000001e-06,
|
| 13 |
+
"loss": 0.6624,
|
| 14 |
+
"step": 8
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.01,
|
| 18 |
+
"learning_rate": 8.000000000000001e-06,
|
| 19 |
+
"loss": 0.604,
|
| 20 |
+
"step": 16
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.02,
|
| 24 |
+
"learning_rate": 1.2e-05,
|
| 25 |
+
"loss": 0.4918,
|
| 26 |
+
"step": 24
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 0.02,
|
| 30 |
+
"eval_accuracy": 0.8941295546558704,
|
| 31 |
+
"eval_f1": 0.9441060168857539,
|
| 32 |
+
"eval_loss": 0.3381046652793884,
|
| 33 |
+
"eval_precision": 0.8941295546558704,
|
| 34 |
+
"eval_recall": 1.0,
|
| 35 |
+
"eval_runtime": 63.6445,
|
| 36 |
+
"eval_samples_per_second": 155.237,
|
| 37 |
+
"eval_steps_per_second": 19.405,
|
| 38 |
+
"step": 25
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.03,
|
| 42 |
+
"learning_rate": 1.6000000000000003e-05,
|
| 43 |
+
"loss": 0.2579,
|
| 44 |
+
"step": 32
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.04,
|
| 48 |
+
"learning_rate": 2e-05,
|
| 49 |
+
"loss": 0.1325,
|
| 50 |
+
"step": 40
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 0.04,
|
| 54 |
+
"learning_rate": 2.4e-05,
|
| 55 |
+
"loss": 0.0186,
|
| 56 |
+
"step": 48
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 0.05,
|
| 60 |
+
"eval_accuracy": 0.9995951417004049,
|
| 61 |
+
"eval_f1": 0.9997736532367587,
|
| 62 |
+
"eval_loss": 0.002655914518982172,
|
| 63 |
+
"eval_precision": 0.9995474089160443,
|
| 64 |
+
"eval_recall": 1.0,
|
| 65 |
+
"eval_runtime": 64.7489,
|
| 66 |
+
"eval_samples_per_second": 152.589,
|
| 67 |
+
"eval_steps_per_second": 19.074,
|
| 68 |
+
"step": 50
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 0.05,
|
| 72 |
+
"learning_rate": 2.8000000000000003e-05,
|
| 73 |
+
"loss": 0.0126,
|
| 74 |
+
"step": 56
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 0.06,
|
| 78 |
+
"learning_rate": 3.2000000000000005e-05,
|
| 79 |
+
"loss": 0.0008,
|
| 80 |
+
"step": 64
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.07,
|
| 84 |
+
"learning_rate": 3.6e-05,
|
| 85 |
+
"loss": 0.0015,
|
| 86 |
+
"step": 72
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.07,
|
| 90 |
+
"eval_accuracy": 0.9997975708502024,
|
| 91 |
+
"eval_f1": 0.9998868138087154,
|
| 92 |
+
"eval_loss": 0.0015935683622956276,
|
| 93 |
+
"eval_precision": 0.9997736532367587,
|
| 94 |
+
"eval_recall": 1.0,
|
| 95 |
+
"eval_runtime": 63.6605,
|
| 96 |
+
"eval_samples_per_second": 155.198,
|
| 97 |
+
"eval_steps_per_second": 19.4,
|
| 98 |
+
"step": 75
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 0.07,
|
| 102 |
+
"learning_rate": 4e-05,
|
| 103 |
+
"loss": 0.0156,
|
| 104 |
+
"step": 80
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 0.08,
|
| 108 |
+
"learning_rate": 4.4000000000000006e-05,
|
| 109 |
+
"loss": 0.0155,
|
| 110 |
+
"step": 88
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 0.09,
|
| 114 |
+
"learning_rate": 4.8e-05,
|
| 115 |
+
"loss": 0.0004,
|
| 116 |
+
"step": 96
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 0.09,
|
| 120 |
+
"eval_accuracy": 0.9997975708502024,
|
| 121 |
+
"eval_f1": 0.9998868138087154,
|
| 122 |
+
"eval_loss": 0.0007984100375324488,
|
| 123 |
+
"eval_precision": 0.9997736532367587,
|
| 124 |
+
"eval_recall": 1.0,
|
| 125 |
+
"eval_runtime": 63.3854,
|
| 126 |
+
"eval_samples_per_second": 155.872,
|
| 127 |
+
"eval_steps_per_second": 19.484,
|
| 128 |
+
"step": 100
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.1,
|
| 132 |
+
"learning_rate": 4.979591836734694e-05,
|
| 133 |
+
"loss": 0.0004,
|
| 134 |
+
"step": 104
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 0.1,
|
| 138 |
+
"learning_rate": 4.938775510204082e-05,
|
| 139 |
+
"loss": 0.0004,
|
| 140 |
+
"step": 112
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 0.11,
|
| 144 |
+
"learning_rate": 4.903061224489796e-05,
|
| 145 |
+
"loss": 0.0118,
|
| 146 |
+
"step": 120
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 0.12,
|
| 150 |
+
"eval_accuracy": 0.9996963562753036,
|
| 151 |
+
"eval_f1": 0.9998302303208647,
|
| 152 |
+
"eval_loss": 0.0019073737785220146,
|
| 153 |
+
"eval_precision": 0.9996605182754328,
|
| 154 |
+
"eval_recall": 1.0,
|
| 155 |
+
"eval_runtime": 63.5966,
|
| 156 |
+
"eval_samples_per_second": 155.354,
|
| 157 |
+
"eval_steps_per_second": 19.419,
|
| 158 |
+
"step": 125
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 0.12,
|
| 162 |
+
"learning_rate": 4.862244897959184e-05,
|
| 163 |
+
"loss": 0.0002,
|
| 164 |
+
"step": 128
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.13,
|
| 168 |
+
"learning_rate": 4.8214285714285716e-05,
|
| 169 |
+
"loss": 0.001,
|
| 170 |
+
"step": 136
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.13,
|
| 174 |
+
"learning_rate": 4.7806122448979595e-05,
|
| 175 |
+
"loss": 0.0113,
|
| 176 |
+
"step": 144
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 0.14,
|
| 180 |
+
"eval_accuracy": 0.9998987854251012,
|
| 181 |
+
"eval_f1": 0.9999434037013979,
|
| 182 |
+
"eval_loss": 0.0006314449128694832,
|
| 183 |
+
"eval_precision": 0.9998868138087154,
|
| 184 |
+
"eval_recall": 1.0,
|
| 185 |
+
"eval_runtime": 63.6772,
|
| 186 |
+
"eval_samples_per_second": 155.158,
|
| 187 |
+
"eval_steps_per_second": 19.395,
|
| 188 |
+
"step": 150
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 0.14,
|
| 192 |
+
"learning_rate": 4.7397959183673474e-05,
|
| 193 |
+
"loss": 0.0001,
|
| 194 |
+
"step": 152
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 0.15,
|
| 198 |
+
"learning_rate": 4.698979591836735e-05,
|
| 199 |
+
"loss": 0.0001,
|
| 200 |
+
"step": 160
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 0.16,
|
| 204 |
+
"learning_rate": 4.6581632653061226e-05,
|
| 205 |
+
"loss": 0.0149,
|
| 206 |
+
"step": 168
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.16,
|
| 210 |
+
"eval_accuracy": 0.9998987854251012,
|
| 211 |
+
"eval_f1": 0.9999433972943906,
|
| 212 |
+
"eval_loss": 0.0006301466492004693,
|
| 213 |
+
"eval_precision": 1.0,
|
| 214 |
+
"eval_recall": 0.9998868009961512,
|
| 215 |
+
"eval_runtime": 63.7491,
|
| 216 |
+
"eval_samples_per_second": 154.982,
|
| 217 |
+
"eval_steps_per_second": 19.373,
|
| 218 |
+
"step": 175
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 0.16,
|
| 222 |
+
"learning_rate": 4.6173469387755105e-05,
|
| 223 |
+
"loss": 0.0017,
|
| 224 |
+
"step": 176
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 0.17,
|
| 228 |
+
"learning_rate": 4.5765306122448984e-05,
|
| 229 |
+
"loss": 0.0137,
|
| 230 |
+
"step": 184
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 0.18,
|
| 234 |
+
"learning_rate": 4.5357142857142856e-05,
|
| 235 |
+
"loss": 0.0001,
|
| 236 |
+
"step": 192
|
| 237 |
+
},
|
| 238 |
+
{
|
| 239 |
+
"epoch": 0.19,
|
| 240 |
+
"learning_rate": 4.4948979591836735e-05,
|
| 241 |
+
"loss": 0.0003,
|
| 242 |
+
"step": 200
|
| 243 |
+
},
|
| 244 |
+
{
|
| 245 |
+
"epoch": 0.19,
|
| 246 |
+
"eval_accuracy": 0.9997975708502024,
|
| 247 |
+
"eval_f1": 0.9998867881806861,
|
| 248 |
+
"eval_loss": 0.0011095332447439432,
|
| 249 |
+
"eval_precision": 1.0,
|
| 250 |
+
"eval_recall": 0.9997736019923025,
|
| 251 |
+
"eval_runtime": 63.9897,
|
| 252 |
+
"eval_samples_per_second": 154.4,
|
| 253 |
+
"eval_steps_per_second": 19.3,
|
| 254 |
+
"step": 200
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.19,
|
| 258 |
+
"learning_rate": 4.4540816326530614e-05,
|
| 259 |
+
"loss": 0.0001,
|
| 260 |
+
"step": 208
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 0.2,
|
| 264 |
+
"learning_rate": 4.4132653061224493e-05,
|
| 265 |
+
"loss": 0.0001,
|
| 266 |
+
"step": 216
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 0.21,
|
| 270 |
+
"learning_rate": 4.372448979591837e-05,
|
| 271 |
+
"loss": 0.0001,
|
| 272 |
+
"step": 224
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 0.21,
|
| 276 |
+
"eval_accuracy": 0.9998987854251012,
|
| 277 |
+
"eval_f1": 0.9999434037013979,
|
| 278 |
+
"eval_loss": 0.00041871442226693034,
|
| 279 |
+
"eval_precision": 0.9998868138087154,
|
| 280 |
+
"eval_recall": 1.0,
|
| 281 |
+
"eval_runtime": 63.1331,
|
| 282 |
+
"eval_samples_per_second": 156.495,
|
| 283 |
+
"eval_steps_per_second": 19.562,
|
| 284 |
+
"step": 225
|
| 285 |
+
}
|
| 286 |
+
],
|
| 287 |
+
"max_steps": 1080,
|
| 288 |
+
"num_train_epochs": 1,
|
| 289 |
+
"total_flos": 3788799197184000.0,
|
| 290 |
+
"trial_name": null,
|
| 291 |
+
"trial_params": null
|
| 292 |
+
}
|