Upload trainer_state.json
Browse files- trainer_state.json +223 -0
trainer_state.json
ADDED
|
@@ -0,0 +1,223 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.8896821741133119,
|
| 3 |
+
"best_model_checkpoint": "model_ajustado/checkpoint-3258",
|
| 4 |
+
"epoch": 3.0,
|
| 5 |
+
"global_step": 3258,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.1,
|
| 12 |
+
"learning_rate": 9.801104972375691e-06,
|
| 13 |
+
"loss": 0.6336,
|
| 14 |
+
"step": 108
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.2,
|
| 18 |
+
"learning_rate": 9.602209944751382e-06,
|
| 19 |
+
"loss": 0.445,
|
| 20 |
+
"step": 216
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.3,
|
| 24 |
+
"learning_rate": 9.403314917127072e-06,
|
| 25 |
+
"loss": 0.4139,
|
| 26 |
+
"step": 324
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 0.4,
|
| 30 |
+
"learning_rate": 9.204419889502764e-06,
|
| 31 |
+
"loss": 0.3952,
|
| 32 |
+
"step": 432
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.5,
|
| 36 |
+
"learning_rate": 9.005524861878454e-06,
|
| 37 |
+
"loss": 0.374,
|
| 38 |
+
"step": 540
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.6,
|
| 42 |
+
"learning_rate": 8.806629834254145e-06,
|
| 43 |
+
"loss": 0.3476,
|
| 44 |
+
"step": 648
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.7,
|
| 48 |
+
"learning_rate": 8.607734806629835e-06,
|
| 49 |
+
"loss": 0.3492,
|
| 50 |
+
"step": 756
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 0.8,
|
| 54 |
+
"learning_rate": 8.408839779005525e-06,
|
| 55 |
+
"loss": 0.3277,
|
| 56 |
+
"step": 864
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 0.9,
|
| 60 |
+
"learning_rate": 8.209944751381216e-06,
|
| 61 |
+
"loss": 0.3479,
|
| 62 |
+
"step": 972
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 0.99,
|
| 66 |
+
"learning_rate": 8.011049723756906e-06,
|
| 67 |
+
"loss": 0.3279,
|
| 68 |
+
"step": 1080
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 1.0,
|
| 72 |
+
"eval_accuracy": 0.8813910640257946,
|
| 73 |
+
"eval_loss": 0.31820762157440186,
|
| 74 |
+
"eval_runtime": 24.2259,
|
| 75 |
+
"eval_samples_per_second": 179.23,
|
| 76 |
+
"eval_steps_per_second": 11.228,
|
| 77 |
+
"step": 1086
|
| 78 |
+
},
|
| 79 |
+
{
|
| 80 |
+
"epoch": 1.09,
|
| 81 |
+
"learning_rate": 7.812154696132598e-06,
|
| 82 |
+
"loss": 0.2958,
|
| 83 |
+
"step": 1188
|
| 84 |
+
},
|
| 85 |
+
{
|
| 86 |
+
"epoch": 1.19,
|
| 87 |
+
"learning_rate": 7.613259668508289e-06,
|
| 88 |
+
"loss": 0.2934,
|
| 89 |
+
"step": 1296
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 1.29,
|
| 93 |
+
"learning_rate": 7.414364640883979e-06,
|
| 94 |
+
"loss": 0.2638,
|
| 95 |
+
"step": 1404
|
| 96 |
+
},
|
| 97 |
+
{
|
| 98 |
+
"epoch": 1.39,
|
| 99 |
+
"learning_rate": 7.215469613259669e-06,
|
| 100 |
+
"loss": 0.2794,
|
| 101 |
+
"step": 1512
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 1.49,
|
| 105 |
+
"learning_rate": 7.01657458563536e-06,
|
| 106 |
+
"loss": 0.258,
|
| 107 |
+
"step": 1620
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 1.59,
|
| 111 |
+
"learning_rate": 6.81767955801105e-06,
|
| 112 |
+
"loss": 0.2662,
|
| 113 |
+
"step": 1728
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"epoch": 1.69,
|
| 117 |
+
"learning_rate": 6.6187845303867405e-06,
|
| 118 |
+
"loss": 0.3038,
|
| 119 |
+
"step": 1836
|
| 120 |
+
},
|
| 121 |
+
{
|
| 122 |
+
"epoch": 1.79,
|
| 123 |
+
"learning_rate": 6.419889502762432e-06,
|
| 124 |
+
"loss": 0.2594,
|
| 125 |
+
"step": 1944
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"epoch": 1.89,
|
| 129 |
+
"learning_rate": 6.220994475138122e-06,
|
| 130 |
+
"loss": 0.2475,
|
| 131 |
+
"step": 2052
|
| 132 |
+
},
|
| 133 |
+
{
|
| 134 |
+
"epoch": 1.99,
|
| 135 |
+
"learning_rate": 6.0220994475138124e-06,
|
| 136 |
+
"loss": 0.2922,
|
| 137 |
+
"step": 2160
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 2.0,
|
| 141 |
+
"eval_accuracy": 0.8871487793643482,
|
| 142 |
+
"eval_loss": 0.2970804274082184,
|
| 143 |
+
"eval_runtime": 24.4656,
|
| 144 |
+
"eval_samples_per_second": 177.473,
|
| 145 |
+
"eval_steps_per_second": 11.118,
|
| 146 |
+
"step": 2172
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 2.09,
|
| 150 |
+
"learning_rate": 5.823204419889504e-06,
|
| 151 |
+
"loss": 0.2174,
|
| 152 |
+
"step": 2268
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 2.19,
|
| 156 |
+
"learning_rate": 5.624309392265194e-06,
|
| 157 |
+
"loss": 0.2346,
|
| 158 |
+
"step": 2376
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 2.29,
|
| 162 |
+
"learning_rate": 5.425414364640884e-06,
|
| 163 |
+
"loss": 0.2061,
|
| 164 |
+
"step": 2484
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 2.39,
|
| 168 |
+
"learning_rate": 5.226519337016575e-06,
|
| 169 |
+
"loss": 0.2441,
|
| 170 |
+
"step": 2592
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 2.49,
|
| 174 |
+
"learning_rate": 5.027624309392266e-06,
|
| 175 |
+
"loss": 0.2284,
|
| 176 |
+
"step": 2700
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 2.59,
|
| 180 |
+
"learning_rate": 4.828729281767956e-06,
|
| 181 |
+
"loss": 0.2147,
|
| 182 |
+
"step": 2808
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 2.69,
|
| 186 |
+
"learning_rate": 4.629834254143647e-06,
|
| 187 |
+
"loss": 0.2296,
|
| 188 |
+
"step": 2916
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 2.78,
|
| 192 |
+
"learning_rate": 4.430939226519337e-06,
|
| 193 |
+
"loss": 0.2489,
|
| 194 |
+
"step": 3024
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 2.88,
|
| 198 |
+
"learning_rate": 4.2320441988950274e-06,
|
| 199 |
+
"loss": 0.2199,
|
| 200 |
+
"step": 3132
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 2.98,
|
| 204 |
+
"learning_rate": 4.033149171270719e-06,
|
| 205 |
+
"loss": 0.2178,
|
| 206 |
+
"step": 3240
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 3.0,
|
| 210 |
+
"eval_accuracy": 0.8896821741133119,
|
| 211 |
+
"eval_loss": 0.3544089198112488,
|
| 212 |
+
"eval_runtime": 24.4186,
|
| 213 |
+
"eval_samples_per_second": 177.816,
|
| 214 |
+
"eval_steps_per_second": 11.139,
|
| 215 |
+
"step": 3258
|
| 216 |
+
}
|
| 217 |
+
],
|
| 218 |
+
"max_steps": 5430,
|
| 219 |
+
"num_train_epochs": 5,
|
| 220 |
+
"total_flos": 2646085391685036.0,
|
| 221 |
+
"trial_name": null,
|
| 222 |
+
"trial_params": null
|
| 223 |
+
}
|