Upload trainer_state.json
Browse files- trainer_state.json +226 -0
trainer_state.json
ADDED
|
@@ -0,0 +1,226 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 10.0,
|
| 5 |
+
"global_step": 17610,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 1.0,
|
| 12 |
+
"learning_rate": 1.8e-05,
|
| 13 |
+
"loss": 0.0167,
|
| 14 |
+
"step": 1761
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 1.0,
|
| 18 |
+
"eval_EG_f1": 0.9978260869565218,
|
| 19 |
+
"eval_ET_f1": 0.9987063389391979,
|
| 20 |
+
"eval_TE_f1": 0.9969821673525378,
|
| 21 |
+
"eval_loss": 0.005893823690712452,
|
| 22 |
+
"eval_overall_accuracy": 0.998933354402876,
|
| 23 |
+
"eval_overall_f1": 0.9976327358809605,
|
| 24 |
+
"eval_overall_precision": 0.9975202885482417,
|
| 25 |
+
"eval_overall_recall": 0.9977452085682075,
|
| 26 |
+
"eval_runtime": 59.0341,
|
| 27 |
+
"eval_samples_per_second": 53.037,
|
| 28 |
+
"eval_steps_per_second": 3.32,
|
| 29 |
+
"step": 1761
|
| 30 |
+
},
|
| 31 |
+
{
|
| 32 |
+
"epoch": 2.0,
|
| 33 |
+
"learning_rate": 1.6000000000000003e-05,
|
| 34 |
+
"loss": 0.0048,
|
| 35 |
+
"step": 3522
|
| 36 |
+
},
|
| 37 |
+
{
|
| 38 |
+
"epoch": 2.0,
|
| 39 |
+
"eval_EG_f1": 0.9950846531949754,
|
| 40 |
+
"eval_ET_f1": 0.9974059662775615,
|
| 41 |
+
"eval_TE_f1": 0.9942763695829926,
|
| 42 |
+
"eval_loss": 0.021830907091498375,
|
| 43 |
+
"eval_overall_accuracy": 0.9975506656658634,
|
| 44 |
+
"eval_overall_f1": 0.9951538374845036,
|
| 45 |
+
"eval_overall_precision": 0.9948174853537629,
|
| 46 |
+
"eval_overall_recall": 0.9954904171364148,
|
| 47 |
+
"eval_runtime": 58.5084,
|
| 48 |
+
"eval_samples_per_second": 53.514,
|
| 49 |
+
"eval_steps_per_second": 3.35,
|
| 50 |
+
"step": 3522
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 3.0,
|
| 54 |
+
"learning_rate": 1.4e-05,
|
| 55 |
+
"loss": 0.003,
|
| 56 |
+
"step": 5283
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 3.0,
|
| 60 |
+
"eval_EG_f1": 0.9939989088925258,
|
| 61 |
+
"eval_ET_f1": 0.9987063389391979,
|
| 62 |
+
"eval_TE_f1": 0.9942638623326959,
|
| 63 |
+
"eval_loss": 0.022889936342835426,
|
| 64 |
+
"eval_overall_accuracy": 0.9971556117410026,
|
| 65 |
+
"eval_overall_f1": 0.9949284345768061,
|
| 66 |
+
"eval_overall_precision": 0.9945921586300135,
|
| 67 |
+
"eval_overall_recall": 0.9952649379932357,
|
| 68 |
+
"eval_runtime": 57.8629,
|
| 69 |
+
"eval_samples_per_second": 54.111,
|
| 70 |
+
"eval_steps_per_second": 3.387,
|
| 71 |
+
"step": 5283
|
| 72 |
+
},
|
| 73 |
+
{
|
| 74 |
+
"epoch": 4.0,
|
| 75 |
+
"learning_rate": 1.2e-05,
|
| 76 |
+
"loss": 0.002,
|
| 77 |
+
"step": 7044
|
| 78 |
+
},
|
| 79 |
+
{
|
| 80 |
+
"epoch": 4.0,
|
| 81 |
+
"eval_EG_f1": 0.9961977186311787,
|
| 82 |
+
"eval_ET_f1": 0.9987046632124352,
|
| 83 |
+
"eval_TE_f1": 0.9958847736625515,
|
| 84 |
+
"eval_loss": 0.017046082764863968,
|
| 85 |
+
"eval_overall_accuracy": 0.9982222573381266,
|
| 86 |
+
"eval_overall_f1": 0.9965054672528464,
|
| 87 |
+
"eval_overall_precision": 0.9963931469792606,
|
| 88 |
+
"eval_overall_recall": 0.9966178128523112,
|
| 89 |
+
"eval_runtime": 57.0371,
|
| 90 |
+
"eval_samples_per_second": 54.894,
|
| 91 |
+
"eval_steps_per_second": 3.436,
|
| 92 |
+
"step": 7044
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 5.0,
|
| 96 |
+
"learning_rate": 1e-05,
|
| 97 |
+
"loss": 0.0013,
|
| 98 |
+
"step": 8805
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 5.0,
|
| 102 |
+
"eval_EG_f1": 0.9970068027210883,
|
| 103 |
+
"eval_ET_f1": 0.9987046632124352,
|
| 104 |
+
"eval_TE_f1": 0.9961643835616438,
|
| 105 |
+
"eval_loss": 0.011512575671076775,
|
| 106 |
+
"eval_overall_accuracy": 0.9984987950855292,
|
| 107 |
+
"eval_overall_f1": 0.9969556883526891,
|
| 108 |
+
"eval_overall_precision": 0.9970681100586378,
|
| 109 |
+
"eval_overall_recall": 0.9968432919954904,
|
| 110 |
+
"eval_runtime": 57.8352,
|
| 111 |
+
"eval_samples_per_second": 54.137,
|
| 112 |
+
"eval_steps_per_second": 3.389,
|
| 113 |
+
"step": 8805
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"epoch": 6.0,
|
| 117 |
+
"learning_rate": 8.000000000000001e-06,
|
| 118 |
+
"loss": 0.0015,
|
| 119 |
+
"step": 10566
|
| 120 |
+
},
|
| 121 |
+
{
|
| 122 |
+
"epoch": 6.0,
|
| 123 |
+
"eval_EG_f1": 0.9948271167982575,
|
| 124 |
+
"eval_ET_f1": 0.9980557355800389,
|
| 125 |
+
"eval_TE_f1": 0.9948073244055754,
|
| 126 |
+
"eval_loss": 0.01284080371260643,
|
| 127 |
+
"eval_overall_accuracy": 0.9979457195907241,
|
| 128 |
+
"eval_overall_f1": 0.9953802816901409,
|
| 129 |
+
"eval_overall_precision": 0.9948198198198198,
|
| 130 |
+
"eval_overall_recall": 0.9959413754227734,
|
| 131 |
+
"eval_runtime": 57.2307,
|
| 132 |
+
"eval_samples_per_second": 54.708,
|
| 133 |
+
"eval_steps_per_second": 3.425,
|
| 134 |
+
"step": 10566
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 7.0,
|
| 138 |
+
"learning_rate": 6e-06,
|
| 139 |
+
"loss": 0.0003,
|
| 140 |
+
"step": 12327
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 7.0,
|
| 144 |
+
"eval_EG_f1": 0.9953615279672577,
|
| 145 |
+
"eval_ET_f1": 0.9993527508090615,
|
| 146 |
+
"eval_TE_f1": 0.9959049959049959,
|
| 147 |
+
"eval_loss": 0.021282300353050232,
|
| 148 |
+
"eval_overall_accuracy": 0.9981037411606685,
|
| 149 |
+
"eval_overall_f1": 0.9962808520229911,
|
| 150 |
+
"eval_overall_precision": 0.9959441189725101,
|
| 151 |
+
"eval_overall_recall": 0.9966178128523112,
|
| 152 |
+
"eval_runtime": 57.2264,
|
| 153 |
+
"eval_samples_per_second": 54.713,
|
| 154 |
+
"eval_steps_per_second": 3.425,
|
| 155 |
+
"step": 12327
|
| 156 |
+
},
|
| 157 |
+
{
|
| 158 |
+
"epoch": 8.0,
|
| 159 |
+
"learning_rate": 4.000000000000001e-06,
|
| 160 |
+
"loss": 0.0005,
|
| 161 |
+
"step": 14088
|
| 162 |
+
},
|
| 163 |
+
{
|
| 164 |
+
"epoch": 8.0,
|
| 165 |
+
"eval_EG_f1": 0.9961852861035423,
|
| 166 |
+
"eval_ET_f1": 0.9993527508090615,
|
| 167 |
+
"eval_TE_f1": 0.9961706783369804,
|
| 168 |
+
"eval_loss": 0.01351018063724041,
|
| 169 |
+
"eval_overall_accuracy": 0.9983802789080709,
|
| 170 |
+
"eval_overall_f1": 0.9967309209784693,
|
| 171 |
+
"eval_overall_precision": 0.9966185752930569,
|
| 172 |
+
"eval_overall_recall": 0.9968432919954904,
|
| 173 |
+
"eval_runtime": 57.0797,
|
| 174 |
+
"eval_samples_per_second": 54.853,
|
| 175 |
+
"eval_steps_per_second": 3.434,
|
| 176 |
+
"step": 14088
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 9.0,
|
| 180 |
+
"learning_rate": 2.0000000000000003e-06,
|
| 181 |
+
"loss": 0.0002,
|
| 182 |
+
"step": 15849
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 9.0,
|
| 186 |
+
"eval_EG_f1": 0.9972781709308656,
|
| 187 |
+
"eval_ET_f1": 0.9993527508090615,
|
| 188 |
+
"eval_TE_f1": 0.9969871268145714,
|
| 189 |
+
"eval_loss": 0.014631124213337898,
|
| 190 |
+
"eval_overall_accuracy": 0.9986568166554735,
|
| 191 |
+
"eval_overall_f1": 0.9975197294250282,
|
| 192 |
+
"eval_overall_precision": 0.9975197294250282,
|
| 193 |
+
"eval_overall_recall": 0.9975197294250282,
|
| 194 |
+
"eval_runtime": 57.0354,
|
| 195 |
+
"eval_samples_per_second": 54.896,
|
| 196 |
+
"eval_steps_per_second": 3.436,
|
| 197 |
+
"step": 15849
|
| 198 |
+
},
|
| 199 |
+
{
|
| 200 |
+
"epoch": 10.0,
|
| 201 |
+
"learning_rate": 0.0,
|
| 202 |
+
"loss": 0.0001,
|
| 203 |
+
"step": 17610
|
| 204 |
+
},
|
| 205 |
+
{
|
| 206 |
+
"epoch": 10.0,
|
| 207 |
+
"eval_EG_f1": 0.9970051728832018,
|
| 208 |
+
"eval_ET_f1": 0.9993527508090615,
|
| 209 |
+
"eval_TE_f1": 0.9967141292442497,
|
| 210 |
+
"eval_loss": 0.014984946697950363,
|
| 211 |
+
"eval_overall_accuracy": 0.9984987950855292,
|
| 212 |
+
"eval_overall_f1": 0.9972942502818489,
|
| 213 |
+
"eval_overall_precision": 0.9972942502818489,
|
| 214 |
+
"eval_overall_recall": 0.9972942502818489,
|
| 215 |
+
"eval_runtime": 56.8354,
|
| 216 |
+
"eval_samples_per_second": 55.089,
|
| 217 |
+
"eval_steps_per_second": 3.449,
|
| 218 |
+
"step": 17610
|
| 219 |
+
}
|
| 220 |
+
],
|
| 221 |
+
"max_steps": 17610,
|
| 222 |
+
"num_train_epochs": 10,
|
| 223 |
+
"total_flos": 3085811947058688.0,
|
| 224 |
+
"trial_name": null,
|
| 225 |
+
"trial_params": null
|
| 226 |
+
}
|