Create trainer_state.json
Browse files- trainer_state.json +136 -0
trainer_state.json
ADDED
|
@@ -0,0 +1,136 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 1.4365631341934204,
|
| 3 |
+
"best_model_checkpoint": "test_/checkpoint-1128",
|
| 4 |
+
"epoch": 6.0,
|
| 5 |
+
"global_step": 1128,
|
| 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": 0.001,
|
| 13 |
+
"loss": 1.6043,
|
| 14 |
+
"step": 188
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 1.0,
|
| 18 |
+
"eval_loss": 1.5687600374221802,
|
| 19 |
+
"eval_micro_f1": 0.24933333333333332,
|
| 20 |
+
"eval_runtime": 51.1135,
|
| 21 |
+
"eval_samples_per_second": 58.693,
|
| 22 |
+
"eval_steps_per_second": 0.47,
|
| 23 |
+
"eval_unavg_f1_rating1": 0.33407572383073497,
|
| 24 |
+
"eval_unavg_f1_rating2": 0.0,
|
| 25 |
+
"eval_unavg_f1_rating3": 0.0,
|
| 26 |
+
"eval_unavg_f1_rating4": 0.36225728155339804,
|
| 27 |
+
"eval_unavg_f1_rating5": 0.0034305317324185253,
|
| 28 |
+
"step": 188
|
| 29 |
+
},
|
| 30 |
+
{
|
| 31 |
+
"epoch": 2.0,
|
| 32 |
+
"learning_rate": 0.001,
|
| 33 |
+
"loss": 1.5576,
|
| 34 |
+
"step": 376
|
| 35 |
+
},
|
| 36 |
+
{
|
| 37 |
+
"epoch": 2.0,
|
| 38 |
+
"eval_loss": 1.5380204916000366,
|
| 39 |
+
"eval_micro_f1": 0.29033333333333333,
|
| 40 |
+
"eval_runtime": 51.1315,
|
| 41 |
+
"eval_samples_per_second": 58.672,
|
| 42 |
+
"eval_steps_per_second": 0.469,
|
| 43 |
+
"eval_unavg_f1_rating1": 0.2691823899371069,
|
| 44 |
+
"eval_unavg_f1_rating2": 0.20084566596194503,
|
| 45 |
+
"eval_unavg_f1_rating3": 0.3557312252964427,
|
| 46 |
+
"eval_unavg_f1_rating4": 0.0,
|
| 47 |
+
"eval_unavg_f1_rating5": 0.4060676779463244,
|
| 48 |
+
"step": 376
|
| 49 |
+
},
|
| 50 |
+
{
|
| 51 |
+
"epoch": 3.0,
|
| 52 |
+
"learning_rate": 0.001,
|
| 53 |
+
"loss": 1.5317,
|
| 54 |
+
"step": 564
|
| 55 |
+
},
|
| 56 |
+
{
|
| 57 |
+
"epoch": 3.0,
|
| 58 |
+
"eval_loss": 1.4792137145996094,
|
| 59 |
+
"eval_micro_f1": 0.362,
|
| 60 |
+
"eval_runtime": 51.3091,
|
| 61 |
+
"eval_samples_per_second": 58.469,
|
| 62 |
+
"eval_steps_per_second": 0.468,
|
| 63 |
+
"eval_unavg_f1_rating1": 0.4780600461893765,
|
| 64 |
+
"eval_unavg_f1_rating2": 0.26759339704604695,
|
| 65 |
+
"eval_unavg_f1_rating3": 0.25631431244153413,
|
| 66 |
+
"eval_unavg_f1_rating4": 0.07207207207207207,
|
| 67 |
+
"eval_unavg_f1_rating5": 0.5166425470332849,
|
| 68 |
+
"step": 564
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 4.0,
|
| 72 |
+
"learning_rate": 0.001,
|
| 73 |
+
"loss": 1.5124,
|
| 74 |
+
"step": 752
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 4.0,
|
| 78 |
+
"eval_loss": 1.491315245628357,
|
| 79 |
+
"eval_micro_f1": 0.3546666666666667,
|
| 80 |
+
"eval_runtime": 51.1691,
|
| 81 |
+
"eval_samples_per_second": 58.629,
|
| 82 |
+
"eval_steps_per_second": 0.469,
|
| 83 |
+
"eval_unavg_f1_rating1": 0.4857908847184987,
|
| 84 |
+
"eval_unavg_f1_rating2": 0.0,
|
| 85 |
+
"eval_unavg_f1_rating3": 0.3111111111111111,
|
| 86 |
+
"eval_unavg_f1_rating4": 0.016,
|
| 87 |
+
"eval_unavg_f1_rating5": 0.5133757961783438,
|
| 88 |
+
"step": 752
|
| 89 |
+
},
|
| 90 |
+
{
|
| 91 |
+
"epoch": 5.0,
|
| 92 |
+
"learning_rate": 0.001,
|
| 93 |
+
"loss": 1.4943,
|
| 94 |
+
"step": 940
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 5.0,
|
| 98 |
+
"eval_loss": 1.453855276107788,
|
| 99 |
+
"eval_micro_f1": 0.349,
|
| 100 |
+
"eval_runtime": 51.2727,
|
| 101 |
+
"eval_samples_per_second": 58.511,
|
| 102 |
+
"eval_steps_per_second": 0.468,
|
| 103 |
+
"eval_unavg_f1_rating1": 0.3043478260869565,
|
| 104 |
+
"eval_unavg_f1_rating2": 0.2603773584905661,
|
| 105 |
+
"eval_unavg_f1_rating3": 0.35506094329623744,
|
| 106 |
+
"eval_unavg_f1_rating4": 0.2099056603773585,
|
| 107 |
+
"eval_unavg_f1_rating5": 0.5144061841180605,
|
| 108 |
+
"step": 940
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 6.0,
|
| 112 |
+
"learning_rate": 0.001,
|
| 113 |
+
"loss": 1.4977,
|
| 114 |
+
"step": 1128
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 6.0,
|
| 118 |
+
"eval_loss": 1.4365631341934204,
|
| 119 |
+
"eval_micro_f1": 0.36499999999999994,
|
| 120 |
+
"eval_runtime": 51.4954,
|
| 121 |
+
"eval_samples_per_second": 58.258,
|
| 122 |
+
"eval_steps_per_second": 0.466,
|
| 123 |
+
"eval_unavg_f1_rating1": 0.46589018302828616,
|
| 124 |
+
"eval_unavg_f1_rating2": 0.027439024390243903,
|
| 125 |
+
"eval_unavg_f1_rating3": 0.36269963936115407,
|
| 126 |
+
"eval_unavg_f1_rating4": 0.1606425702811245,
|
| 127 |
+
"eval_unavg_f1_rating5": 0.5419532324621733,
|
| 128 |
+
"step": 1128
|
| 129 |
+
}
|
| 130 |
+
],
|
| 131 |
+
"max_steps": 1880,
|
| 132 |
+
"num_train_epochs": 10,
|
| 133 |
+
"total_flos": 1.7679088885879104e+16,
|
| 134 |
+
"trial_name": null,
|
| 135 |
+
"trial_params": null
|
| 136 |
+
}
|