Upload trainer_state.json
Browse files- trainer_state.json +133 -0
trainer_state.json
ADDED
|
@@ -0,0 +1,133 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 100.0,
|
| 5 |
+
"global_step": 6300,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 15.87,
|
| 12 |
+
"learning_rate": 3.3650793650793656e-05,
|
| 13 |
+
"loss": 0.067,
|
| 14 |
+
"step": 1000
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 15.87,
|
| 18 |
+
"eval_accuracy": 0.9120488561846123,
|
| 19 |
+
"eval_f1": 0.5542691751085383,
|
| 20 |
+
"eval_loss": 0.5247976183891296,
|
| 21 |
+
"eval_precision": 0.5292491939198526,
|
| 22 |
+
"eval_recall": 0.5817721518987342,
|
| 23 |
+
"eval_runtime": 6.9983,
|
| 24 |
+
"eval_samples_per_second": 71.445,
|
| 25 |
+
"eval_steps_per_second": 9.002,
|
| 26 |
+
"step": 1000
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 31.75,
|
| 30 |
+
"learning_rate": 2.7301587301587305e-05,
|
| 31 |
+
"loss": 0.0049,
|
| 32 |
+
"step": 2000
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 31.75,
|
| 36 |
+
"eval_accuracy": 0.9141012163988587,
|
| 37 |
+
"eval_f1": 0.5845740785940932,
|
| 38 |
+
"eval_loss": 0.6428591012954712,
|
| 39 |
+
"eval_precision": 0.5643261074458058,
|
| 40 |
+
"eval_recall": 0.6063291139240506,
|
| 41 |
+
"eval_runtime": 7.0179,
|
| 42 |
+
"eval_samples_per_second": 71.246,
|
| 43 |
+
"eval_steps_per_second": 8.977,
|
| 44 |
+
"step": 2000
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 47.62,
|
| 48 |
+
"learning_rate": 2.0952380952380954e-05,
|
| 49 |
+
"loss": 0.0022,
|
| 50 |
+
"step": 3000
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 47.62,
|
| 54 |
+
"eval_accuracy": 0.9139510437002553,
|
| 55 |
+
"eval_f1": 0.5792592592592594,
|
| 56 |
+
"eval_loss": 0.6684653162956238,
|
| 57 |
+
"eval_precision": 0.5653012048192771,
|
| 58 |
+
"eval_recall": 0.5939240506329114,
|
| 59 |
+
"eval_runtime": 6.9798,
|
| 60 |
+
"eval_samples_per_second": 71.636,
|
| 61 |
+
"eval_steps_per_second": 9.026,
|
| 62 |
+
"step": 3000
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 63.49,
|
| 66 |
+
"learning_rate": 1.4603174603174603e-05,
|
| 67 |
+
"loss": 0.0013,
|
| 68 |
+
"step": 4000
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 63.49,
|
| 72 |
+
"eval_accuracy": 0.9137007558692496,
|
| 73 |
+
"eval_f1": 0.5624591236102028,
|
| 74 |
+
"eval_loss": 0.6453785300254822,
|
| 75 |
+
"eval_precision": 0.5818673883626523,
|
| 76 |
+
"eval_recall": 0.5443037974683544,
|
| 77 |
+
"eval_runtime": 6.9826,
|
| 78 |
+
"eval_samples_per_second": 71.607,
|
| 79 |
+
"eval_steps_per_second": 9.022,
|
| 80 |
+
"step": 4000
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 79.37,
|
| 84 |
+
"learning_rate": 8.253968253968254e-06,
|
| 85 |
+
"loss": 0.0004,
|
| 86 |
+
"step": 5000
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 79.37,
|
| 90 |
+
"eval_accuracy": 0.9153025979876859,
|
| 91 |
+
"eval_f1": 0.5703675246510438,
|
| 92 |
+
"eval_loss": 0.7095718383789062,
|
| 93 |
+
"eval_precision": 0.5770925110132159,
|
| 94 |
+
"eval_recall": 0.5637974683544303,
|
| 95 |
+
"eval_runtime": 7.0736,
|
| 96 |
+
"eval_samples_per_second": 70.685,
|
| 97 |
+
"eval_steps_per_second": 8.906,
|
| 98 |
+
"step": 5000
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 95.24,
|
| 102 |
+
"learning_rate": 1.904761904761905e-06,
|
| 103 |
+
"loss": 0.0003,
|
| 104 |
+
"step": 6000
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 95.24,
|
| 108 |
+
"eval_accuracy": 0.9154777994693898,
|
| 109 |
+
"eval_f1": 0.581075973289656,
|
| 110 |
+
"eval_loss": 0.7306210994720459,
|
| 111 |
+
"eval_precision": 0.5783797341359418,
|
| 112 |
+
"eval_recall": 0.5837974683544304,
|
| 113 |
+
"eval_runtime": 7.1135,
|
| 114 |
+
"eval_samples_per_second": 70.289,
|
| 115 |
+
"eval_steps_per_second": 8.856,
|
| 116 |
+
"step": 6000
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 100.0,
|
| 120 |
+
"step": 6300,
|
| 121 |
+
"total_flos": 9.28710140928e+16,
|
| 122 |
+
"train_loss": 0.012086462969226497,
|
| 123 |
+
"train_runtime": 3307.3744,
|
| 124 |
+
"train_samples_per_second": 30.235,
|
| 125 |
+
"train_steps_per_second": 1.905
|
| 126 |
+
}
|
| 127 |
+
],
|
| 128 |
+
"max_steps": 6300,
|
| 129 |
+
"num_train_epochs": 100,
|
| 130 |
+
"total_flos": 9.28710140928e+16,
|
| 131 |
+
"trial_name": null,
|
| 132 |
+
"trial_params": null
|
| 133 |
+
}
|