Upload trainer_state.json with huggingface_hub
Browse files- trainer_state.json +115 -0
trainer_state.json
ADDED
|
@@ -0,0 +1,115 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 3.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 684,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.22,
|
| 13 |
+
"learning_rate": 1.8538011695906434e-05,
|
| 14 |
+
"loss": 2.3079,
|
| 15 |
+
"step": 50
|
| 16 |
+
},
|
| 17 |
+
{
|
| 18 |
+
"epoch": 0.44,
|
| 19 |
+
"learning_rate": 1.7076023391812867e-05,
|
| 20 |
+
"loss": 1.5695,
|
| 21 |
+
"step": 100
|
| 22 |
+
},
|
| 23 |
+
{
|
| 24 |
+
"epoch": 0.66,
|
| 25 |
+
"learning_rate": 1.56140350877193e-05,
|
| 26 |
+
"loss": 1.1902,
|
| 27 |
+
"step": 150
|
| 28 |
+
},
|
| 29 |
+
{
|
| 30 |
+
"epoch": 0.88,
|
| 31 |
+
"learning_rate": 1.4152046783625733e-05,
|
| 32 |
+
"loss": 1.0055,
|
| 33 |
+
"step": 200
|
| 34 |
+
},
|
| 35 |
+
{
|
| 36 |
+
"epoch": 1.1,
|
| 37 |
+
"learning_rate": 1.2690058479532166e-05,
|
| 38 |
+
"loss": 0.8595,
|
| 39 |
+
"step": 250
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 1.32,
|
| 43 |
+
"learning_rate": 1.1228070175438597e-05,
|
| 44 |
+
"loss": 0.7444,
|
| 45 |
+
"step": 300
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 1.54,
|
| 49 |
+
"learning_rate": 9.76608187134503e-06,
|
| 50 |
+
"loss": 0.6758,
|
| 51 |
+
"step": 350
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 1.75,
|
| 55 |
+
"learning_rate": 8.304093567251463e-06,
|
| 56 |
+
"loss": 0.6153,
|
| 57 |
+
"step": 400
|
| 58 |
+
},
|
| 59 |
+
{
|
| 60 |
+
"epoch": 1.97,
|
| 61 |
+
"learning_rate": 6.842105263157896e-06,
|
| 62 |
+
"loss": 0.5803,
|
| 63 |
+
"step": 450
|
| 64 |
+
},
|
| 65 |
+
{
|
| 66 |
+
"epoch": 2.19,
|
| 67 |
+
"learning_rate": 5.380116959064328e-06,
|
| 68 |
+
"loss": 0.4988,
|
| 69 |
+
"step": 500
|
| 70 |
+
},
|
| 71 |
+
{
|
| 72 |
+
"epoch": 2.19,
|
| 73 |
+
"eval_accuracy": 0.8061168760240306,
|
| 74 |
+
"eval_loss": 0.6808876395225525,
|
| 75 |
+
"eval_runtime": 1.655,
|
| 76 |
+
"eval_samples_per_second": 1106.366,
|
| 77 |
+
"eval_steps_per_second": 69.488,
|
| 78 |
+
"step": 500
|
| 79 |
+
},
|
| 80 |
+
{
|
| 81 |
+
"epoch": 2.41,
|
| 82 |
+
"learning_rate": 3.9181286549707605e-06,
|
| 83 |
+
"loss": 0.4889,
|
| 84 |
+
"step": 550
|
| 85 |
+
},
|
| 86 |
+
{
|
| 87 |
+
"epoch": 2.63,
|
| 88 |
+
"learning_rate": 2.456140350877193e-06,
|
| 89 |
+
"loss": 0.4063,
|
| 90 |
+
"step": 600
|
| 91 |
+
},
|
| 92 |
+
{
|
| 93 |
+
"epoch": 2.85,
|
| 94 |
+
"learning_rate": 9.941520467836258e-07,
|
| 95 |
+
"loss": 0.3951,
|
| 96 |
+
"step": 650
|
| 97 |
+
},
|
| 98 |
+
{
|
| 99 |
+
"epoch": 3.0,
|
| 100 |
+
"step": 684,
|
| 101 |
+
"total_flos": 631486788074352.0,
|
| 102 |
+
"train_loss": 0.8504555141716673,
|
| 103 |
+
"train_runtime": 83.9962,
|
| 104 |
+
"train_samples_per_second": 260.476,
|
| 105 |
+
"train_steps_per_second": 8.143
|
| 106 |
+
}
|
| 107 |
+
],
|
| 108 |
+
"logging_steps": 50,
|
| 109 |
+
"max_steps": 684,
|
| 110 |
+
"num_train_epochs": 3,
|
| 111 |
+
"save_steps": 500,
|
| 112 |
+
"total_flos": 631486788074352.0,
|
| 113 |
+
"trial_name": null,
|
| 114 |
+
"trial_params": null
|
| 115 |
+
}
|