Commit
•
ddd638f
1
Parent(s):
965e1e3
Upload trainer_state.json
Browse files- trainer_state.json +107 -0
trainer_state.json
ADDED
|
@@ -0,0 +1,107 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 30.0,
|
| 5 |
+
"global_step": 2490,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 4.82,
|
| 12 |
+
"learning_rate": 0.00023999999999999998,
|
| 13 |
+
"loss": 4.3036,
|
| 14 |
+
"step": 400
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 4.82,
|
| 18 |
+
"eval_loss": 0.6421348452568054,
|
| 19 |
+
"eval_runtime": 266.3964,
|
| 20 |
+
"eval_samples_per_second": 6.922,
|
| 21 |
+
"eval_wer": 0.7364348312511654,
|
| 22 |
+
"step": 400
|
| 23 |
+
},
|
| 24 |
+
{
|
| 25 |
+
"epoch": 9.64,
|
| 26 |
+
"learning_rate": 0.0002547738693467337,
|
| 27 |
+
"loss": 0.2971,
|
| 28 |
+
"step": 800
|
| 29 |
+
},
|
| 30 |
+
{
|
| 31 |
+
"epoch": 9.64,
|
| 32 |
+
"eval_loss": 0.39936190843582153,
|
| 33 |
+
"eval_runtime": 260.3352,
|
| 34 |
+
"eval_samples_per_second": 7.083,
|
| 35 |
+
"eval_wer": 0.4973895207906023,
|
| 36 |
+
"step": 800
|
| 37 |
+
},
|
| 38 |
+
{
|
| 39 |
+
"epoch": 14.46,
|
| 40 |
+
"learning_rate": 0.0001944723618090452,
|
| 41 |
+
"loss": 0.1098,
|
| 42 |
+
"step": 1200
|
| 43 |
+
},
|
| 44 |
+
{
|
| 45 |
+
"epoch": 14.46,
|
| 46 |
+
"eval_loss": 0.4166050851345062,
|
| 47 |
+
"eval_runtime": 265.7729,
|
| 48 |
+
"eval_samples_per_second": 6.938,
|
| 49 |
+
"eval_wer": 0.4447137796009696,
|
| 50 |
+
"step": 1200
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 19.28,
|
| 54 |
+
"learning_rate": 0.00013417085427135678,
|
| 55 |
+
"loss": 0.0664,
|
| 56 |
+
"step": 1600
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 19.28,
|
| 60 |
+
"eval_loss": 0.4249325096607208,
|
| 61 |
+
"eval_runtime": 265.3532,
|
| 62 |
+
"eval_samples_per_second": 6.949,
|
| 63 |
+
"eval_wer": 0.4381875815774753,
|
| 64 |
+
"step": 1600
|
| 65 |
+
},
|
| 66 |
+
{
|
| 67 |
+
"epoch": 24.1,
|
| 68 |
+
"learning_rate": 7.386934673366834e-05,
|
| 69 |
+
"loss": 0.0499,
|
| 70 |
+
"step": 2000
|
| 71 |
+
},
|
| 72 |
+
{
|
| 73 |
+
"epoch": 24.1,
|
| 74 |
+
"eval_loss": 0.42595329880714417,
|
| 75 |
+
"eval_runtime": 264.3813,
|
| 76 |
+
"eval_samples_per_second": 6.975,
|
| 77 |
+
"eval_wer": 0.42401640872645907,
|
| 78 |
+
"step": 2000
|
| 79 |
+
},
|
| 80 |
+
{
|
| 81 |
+
"epoch": 28.92,
|
| 82 |
+
"learning_rate": 1.35678391959799e-05,
|
| 83 |
+
"loss": 0.0407,
|
| 84 |
+
"step": 2400
|
| 85 |
+
},
|
| 86 |
+
{
|
| 87 |
+
"epoch": 28.92,
|
| 88 |
+
"eval_loss": 0.43581095337867737,
|
| 89 |
+
"eval_runtime": 272.7984,
|
| 90 |
+
"eval_samples_per_second": 6.76,
|
| 91 |
+
"eval_wer": 0.4190751445086705,
|
| 92 |
+
"step": 2400
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 30.0,
|
| 96 |
+
"step": 2490,
|
| 97 |
+
"total_flos": 1.3821944739107277e+19,
|
| 98 |
+
"train_runtime": 26891.3209,
|
| 99 |
+
"train_samples_per_second": 0.093
|
| 100 |
+
}
|
| 101 |
+
],
|
| 102 |
+
"max_steps": 2490,
|
| 103 |
+
"num_train_epochs": 30,
|
| 104 |
+
"total_flos": 1.3821944739107277e+19,
|
| 105 |
+
"trial_name": null,
|
| 106 |
+
"trial_params": null
|
| 107 |
+
}
|