End of training
Browse files- all_results.json +8 -0
- train_results.json +8 -0
- trainer_state.json +1029 -0
all_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"total_flos": 1.24283947843584e+18,
|
| 4 |
+
"train_loss": 0.03442937224805355,
|
| 5 |
+
"train_runtime": 6631.726,
|
| 6 |
+
"train_samples_per_second": 36.19,
|
| 7 |
+
"train_steps_per_second": 0.283
|
| 8 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"total_flos": 1.24283947843584e+18,
|
| 4 |
+
"train_loss": 0.03442937224805355,
|
| 5 |
+
"train_runtime": 6631.726,
|
| 6 |
+
"train_samples_per_second": 36.19,
|
| 7 |
+
"train_steps_per_second": 0.283
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1029 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.0046388679184019566,
|
| 3 |
+
"best_model_checkpoint": "./mobilevitv2_Liveness_detection_v1.0/checkpoint-1792",
|
| 4 |
+
"epoch": 3.0,
|
| 5 |
+
"eval_steps": 128,
|
| 6 |
+
"global_step": 1875,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.0256,
|
| 13 |
+
"grad_norm": 0.686086893081665,
|
| 14 |
+
"learning_rate": 4.9573333333333335e-05,
|
| 15 |
+
"loss": 0.6647,
|
| 16 |
+
"step": 16
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.0512,
|
| 20 |
+
"grad_norm": 0.7888203859329224,
|
| 21 |
+
"learning_rate": 4.914666666666667e-05,
|
| 22 |
+
"loss": 0.6063,
|
| 23 |
+
"step": 32
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.0768,
|
| 27 |
+
"grad_norm": 0.9471352100372314,
|
| 28 |
+
"learning_rate": 4.872000000000001e-05,
|
| 29 |
+
"loss": 0.5254,
|
| 30 |
+
"step": 48
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.1024,
|
| 34 |
+
"grad_norm": 0.9644356966018677,
|
| 35 |
+
"learning_rate": 4.8293333333333334e-05,
|
| 36 |
+
"loss": 0.437,
|
| 37 |
+
"step": 64
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.128,
|
| 41 |
+
"grad_norm": 1.174842119216919,
|
| 42 |
+
"learning_rate": 4.7866666666666674e-05,
|
| 43 |
+
"loss": 0.3153,
|
| 44 |
+
"step": 80
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.1536,
|
| 48 |
+
"grad_norm": 1.0370259284973145,
|
| 49 |
+
"learning_rate": 4.744e-05,
|
| 50 |
+
"loss": 0.2325,
|
| 51 |
+
"step": 96
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.1792,
|
| 55 |
+
"grad_norm": 0.7825124859809875,
|
| 56 |
+
"learning_rate": 4.701333333333334e-05,
|
| 57 |
+
"loss": 0.1471,
|
| 58 |
+
"step": 112
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.2048,
|
| 62 |
+
"grad_norm": 0.6782599687576294,
|
| 63 |
+
"learning_rate": 4.6586666666666666e-05,
|
| 64 |
+
"loss": 0.1093,
|
| 65 |
+
"step": 128
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.2048,
|
| 69 |
+
"eval_accuracy": 0.9929,
|
| 70 |
+
"eval_f1": 0.992897770729339,
|
| 71 |
+
"eval_loss": 0.06788957864046097,
|
| 72 |
+
"eval_precision": 0.9929058981404609,
|
| 73 |
+
"eval_recall": 0.9929,
|
| 74 |
+
"eval_runtime": 247.0578,
|
| 75 |
+
"eval_samples_per_second": 80.953,
|
| 76 |
+
"eval_steps_per_second": 10.119,
|
| 77 |
+
"step": 128
|
| 78 |
+
},
|
| 79 |
+
{
|
| 80 |
+
"epoch": 0.2304,
|
| 81 |
+
"grad_norm": 0.5946743488311768,
|
| 82 |
+
"learning_rate": 4.6160000000000005e-05,
|
| 83 |
+
"loss": 0.0675,
|
| 84 |
+
"step": 144
|
| 85 |
+
},
|
| 86 |
+
{
|
| 87 |
+
"epoch": 0.256,
|
| 88 |
+
"grad_norm": 0.6635921597480774,
|
| 89 |
+
"learning_rate": 4.573333333333333e-05,
|
| 90 |
+
"loss": 0.0477,
|
| 91 |
+
"step": 160
|
| 92 |
+
},
|
| 93 |
+
{
|
| 94 |
+
"epoch": 0.2816,
|
| 95 |
+
"grad_norm": 0.22356025874614716,
|
| 96 |
+
"learning_rate": 4.530666666666667e-05,
|
| 97 |
+
"loss": 0.0434,
|
| 98 |
+
"step": 176
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 0.3072,
|
| 102 |
+
"grad_norm": 0.7422251105308533,
|
| 103 |
+
"learning_rate": 4.488e-05,
|
| 104 |
+
"loss": 0.0332,
|
| 105 |
+
"step": 192
|
| 106 |
+
},
|
| 107 |
+
{
|
| 108 |
+
"epoch": 0.3328,
|
| 109 |
+
"grad_norm": 0.3897477090358734,
|
| 110 |
+
"learning_rate": 4.445333333333334e-05,
|
| 111 |
+
"loss": 0.0294,
|
| 112 |
+
"step": 208
|
| 113 |
+
},
|
| 114 |
+
{
|
| 115 |
+
"epoch": 0.3584,
|
| 116 |
+
"grad_norm": 0.7231374979019165,
|
| 117 |
+
"learning_rate": 4.402666666666666e-05,
|
| 118 |
+
"loss": 0.0251,
|
| 119 |
+
"step": 224
|
| 120 |
+
},
|
| 121 |
+
{
|
| 122 |
+
"epoch": 0.384,
|
| 123 |
+
"grad_norm": 0.6253751516342163,
|
| 124 |
+
"learning_rate": 4.36e-05,
|
| 125 |
+
"loss": 0.0218,
|
| 126 |
+
"step": 240
|
| 127 |
+
},
|
| 128 |
+
{
|
| 129 |
+
"epoch": 0.4096,
|
| 130 |
+
"grad_norm": 0.14728830754756927,
|
| 131 |
+
"learning_rate": 4.3173333333333336e-05,
|
| 132 |
+
"loss": 0.0234,
|
| 133 |
+
"step": 256
|
| 134 |
+
},
|
| 135 |
+
{
|
| 136 |
+
"epoch": 0.4096,
|
| 137 |
+
"eval_accuracy": 0.9962,
|
| 138 |
+
"eval_f1": 0.99619730556991,
|
| 139 |
+
"eval_loss": 0.016960417851805687,
|
| 140 |
+
"eval_precision": 0.9962127076373954,
|
| 141 |
+
"eval_recall": 0.9962,
|
| 142 |
+
"eval_runtime": 219.599,
|
| 143 |
+
"eval_samples_per_second": 91.075,
|
| 144 |
+
"eval_steps_per_second": 11.384,
|
| 145 |
+
"step": 256
|
| 146 |
+
},
|
| 147 |
+
{
|
| 148 |
+
"epoch": 0.4352,
|
| 149 |
+
"grad_norm": 0.45428574085235596,
|
| 150 |
+
"learning_rate": 4.274666666666667e-05,
|
| 151 |
+
"loss": 0.0174,
|
| 152 |
+
"step": 272
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 0.4608,
|
| 156 |
+
"grad_norm": 0.11591943353414536,
|
| 157 |
+
"learning_rate": 4.232e-05,
|
| 158 |
+
"loss": 0.0195,
|
| 159 |
+
"step": 288
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 0.4864,
|
| 163 |
+
"grad_norm": 0.34196004271507263,
|
| 164 |
+
"learning_rate": 4.1893333333333334e-05,
|
| 165 |
+
"loss": 0.0239,
|
| 166 |
+
"step": 304
|
| 167 |
+
},
|
| 168 |
+
{
|
| 169 |
+
"epoch": 0.512,
|
| 170 |
+
"grad_norm": 0.18962785601615906,
|
| 171 |
+
"learning_rate": 4.146666666666667e-05,
|
| 172 |
+
"loss": 0.0203,
|
| 173 |
+
"step": 320
|
| 174 |
+
},
|
| 175 |
+
{
|
| 176 |
+
"epoch": 0.5376,
|
| 177 |
+
"grad_norm": 0.05710538849234581,
|
| 178 |
+
"learning_rate": 4.104e-05,
|
| 179 |
+
"loss": 0.0247,
|
| 180 |
+
"step": 336
|
| 181 |
+
},
|
| 182 |
+
{
|
| 183 |
+
"epoch": 0.5632,
|
| 184 |
+
"grad_norm": 1.2746449708938599,
|
| 185 |
+
"learning_rate": 4.061333333333334e-05,
|
| 186 |
+
"loss": 0.017,
|
| 187 |
+
"step": 352
|
| 188 |
+
},
|
| 189 |
+
{
|
| 190 |
+
"epoch": 0.5888,
|
| 191 |
+
"grad_norm": 0.46703022718429565,
|
| 192 |
+
"learning_rate": 4.0186666666666666e-05,
|
| 193 |
+
"loss": 0.0112,
|
| 194 |
+
"step": 368
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 0.6144,
|
| 198 |
+
"grad_norm": 2.4929864406585693,
|
| 199 |
+
"learning_rate": 3.9760000000000006e-05,
|
| 200 |
+
"loss": 0.0186,
|
| 201 |
+
"step": 384
|
| 202 |
+
},
|
| 203 |
+
{
|
| 204 |
+
"epoch": 0.6144,
|
| 205 |
+
"eval_accuracy": 0.9973,
|
| 206 |
+
"eval_f1": 0.9973011575273466,
|
| 207 |
+
"eval_loss": 0.013053582981228828,
|
| 208 |
+
"eval_precision": 0.9973004220662013,
|
| 209 |
+
"eval_recall": 0.9973,
|
| 210 |
+
"eval_runtime": 215.8374,
|
| 211 |
+
"eval_samples_per_second": 92.662,
|
| 212 |
+
"eval_steps_per_second": 11.583,
|
| 213 |
+
"step": 384
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 0.64,
|
| 217 |
+
"grad_norm": 0.2726776897907257,
|
| 218 |
+
"learning_rate": 3.933333333333333e-05,
|
| 219 |
+
"loss": 0.009,
|
| 220 |
+
"step": 400
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 0.6656,
|
| 224 |
+
"grad_norm": 0.04591584578156471,
|
| 225 |
+
"learning_rate": 3.890666666666667e-05,
|
| 226 |
+
"loss": 0.0137,
|
| 227 |
+
"step": 416
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 0.6912,
|
| 231 |
+
"grad_norm": 0.2672736644744873,
|
| 232 |
+
"learning_rate": 3.848e-05,
|
| 233 |
+
"loss": 0.0156,
|
| 234 |
+
"step": 432
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.7168,
|
| 238 |
+
"grad_norm": 0.07182446122169495,
|
| 239 |
+
"learning_rate": 3.805333333333334e-05,
|
| 240 |
+
"loss": 0.0168,
|
| 241 |
+
"step": 448
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.7424,
|
| 245 |
+
"grad_norm": 0.029036881402134895,
|
| 246 |
+
"learning_rate": 3.762666666666667e-05,
|
| 247 |
+
"loss": 0.0104,
|
| 248 |
+
"step": 464
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.768,
|
| 252 |
+
"grad_norm": 1.6431952714920044,
|
| 253 |
+
"learning_rate": 3.72e-05,
|
| 254 |
+
"loss": 0.0145,
|
| 255 |
+
"step": 480
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 0.7936,
|
| 259 |
+
"grad_norm": 0.5779602527618408,
|
| 260 |
+
"learning_rate": 3.6773333333333336e-05,
|
| 261 |
+
"loss": 0.0099,
|
| 262 |
+
"step": 496
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 0.8192,
|
| 266 |
+
"grad_norm": 0.07454531639814377,
|
| 267 |
+
"learning_rate": 3.634666666666667e-05,
|
| 268 |
+
"loss": 0.0068,
|
| 269 |
+
"step": 512
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 0.8192,
|
| 273 |
+
"eval_accuracy": 0.99805,
|
| 274 |
+
"eval_f1": 0.9980503428173515,
|
| 275 |
+
"eval_loss": 0.008904083631932735,
|
| 276 |
+
"eval_precision": 0.9980497728736165,
|
| 277 |
+
"eval_recall": 0.99805,
|
| 278 |
+
"eval_runtime": 216.611,
|
| 279 |
+
"eval_samples_per_second": 92.331,
|
| 280 |
+
"eval_steps_per_second": 11.541,
|
| 281 |
+
"step": 512
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 0.8448,
|
| 285 |
+
"grad_norm": 0.0550871342420578,
|
| 286 |
+
"learning_rate": 3.592e-05,
|
| 287 |
+
"loss": 0.0095,
|
| 288 |
+
"step": 528
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"epoch": 0.8704,
|
| 292 |
+
"grad_norm": 0.13159841299057007,
|
| 293 |
+
"learning_rate": 3.5493333333333335e-05,
|
| 294 |
+
"loss": 0.0089,
|
| 295 |
+
"step": 544
|
| 296 |
+
},
|
| 297 |
+
{
|
| 298 |
+
"epoch": 0.896,
|
| 299 |
+
"grad_norm": 0.6349660754203796,
|
| 300 |
+
"learning_rate": 3.506666666666667e-05,
|
| 301 |
+
"loss": 0.0096,
|
| 302 |
+
"step": 560
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 0.9216,
|
| 306 |
+
"grad_norm": 0.7711471915245056,
|
| 307 |
+
"learning_rate": 3.464e-05,
|
| 308 |
+
"loss": 0.015,
|
| 309 |
+
"step": 576
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 0.9472,
|
| 313 |
+
"grad_norm": 0.05490660294890404,
|
| 314 |
+
"learning_rate": 3.424e-05,
|
| 315 |
+
"loss": 0.0113,
|
| 316 |
+
"step": 592
|
| 317 |
+
},
|
| 318 |
+
{
|
| 319 |
+
"epoch": 0.9728,
|
| 320 |
+
"grad_norm": 1.9179311990737915,
|
| 321 |
+
"learning_rate": 3.381333333333334e-05,
|
| 322 |
+
"loss": 0.0084,
|
| 323 |
+
"step": 608
|
| 324 |
+
},
|
| 325 |
+
{
|
| 326 |
+
"epoch": 0.9984,
|
| 327 |
+
"grad_norm": 0.04221022129058838,
|
| 328 |
+
"learning_rate": 3.338666666666666e-05,
|
| 329 |
+
"loss": 0.0166,
|
| 330 |
+
"step": 624
|
| 331 |
+
},
|
| 332 |
+
{
|
| 333 |
+
"epoch": 1.024,
|
| 334 |
+
"grad_norm": 0.21213921904563904,
|
| 335 |
+
"learning_rate": 3.296e-05,
|
| 336 |
+
"loss": 0.0049,
|
| 337 |
+
"step": 640
|
| 338 |
+
},
|
| 339 |
+
{
|
| 340 |
+
"epoch": 1.024,
|
| 341 |
+
"eval_accuracy": 0.9985,
|
| 342 |
+
"eval_f1": 0.9984999412798902,
|
| 343 |
+
"eval_loss": 0.006704141851514578,
|
| 344 |
+
"eval_precision": 0.9985001400955447,
|
| 345 |
+
"eval_recall": 0.9985,
|
| 346 |
+
"eval_runtime": 215.2961,
|
| 347 |
+
"eval_samples_per_second": 92.895,
|
| 348 |
+
"eval_steps_per_second": 11.612,
|
| 349 |
+
"step": 640
|
| 350 |
+
},
|
| 351 |
+
{
|
| 352 |
+
"epoch": 1.0496,
|
| 353 |
+
"grad_norm": 0.24302862584590912,
|
| 354 |
+
"learning_rate": 3.253333333333333e-05,
|
| 355 |
+
"loss": 0.0055,
|
| 356 |
+
"step": 656
|
| 357 |
+
},
|
| 358 |
+
{
|
| 359 |
+
"epoch": 1.0752,
|
| 360 |
+
"grad_norm": 0.17449580132961273,
|
| 361 |
+
"learning_rate": 3.210666666666667e-05,
|
| 362 |
+
"loss": 0.0078,
|
| 363 |
+
"step": 672
|
| 364 |
+
},
|
| 365 |
+
{
|
| 366 |
+
"epoch": 1.1008,
|
| 367 |
+
"grad_norm": 1.3184378147125244,
|
| 368 |
+
"learning_rate": 3.168e-05,
|
| 369 |
+
"loss": 0.0073,
|
| 370 |
+
"step": 688
|
| 371 |
+
},
|
| 372 |
+
{
|
| 373 |
+
"epoch": 1.1264,
|
| 374 |
+
"grad_norm": 0.8744115829467773,
|
| 375 |
+
"learning_rate": 3.1253333333333335e-05,
|
| 376 |
+
"loss": 0.0074,
|
| 377 |
+
"step": 704
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 1.152,
|
| 381 |
+
"grad_norm": 0.07388653606176376,
|
| 382 |
+
"learning_rate": 3.082666666666667e-05,
|
| 383 |
+
"loss": 0.0039,
|
| 384 |
+
"step": 720
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 1.1776,
|
| 388 |
+
"grad_norm": 0.035437047481536865,
|
| 389 |
+
"learning_rate": 3.04e-05,
|
| 390 |
+
"loss": 0.008,
|
| 391 |
+
"step": 736
|
| 392 |
+
},
|
| 393 |
+
{
|
| 394 |
+
"epoch": 1.2032,
|
| 395 |
+
"grad_norm": 0.06002597510814667,
|
| 396 |
+
"learning_rate": 2.9973333333333337e-05,
|
| 397 |
+
"loss": 0.0033,
|
| 398 |
+
"step": 752
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 1.2288000000000001,
|
| 402 |
+
"grad_norm": 0.01166555192321539,
|
| 403 |
+
"learning_rate": 2.9546666666666666e-05,
|
| 404 |
+
"loss": 0.0113,
|
| 405 |
+
"step": 768
|
| 406 |
+
},
|
| 407 |
+
{
|
| 408 |
+
"epoch": 1.2288000000000001,
|
| 409 |
+
"eval_accuracy": 0.99835,
|
| 410 |
+
"eval_f1": 0.9983500322782247,
|
| 411 |
+
"eval_loss": 0.006390445865690708,
|
| 412 |
+
"eval_precision": 0.9983499411063025,
|
| 413 |
+
"eval_recall": 0.99835,
|
| 414 |
+
"eval_runtime": 215.8826,
|
| 415 |
+
"eval_samples_per_second": 92.643,
|
| 416 |
+
"eval_steps_per_second": 11.58,
|
| 417 |
+
"step": 768
|
| 418 |
+
},
|
| 419 |
+
{
|
| 420 |
+
"epoch": 1.2544,
|
| 421 |
+
"grad_norm": 2.266014575958252,
|
| 422 |
+
"learning_rate": 2.9120000000000002e-05,
|
| 423 |
+
"loss": 0.0089,
|
| 424 |
+
"step": 784
|
| 425 |
+
},
|
| 426 |
+
{
|
| 427 |
+
"epoch": 1.28,
|
| 428 |
+
"grad_norm": 1.0778322219848633,
|
| 429 |
+
"learning_rate": 2.8693333333333332e-05,
|
| 430 |
+
"loss": 0.0041,
|
| 431 |
+
"step": 800
|
| 432 |
+
},
|
| 433 |
+
{
|
| 434 |
+
"epoch": 1.3056,
|
| 435 |
+
"grad_norm": 0.18685859441757202,
|
| 436 |
+
"learning_rate": 2.8266666666666668e-05,
|
| 437 |
+
"loss": 0.0041,
|
| 438 |
+
"step": 816
|
| 439 |
+
},
|
| 440 |
+
{
|
| 441 |
+
"epoch": 1.3312,
|
| 442 |
+
"grad_norm": 0.09057345986366272,
|
| 443 |
+
"learning_rate": 2.7839999999999998e-05,
|
| 444 |
+
"loss": 0.0071,
|
| 445 |
+
"step": 832
|
| 446 |
+
},
|
| 447 |
+
{
|
| 448 |
+
"epoch": 1.3568,
|
| 449 |
+
"grad_norm": 0.28758856654167175,
|
| 450 |
+
"learning_rate": 2.7413333333333334e-05,
|
| 451 |
+
"loss": 0.0045,
|
| 452 |
+
"step": 848
|
| 453 |
+
},
|
| 454 |
+
{
|
| 455 |
+
"epoch": 1.3824,
|
| 456 |
+
"grad_norm": 0.9198317527770996,
|
| 457 |
+
"learning_rate": 2.6986666666666667e-05,
|
| 458 |
+
"loss": 0.0107,
|
| 459 |
+
"step": 864
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 1.408,
|
| 463 |
+
"grad_norm": 0.14636975526809692,
|
| 464 |
+
"learning_rate": 2.6560000000000003e-05,
|
| 465 |
+
"loss": 0.004,
|
| 466 |
+
"step": 880
|
| 467 |
+
},
|
| 468 |
+
{
|
| 469 |
+
"epoch": 1.4336,
|
| 470 |
+
"grad_norm": 0.038640450686216354,
|
| 471 |
+
"learning_rate": 2.6133333333333333e-05,
|
| 472 |
+
"loss": 0.0061,
|
| 473 |
+
"step": 896
|
| 474 |
+
},
|
| 475 |
+
{
|
| 476 |
+
"epoch": 1.4336,
|
| 477 |
+
"eval_accuracy": 0.99835,
|
| 478 |
+
"eval_f1": 0.9983498384304894,
|
| 479 |
+
"eval_loss": 0.0060436660423874855,
|
| 480 |
+
"eval_precision": 0.9983504647077259,
|
| 481 |
+
"eval_recall": 0.99835,
|
| 482 |
+
"eval_runtime": 219.3162,
|
| 483 |
+
"eval_samples_per_second": 91.193,
|
| 484 |
+
"eval_steps_per_second": 11.399,
|
| 485 |
+
"step": 896
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 1.4592,
|
| 489 |
+
"grad_norm": 2.163743257522583,
|
| 490 |
+
"learning_rate": 2.570666666666667e-05,
|
| 491 |
+
"loss": 0.0071,
|
| 492 |
+
"step": 912
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 1.4848,
|
| 496 |
+
"grad_norm": 1.2653355598449707,
|
| 497 |
+
"learning_rate": 2.5280000000000005e-05,
|
| 498 |
+
"loss": 0.0046,
|
| 499 |
+
"step": 928
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 1.5104,
|
| 503 |
+
"grad_norm": 0.025220897048711777,
|
| 504 |
+
"learning_rate": 2.4853333333333335e-05,
|
| 505 |
+
"loss": 0.0039,
|
| 506 |
+
"step": 944
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 1.536,
|
| 510 |
+
"grad_norm": 0.015397891402244568,
|
| 511 |
+
"learning_rate": 2.4426666666666668e-05,
|
| 512 |
+
"loss": 0.0065,
|
| 513 |
+
"step": 960
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 1.5615999999999999,
|
| 517 |
+
"grad_norm": 0.06891310960054398,
|
| 518 |
+
"learning_rate": 2.4e-05,
|
| 519 |
+
"loss": 0.0049,
|
| 520 |
+
"step": 976
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 1.5872000000000002,
|
| 524 |
+
"grad_norm": 0.23304007947444916,
|
| 525 |
+
"learning_rate": 2.3573333333333334e-05,
|
| 526 |
+
"loss": 0.0125,
|
| 527 |
+
"step": 992
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 1.6128,
|
| 531 |
+
"grad_norm": 0.010659678839147091,
|
| 532 |
+
"learning_rate": 2.3146666666666666e-05,
|
| 533 |
+
"loss": 0.0044,
|
| 534 |
+
"step": 1008
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 1.6383999999999999,
|
| 538 |
+
"grad_norm": 0.0605621375143528,
|
| 539 |
+
"learning_rate": 2.2720000000000003e-05,
|
| 540 |
+
"loss": 0.0025,
|
| 541 |
+
"step": 1024
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 1.6383999999999999,
|
| 545 |
+
"eval_accuracy": 0.99835,
|
| 546 |
+
"eval_f1": 0.9983497737194017,
|
| 547 |
+
"eval_loss": 0.005817316006869078,
|
| 548 |
+
"eval_precision": 0.9983507300357946,
|
| 549 |
+
"eval_recall": 0.99835,
|
| 550 |
+
"eval_runtime": 214.0685,
|
| 551 |
+
"eval_samples_per_second": 93.428,
|
| 552 |
+
"eval_steps_per_second": 11.679,
|
| 553 |
+
"step": 1024
|
| 554 |
+
},
|
| 555 |
+
{
|
| 556 |
+
"epoch": 1.6640000000000001,
|
| 557 |
+
"grad_norm": 0.9562625288963318,
|
| 558 |
+
"learning_rate": 2.2293333333333336e-05,
|
| 559 |
+
"loss": 0.0081,
|
| 560 |
+
"step": 1040
|
| 561 |
+
},
|
| 562 |
+
{
|
| 563 |
+
"epoch": 1.6896,
|
| 564 |
+
"grad_norm": 1.08597731590271,
|
| 565 |
+
"learning_rate": 2.186666666666667e-05,
|
| 566 |
+
"loss": 0.0035,
|
| 567 |
+
"step": 1056
|
| 568 |
+
},
|
| 569 |
+
{
|
| 570 |
+
"epoch": 1.7151999999999998,
|
| 571 |
+
"grad_norm": 0.08577126264572144,
|
| 572 |
+
"learning_rate": 2.144e-05,
|
| 573 |
+
"loss": 0.004,
|
| 574 |
+
"step": 1072
|
| 575 |
+
},
|
| 576 |
+
{
|
| 577 |
+
"epoch": 1.7408000000000001,
|
| 578 |
+
"grad_norm": 0.007748621515929699,
|
| 579 |
+
"learning_rate": 2.1013333333333334e-05,
|
| 580 |
+
"loss": 0.0055,
|
| 581 |
+
"step": 1088
|
| 582 |
+
},
|
| 583 |
+
{
|
| 584 |
+
"epoch": 1.7664,
|
| 585 |
+
"grad_norm": 0.5899467468261719,
|
| 586 |
+
"learning_rate": 2.0586666666666667e-05,
|
| 587 |
+
"loss": 0.0059,
|
| 588 |
+
"step": 1104
|
| 589 |
+
},
|
| 590 |
+
{
|
| 591 |
+
"epoch": 1.792,
|
| 592 |
+
"grad_norm": 1.230574369430542,
|
| 593 |
+
"learning_rate": 2.016e-05,
|
| 594 |
+
"loss": 0.0052,
|
| 595 |
+
"step": 1120
|
| 596 |
+
},
|
| 597 |
+
{
|
| 598 |
+
"epoch": 1.8176,
|
| 599 |
+
"grad_norm": 0.4939228296279907,
|
| 600 |
+
"learning_rate": 1.9733333333333333e-05,
|
| 601 |
+
"loss": 0.0083,
|
| 602 |
+
"step": 1136
|
| 603 |
+
},
|
| 604 |
+
{
|
| 605 |
+
"epoch": 1.8432,
|
| 606 |
+
"grad_norm": 0.02068273536860943,
|
| 607 |
+
"learning_rate": 1.9306666666666666e-05,
|
| 608 |
+
"loss": 0.0019,
|
| 609 |
+
"step": 1152
|
| 610 |
+
},
|
| 611 |
+
{
|
| 612 |
+
"epoch": 1.8432,
|
| 613 |
+
"eval_accuracy": 0.99865,
|
| 614 |
+
"eval_f1": 0.9986499735808153,
|
| 615 |
+
"eval_loss": 0.0053438348695635796,
|
| 616 |
+
"eval_precision": 0.998650058503282,
|
| 617 |
+
"eval_recall": 0.99865,
|
| 618 |
+
"eval_runtime": 222.9408,
|
| 619 |
+
"eval_samples_per_second": 89.71,
|
| 620 |
+
"eval_steps_per_second": 11.214,
|
| 621 |
+
"step": 1152
|
| 622 |
+
},
|
| 623 |
+
{
|
| 624 |
+
"epoch": 1.8688,
|
| 625 |
+
"grad_norm": 1.5578925609588623,
|
| 626 |
+
"learning_rate": 1.888e-05,
|
| 627 |
+
"loss": 0.0063,
|
| 628 |
+
"step": 1168
|
| 629 |
+
},
|
| 630 |
+
{
|
| 631 |
+
"epoch": 1.8944,
|
| 632 |
+
"grad_norm": 0.00804234016686678,
|
| 633 |
+
"learning_rate": 1.8453333333333335e-05,
|
| 634 |
+
"loss": 0.0046,
|
| 635 |
+
"step": 1184
|
| 636 |
+
},
|
| 637 |
+
{
|
| 638 |
+
"epoch": 1.92,
|
| 639 |
+
"grad_norm": 2.660892963409424,
|
| 640 |
+
"learning_rate": 1.8026666666666668e-05,
|
| 641 |
+
"loss": 0.0046,
|
| 642 |
+
"step": 1200
|
| 643 |
+
},
|
| 644 |
+
{
|
| 645 |
+
"epoch": 1.9456,
|
| 646 |
+
"grad_norm": 0.008745341561734676,
|
| 647 |
+
"learning_rate": 1.76e-05,
|
| 648 |
+
"loss": 0.0029,
|
| 649 |
+
"step": 1216
|
| 650 |
+
},
|
| 651 |
+
{
|
| 652 |
+
"epoch": 1.9712,
|
| 653 |
+
"grad_norm": 0.015926124528050423,
|
| 654 |
+
"learning_rate": 1.7173333333333334e-05,
|
| 655 |
+
"loss": 0.0037,
|
| 656 |
+
"step": 1232
|
| 657 |
+
},
|
| 658 |
+
{
|
| 659 |
+
"epoch": 1.9968,
|
| 660 |
+
"grad_norm": 0.009791089221835136,
|
| 661 |
+
"learning_rate": 1.674666666666667e-05,
|
| 662 |
+
"loss": 0.0067,
|
| 663 |
+
"step": 1248
|
| 664 |
+
},
|
| 665 |
+
{
|
| 666 |
+
"epoch": 2.0224,
|
| 667 |
+
"grad_norm": 0.009614923968911171,
|
| 668 |
+
"learning_rate": 1.6320000000000003e-05,
|
| 669 |
+
"loss": 0.0022,
|
| 670 |
+
"step": 1264
|
| 671 |
+
},
|
| 672 |
+
{
|
| 673 |
+
"epoch": 2.048,
|
| 674 |
+
"grad_norm": 0.007969129830598831,
|
| 675 |
+
"learning_rate": 1.5893333333333336e-05,
|
| 676 |
+
"loss": 0.0056,
|
| 677 |
+
"step": 1280
|
| 678 |
+
},
|
| 679 |
+
{
|
| 680 |
+
"epoch": 2.048,
|
| 681 |
+
"eval_accuracy": 0.9987,
|
| 682 |
+
"eval_f1": 0.9986997451712335,
|
| 683 |
+
"eval_loss": 0.005075570661574602,
|
| 684 |
+
"eval_precision": 0.9987010761843749,
|
| 685 |
+
"eval_recall": 0.9987,
|
| 686 |
+
"eval_runtime": 220.6361,
|
| 687 |
+
"eval_samples_per_second": 90.647,
|
| 688 |
+
"eval_steps_per_second": 11.331,
|
| 689 |
+
"step": 1280
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 2.0736,
|
| 693 |
+
"grad_norm": 0.01278685312718153,
|
| 694 |
+
"learning_rate": 1.546666666666667e-05,
|
| 695 |
+
"loss": 0.0029,
|
| 696 |
+
"step": 1296
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 2.0992,
|
| 700 |
+
"grad_norm": 0.013344179838895798,
|
| 701 |
+
"learning_rate": 1.5040000000000002e-05,
|
| 702 |
+
"loss": 0.0029,
|
| 703 |
+
"step": 1312
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 2.1248,
|
| 707 |
+
"grad_norm": 0.021782569587230682,
|
| 708 |
+
"learning_rate": 1.4613333333333335e-05,
|
| 709 |
+
"loss": 0.0021,
|
| 710 |
+
"step": 1328
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 2.1504,
|
| 714 |
+
"grad_norm": 0.036290887743234634,
|
| 715 |
+
"learning_rate": 1.4186666666666667e-05,
|
| 716 |
+
"loss": 0.0031,
|
| 717 |
+
"step": 1344
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 2.176,
|
| 721 |
+
"grad_norm": 0.7045446634292603,
|
| 722 |
+
"learning_rate": 1.376e-05,
|
| 723 |
+
"loss": 0.003,
|
| 724 |
+
"step": 1360
|
| 725 |
+
},
|
| 726 |
+
{
|
| 727 |
+
"epoch": 2.2016,
|
| 728 |
+
"grad_norm": 0.012572742067277431,
|
| 729 |
+
"learning_rate": 1.3333333333333333e-05,
|
| 730 |
+
"loss": 0.0089,
|
| 731 |
+
"step": 1376
|
| 732 |
+
},
|
| 733 |
+
{
|
| 734 |
+
"epoch": 2.2272,
|
| 735 |
+
"grad_norm": 0.05552055686712265,
|
| 736 |
+
"learning_rate": 1.2906666666666666e-05,
|
| 737 |
+
"loss": 0.0011,
|
| 738 |
+
"step": 1392
|
| 739 |
+
},
|
| 740 |
+
{
|
| 741 |
+
"epoch": 2.2528,
|
| 742 |
+
"grad_norm": 0.047569043934345245,
|
| 743 |
+
"learning_rate": 1.248e-05,
|
| 744 |
+
"loss": 0.0015,
|
| 745 |
+
"step": 1408
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 2.2528,
|
| 749 |
+
"eval_accuracy": 0.9987,
|
| 750 |
+
"eval_f1": 0.9987,
|
| 751 |
+
"eval_loss": 0.004999413620680571,
|
| 752 |
+
"eval_precision": 0.9987,
|
| 753 |
+
"eval_recall": 0.9987,
|
| 754 |
+
"eval_runtime": 216.7321,
|
| 755 |
+
"eval_samples_per_second": 92.28,
|
| 756 |
+
"eval_steps_per_second": 11.535,
|
| 757 |
+
"step": 1408
|
| 758 |
+
},
|
| 759 |
+
{
|
| 760 |
+
"epoch": 2.2784,
|
| 761 |
+
"grad_norm": 0.391156405210495,
|
| 762 |
+
"learning_rate": 1.2053333333333334e-05,
|
| 763 |
+
"loss": 0.0038,
|
| 764 |
+
"step": 1424
|
| 765 |
+
},
|
| 766 |
+
{
|
| 767 |
+
"epoch": 2.304,
|
| 768 |
+
"grad_norm": 0.01173941045999527,
|
| 769 |
+
"learning_rate": 1.1626666666666668e-05,
|
| 770 |
+
"loss": 0.0075,
|
| 771 |
+
"step": 1440
|
| 772 |
+
},
|
| 773 |
+
{
|
| 774 |
+
"epoch": 2.3296,
|
| 775 |
+
"grad_norm": 2.2623393535614014,
|
| 776 |
+
"learning_rate": 1.1200000000000001e-05,
|
| 777 |
+
"loss": 0.0093,
|
| 778 |
+
"step": 1456
|
| 779 |
+
},
|
| 780 |
+
{
|
| 781 |
+
"epoch": 2.3552,
|
| 782 |
+
"grad_norm": 0.01715581864118576,
|
| 783 |
+
"learning_rate": 1.0773333333333334e-05,
|
| 784 |
+
"loss": 0.0057,
|
| 785 |
+
"step": 1472
|
| 786 |
+
},
|
| 787 |
+
{
|
| 788 |
+
"epoch": 2.3808,
|
| 789 |
+
"grad_norm": 1.0069034099578857,
|
| 790 |
+
"learning_rate": 1.0346666666666667e-05,
|
| 791 |
+
"loss": 0.0011,
|
| 792 |
+
"step": 1488
|
| 793 |
+
},
|
| 794 |
+
{
|
| 795 |
+
"epoch": 2.4064,
|
| 796 |
+
"grad_norm": 0.20940986275672913,
|
| 797 |
+
"learning_rate": 9.92e-06,
|
| 798 |
+
"loss": 0.0022,
|
| 799 |
+
"step": 1504
|
| 800 |
+
},
|
| 801 |
+
{
|
| 802 |
+
"epoch": 2.432,
|
| 803 |
+
"grad_norm": 0.011296062730252743,
|
| 804 |
+
"learning_rate": 9.493333333333333e-06,
|
| 805 |
+
"loss": 0.0013,
|
| 806 |
+
"step": 1520
|
| 807 |
+
},
|
| 808 |
+
{
|
| 809 |
+
"epoch": 2.4576000000000002,
|
| 810 |
+
"grad_norm": 1.7626785039901733,
|
| 811 |
+
"learning_rate": 9.066666666666667e-06,
|
| 812 |
+
"loss": 0.0055,
|
| 813 |
+
"step": 1536
|
| 814 |
+
},
|
| 815 |
+
{
|
| 816 |
+
"epoch": 2.4576000000000002,
|
| 817 |
+
"eval_accuracy": 0.99875,
|
| 818 |
+
"eval_f1": 0.9987499265863476,
|
| 819 |
+
"eval_loss": 0.004937997553497553,
|
| 820 |
+
"eval_precision": 0.998750197817996,
|
| 821 |
+
"eval_recall": 0.99875,
|
| 822 |
+
"eval_runtime": 216.9352,
|
| 823 |
+
"eval_samples_per_second": 92.193,
|
| 824 |
+
"eval_steps_per_second": 11.524,
|
| 825 |
+
"step": 1536
|
| 826 |
+
},
|
| 827 |
+
{
|
| 828 |
+
"epoch": 2.4832,
|
| 829 |
+
"grad_norm": 0.029261792078614235,
|
| 830 |
+
"learning_rate": 8.64e-06,
|
| 831 |
+
"loss": 0.0038,
|
| 832 |
+
"step": 1552
|
| 833 |
+
},
|
| 834 |
+
{
|
| 835 |
+
"epoch": 2.5088,
|
| 836 |
+
"grad_norm": 0.04974502697587013,
|
| 837 |
+
"learning_rate": 8.213333333333333e-06,
|
| 838 |
+
"loss": 0.0051,
|
| 839 |
+
"step": 1568
|
| 840 |
+
},
|
| 841 |
+
{
|
| 842 |
+
"epoch": 2.5343999999999998,
|
| 843 |
+
"grad_norm": 0.03413296118378639,
|
| 844 |
+
"learning_rate": 7.786666666666668e-06,
|
| 845 |
+
"loss": 0.0013,
|
| 846 |
+
"step": 1584
|
| 847 |
+
},
|
| 848 |
+
{
|
| 849 |
+
"epoch": 2.56,
|
| 850 |
+
"grad_norm": 0.02191867306828499,
|
| 851 |
+
"learning_rate": 7.36e-06,
|
| 852 |
+
"loss": 0.0011,
|
| 853 |
+
"step": 1600
|
| 854 |
+
},
|
| 855 |
+
{
|
| 856 |
+
"epoch": 2.5856,
|
| 857 |
+
"grad_norm": 0.00806102529168129,
|
| 858 |
+
"learning_rate": 6.933333333333334e-06,
|
| 859 |
+
"loss": 0.0032,
|
| 860 |
+
"step": 1616
|
| 861 |
+
},
|
| 862 |
+
{
|
| 863 |
+
"epoch": 2.6112,
|
| 864 |
+
"grad_norm": 0.0360230915248394,
|
| 865 |
+
"learning_rate": 6.5066666666666665e-06,
|
| 866 |
+
"loss": 0.007,
|
| 867 |
+
"step": 1632
|
| 868 |
+
},
|
| 869 |
+
{
|
| 870 |
+
"epoch": 2.6368,
|
| 871 |
+
"grad_norm": 0.024547284469008446,
|
| 872 |
+
"learning_rate": 6.08e-06,
|
| 873 |
+
"loss": 0.007,
|
| 874 |
+
"step": 1648
|
| 875 |
+
},
|
| 876 |
+
{
|
| 877 |
+
"epoch": 2.6624,
|
| 878 |
+
"grad_norm": 0.18587003648281097,
|
| 879 |
+
"learning_rate": 5.653333333333333e-06,
|
| 880 |
+
"loss": 0.0023,
|
| 881 |
+
"step": 1664
|
| 882 |
+
},
|
| 883 |
+
{
|
| 884 |
+
"epoch": 2.6624,
|
| 885 |
+
"eval_accuracy": 0.99885,
|
| 886 |
+
"eval_f1": 0.9988497971544271,
|
| 887 |
+
"eval_loss": 0.004876961465924978,
|
| 888 |
+
"eval_precision": 0.998850864665525,
|
| 889 |
+
"eval_recall": 0.99885,
|
| 890 |
+
"eval_runtime": 216.0532,
|
| 891 |
+
"eval_samples_per_second": 92.57,
|
| 892 |
+
"eval_steps_per_second": 11.571,
|
| 893 |
+
"step": 1664
|
| 894 |
+
},
|
| 895 |
+
{
|
| 896 |
+
"epoch": 2.6879999999999997,
|
| 897 |
+
"grad_norm": 0.8803591728210449,
|
| 898 |
+
"learning_rate": 5.226666666666667e-06,
|
| 899 |
+
"loss": 0.0034,
|
| 900 |
+
"step": 1680
|
| 901 |
+
},
|
| 902 |
+
{
|
| 903 |
+
"epoch": 2.7136,
|
| 904 |
+
"grad_norm": 0.033573221415281296,
|
| 905 |
+
"learning_rate": 4.800000000000001e-06,
|
| 906 |
+
"loss": 0.0057,
|
| 907 |
+
"step": 1696
|
| 908 |
+
},
|
| 909 |
+
{
|
| 910 |
+
"epoch": 2.7392,
|
| 911 |
+
"grad_norm": 0.1415340006351471,
|
| 912 |
+
"learning_rate": 4.3733333333333335e-06,
|
| 913 |
+
"loss": 0.0017,
|
| 914 |
+
"step": 1712
|
| 915 |
+
},
|
| 916 |
+
{
|
| 917 |
+
"epoch": 2.7648,
|
| 918 |
+
"grad_norm": 0.5475676655769348,
|
| 919 |
+
"learning_rate": 3.9466666666666664e-06,
|
| 920 |
+
"loss": 0.0067,
|
| 921 |
+
"step": 1728
|
| 922 |
+
},
|
| 923 |
+
{
|
| 924 |
+
"epoch": 2.7904,
|
| 925 |
+
"grad_norm": 0.062486432492733,
|
| 926 |
+
"learning_rate": 3.52e-06,
|
| 927 |
+
"loss": 0.0033,
|
| 928 |
+
"step": 1744
|
| 929 |
+
},
|
| 930 |
+
{
|
| 931 |
+
"epoch": 2.816,
|
| 932 |
+
"grad_norm": 0.009262731298804283,
|
| 933 |
+
"learning_rate": 3.0933333333333335e-06,
|
| 934 |
+
"loss": 0.0014,
|
| 935 |
+
"step": 1760
|
| 936 |
+
},
|
| 937 |
+
{
|
| 938 |
+
"epoch": 2.8416,
|
| 939 |
+
"grad_norm": 0.07086601853370667,
|
| 940 |
+
"learning_rate": 2.666666666666667e-06,
|
| 941 |
+
"loss": 0.003,
|
| 942 |
+
"step": 1776
|
| 943 |
+
},
|
| 944 |
+
{
|
| 945 |
+
"epoch": 2.8672,
|
| 946 |
+
"grad_norm": 0.04299961403012276,
|
| 947 |
+
"learning_rate": 2.24e-06,
|
| 948 |
+
"loss": 0.0027,
|
| 949 |
+
"step": 1792
|
| 950 |
+
},
|
| 951 |
+
{
|
| 952 |
+
"epoch": 2.8672,
|
| 953 |
+
"eval_accuracy": 0.9988,
|
| 954 |
+
"eval_f1": 0.9987998118880025,
|
| 955 |
+
"eval_loss": 0.0046388679184019566,
|
| 956 |
+
"eval_precision": 0.9988007388473777,
|
| 957 |
+
"eval_recall": 0.9988,
|
| 958 |
+
"eval_runtime": 214.8729,
|
| 959 |
+
"eval_samples_per_second": 93.078,
|
| 960 |
+
"eval_steps_per_second": 11.635,
|
| 961 |
+
"step": 1792
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 2.8928000000000003,
|
| 965 |
+
"grad_norm": 0.012935050763189793,
|
| 966 |
+
"learning_rate": 1.8133333333333335e-06,
|
| 967 |
+
"loss": 0.0023,
|
| 968 |
+
"step": 1808
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 2.9184,
|
| 972 |
+
"grad_norm": 1.5018342733383179,
|
| 973 |
+
"learning_rate": 1.3866666666666666e-06,
|
| 974 |
+
"loss": 0.0099,
|
| 975 |
+
"step": 1824
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 2.944,
|
| 979 |
+
"grad_norm": 0.10675271600484848,
|
| 980 |
+
"learning_rate": 9.6e-07,
|
| 981 |
+
"loss": 0.0013,
|
| 982 |
+
"step": 1840
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 2.9696,
|
| 986 |
+
"grad_norm": 0.010167909786105156,
|
| 987 |
+
"learning_rate": 5.333333333333333e-07,
|
| 988 |
+
"loss": 0.003,
|
| 989 |
+
"step": 1856
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 2.9952,
|
| 993 |
+
"grad_norm": 0.014523031190037727,
|
| 994 |
+
"learning_rate": 1.0666666666666668e-07,
|
| 995 |
+
"loss": 0.0058,
|
| 996 |
+
"step": 1872
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 3.0,
|
| 1000 |
+
"step": 1875,
|
| 1001 |
+
"total_flos": 1.24283947843584e+18,
|
| 1002 |
+
"train_loss": 0.03442937224805355,
|
| 1003 |
+
"train_runtime": 6631.726,
|
| 1004 |
+
"train_samples_per_second": 36.19,
|
| 1005 |
+
"train_steps_per_second": 0.283
|
| 1006 |
+
}
|
| 1007 |
+
],
|
| 1008 |
+
"logging_steps": 16,
|
| 1009 |
+
"max_steps": 1875,
|
| 1010 |
+
"num_input_tokens_seen": 0,
|
| 1011 |
+
"num_train_epochs": 3,
|
| 1012 |
+
"save_steps": 128,
|
| 1013 |
+
"stateful_callbacks": {
|
| 1014 |
+
"TrainerControl": {
|
| 1015 |
+
"args": {
|
| 1016 |
+
"should_epoch_stop": false,
|
| 1017 |
+
"should_evaluate": false,
|
| 1018 |
+
"should_log": false,
|
| 1019 |
+
"should_save": true,
|
| 1020 |
+
"should_training_stop": true
|
| 1021 |
+
},
|
| 1022 |
+
"attributes": {}
|
| 1023 |
+
}
|
| 1024 |
+
},
|
| 1025 |
+
"total_flos": 1.24283947843584e+18,
|
| 1026 |
+
"train_batch_size": 128,
|
| 1027 |
+
"trial_name": null,
|
| 1028 |
+
"trial_params": null
|
| 1029 |
+
}
|