End of training
Browse files
all_results.json
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"eval_accuracy": 0.9886666666666667,
|
| 4 |
+
"eval_loss": 0.0382872000336647,
|
| 5 |
+
"eval_runtime": 35.7639,
|
| 6 |
+
"eval_samples_per_second": 167.767,
|
| 7 |
+
"eval_steps_per_second": 5.257,
|
| 8 |
+
"total_flos": 4.027570334613504e+18,
|
| 9 |
+
"train_loss": 1.9390247602598363,
|
| 10 |
+
"train_runtime": 2519.9488,
|
| 11 |
+
"train_samples_per_second": 64.287,
|
| 12 |
+
"train_steps_per_second": 0.502
|
| 13 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"eval_accuracy": 0.9886666666666667,
|
| 4 |
+
"eval_loss": 0.0382872000336647,
|
| 5 |
+
"eval_runtime": 35.7639,
|
| 6 |
+
"eval_samples_per_second": 167.767,
|
| 7 |
+
"eval_steps_per_second": 5.257
|
| 8 |
+
}
|
runs/Dec26_17-14-13_4f06fcef64d2/events.out.tfevents.1735235839.4f06fcef64d2.551.1
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:218b08f4923bc7cc0795463f72a644ac9acbafcec629d68c7fa0597479126f87
|
| 3 |
+
size 411
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"total_flos": 4.027570334613504e+18,
|
| 4 |
+
"train_loss": 1.9390247602598363,
|
| 5 |
+
"train_runtime": 2519.9488,
|
| 6 |
+
"train_samples_per_second": 64.287,
|
| 7 |
+
"train_steps_per_second": 0.502
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,951 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.9886666666666667,
|
| 3 |
+
"best_model_checkpoint": "swin-tiny-patch4-window7-224-finetuned-eurosat/checkpoint-1266",
|
| 4 |
+
"epoch": 3.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 1266,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.023696682464454975,
|
| 13 |
+
"grad_norm": 15.438875198364258,
|
| 14 |
+
"learning_rate": 3.937007874015748e-06,
|
| 15 |
+
"loss": 9.3952,
|
| 16 |
+
"step": 10
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.04739336492890995,
|
| 20 |
+
"grad_norm": 17.365249633789062,
|
| 21 |
+
"learning_rate": 7.874015748031496e-06,
|
| 22 |
+
"loss": 9.0232,
|
| 23 |
+
"step": 20
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.07109004739336493,
|
| 27 |
+
"grad_norm": 23.01938819885254,
|
| 28 |
+
"learning_rate": 1.1811023622047245e-05,
|
| 29 |
+
"loss": 8.5712,
|
| 30 |
+
"step": 30
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.0947867298578199,
|
| 34 |
+
"grad_norm": 51.1461067199707,
|
| 35 |
+
"learning_rate": 1.5748031496062993e-05,
|
| 36 |
+
"loss": 7.8145,
|
| 37 |
+
"step": 40
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.11848341232227488,
|
| 41 |
+
"grad_norm": 48.782527923583984,
|
| 42 |
+
"learning_rate": 1.9685039370078743e-05,
|
| 43 |
+
"loss": 6.7351,
|
| 44 |
+
"step": 50
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.14218009478672985,
|
| 48 |
+
"grad_norm": 86.34725189208984,
|
| 49 |
+
"learning_rate": 2.362204724409449e-05,
|
| 50 |
+
"loss": 5.2743,
|
| 51 |
+
"step": 60
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.16587677725118483,
|
| 55 |
+
"grad_norm": 49.05707550048828,
|
| 56 |
+
"learning_rate": 2.755905511811024e-05,
|
| 57 |
+
"loss": 4.1303,
|
| 58 |
+
"step": 70
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.1895734597156398,
|
| 62 |
+
"grad_norm": 70.76948547363281,
|
| 63 |
+
"learning_rate": 3.1496062992125985e-05,
|
| 64 |
+
"loss": 3.4811,
|
| 65 |
+
"step": 80
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.2132701421800948,
|
| 69 |
+
"grad_norm": 65.15145111083984,
|
| 70 |
+
"learning_rate": 3.5433070866141735e-05,
|
| 71 |
+
"loss": 3.2691,
|
| 72 |
+
"step": 90
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.23696682464454977,
|
| 76 |
+
"grad_norm": 39.73679733276367,
|
| 77 |
+
"learning_rate": 3.9370078740157485e-05,
|
| 78 |
+
"loss": 2.9577,
|
| 79 |
+
"step": 100
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.26066350710900477,
|
| 83 |
+
"grad_norm": 66.20768737792969,
|
| 84 |
+
"learning_rate": 4.330708661417323e-05,
|
| 85 |
+
"loss": 2.6592,
|
| 86 |
+
"step": 110
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.2843601895734597,
|
| 90 |
+
"grad_norm": 50.88495635986328,
|
| 91 |
+
"learning_rate": 4.724409448818898e-05,
|
| 92 |
+
"loss": 2.4807,
|
| 93 |
+
"step": 120
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.3080568720379147,
|
| 97 |
+
"grad_norm": 43.55908203125,
|
| 98 |
+
"learning_rate": 4.986830553116769e-05,
|
| 99 |
+
"loss": 2.3623,
|
| 100 |
+
"step": 130
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.33175355450236965,
|
| 104 |
+
"grad_norm": 30.119890213012695,
|
| 105 |
+
"learning_rate": 4.9429323968393335e-05,
|
| 106 |
+
"loss": 2.3652,
|
| 107 |
+
"step": 140
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.35545023696682465,
|
| 111 |
+
"grad_norm": 28.67392349243164,
|
| 112 |
+
"learning_rate": 4.8990342405618965e-05,
|
| 113 |
+
"loss": 2.3158,
|
| 114 |
+
"step": 150
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.3791469194312796,
|
| 118 |
+
"grad_norm": 31.85961151123047,
|
| 119 |
+
"learning_rate": 4.85513608428446e-05,
|
| 120 |
+
"loss": 2.3018,
|
| 121 |
+
"step": 160
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.4028436018957346,
|
| 125 |
+
"grad_norm": 75.70964813232422,
|
| 126 |
+
"learning_rate": 4.8112379280070244e-05,
|
| 127 |
+
"loss": 2.1835,
|
| 128 |
+
"step": 170
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.4265402843601896,
|
| 132 |
+
"grad_norm": 24.69963836669922,
|
| 133 |
+
"learning_rate": 4.7673397717295874e-05,
|
| 134 |
+
"loss": 2.0962,
|
| 135 |
+
"step": 180
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.45023696682464454,
|
| 139 |
+
"grad_norm": 46.664432525634766,
|
| 140 |
+
"learning_rate": 4.723441615452151e-05,
|
| 141 |
+
"loss": 2.0229,
|
| 142 |
+
"step": 190
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.47393364928909953,
|
| 146 |
+
"grad_norm": 31.81464195251465,
|
| 147 |
+
"learning_rate": 4.679543459174715e-05,
|
| 148 |
+
"loss": 2.0519,
|
| 149 |
+
"step": 200
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.4976303317535545,
|
| 153 |
+
"grad_norm": 25.131099700927734,
|
| 154 |
+
"learning_rate": 4.635645302897278e-05,
|
| 155 |
+
"loss": 2.1874,
|
| 156 |
+
"step": 210
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.5213270142180095,
|
| 160 |
+
"grad_norm": 23.477203369140625,
|
| 161 |
+
"learning_rate": 4.591747146619842e-05,
|
| 162 |
+
"loss": 2.2174,
|
| 163 |
+
"step": 220
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.5450236966824644,
|
| 167 |
+
"grad_norm": 28.71649742126465,
|
| 168 |
+
"learning_rate": 4.547848990342406e-05,
|
| 169 |
+
"loss": 1.8939,
|
| 170 |
+
"step": 230
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.5687203791469194,
|
| 174 |
+
"grad_norm": 25.678308486938477,
|
| 175 |
+
"learning_rate": 4.503950834064969e-05,
|
| 176 |
+
"loss": 1.9235,
|
| 177 |
+
"step": 240
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.5924170616113744,
|
| 181 |
+
"grad_norm": 17.22383689880371,
|
| 182 |
+
"learning_rate": 4.460052677787533e-05,
|
| 183 |
+
"loss": 1.9575,
|
| 184 |
+
"step": 250
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.6161137440758294,
|
| 188 |
+
"grad_norm": 25.669090270996094,
|
| 189 |
+
"learning_rate": 4.416154521510097e-05,
|
| 190 |
+
"loss": 1.8465,
|
| 191 |
+
"step": 260
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.6398104265402843,
|
| 195 |
+
"grad_norm": 27.523319244384766,
|
| 196 |
+
"learning_rate": 4.37225636523266e-05,
|
| 197 |
+
"loss": 1.9501,
|
| 198 |
+
"step": 270
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.6635071090047393,
|
| 202 |
+
"grad_norm": 23.112300872802734,
|
| 203 |
+
"learning_rate": 4.328358208955224e-05,
|
| 204 |
+
"loss": 1.7979,
|
| 205 |
+
"step": 280
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.6872037914691943,
|
| 209 |
+
"grad_norm": 21.1746768951416,
|
| 210 |
+
"learning_rate": 4.284460052677788e-05,
|
| 211 |
+
"loss": 1.7745,
|
| 212 |
+
"step": 290
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.7109004739336493,
|
| 216 |
+
"grad_norm": 39.38931655883789,
|
| 217 |
+
"learning_rate": 4.240561896400351e-05,
|
| 218 |
+
"loss": 1.8012,
|
| 219 |
+
"step": 300
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.7345971563981043,
|
| 223 |
+
"grad_norm": 32.20198440551758,
|
| 224 |
+
"learning_rate": 4.196663740122915e-05,
|
| 225 |
+
"loss": 2.0618,
|
| 226 |
+
"step": 310
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.7582938388625592,
|
| 230 |
+
"grad_norm": 23.157480239868164,
|
| 231 |
+
"learning_rate": 4.152765583845479e-05,
|
| 232 |
+
"loss": 1.7972,
|
| 233 |
+
"step": 320
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.7819905213270142,
|
| 237 |
+
"grad_norm": 24.55837631225586,
|
| 238 |
+
"learning_rate": 4.108867427568042e-05,
|
| 239 |
+
"loss": 1.8858,
|
| 240 |
+
"step": 330
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.8056872037914692,
|
| 244 |
+
"grad_norm": 18.101133346557617,
|
| 245 |
+
"learning_rate": 4.064969271290606e-05,
|
| 246 |
+
"loss": 1.7343,
|
| 247 |
+
"step": 340
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.8293838862559242,
|
| 251 |
+
"grad_norm": 25.414764404296875,
|
| 252 |
+
"learning_rate": 4.02107111501317e-05,
|
| 253 |
+
"loss": 1.8935,
|
| 254 |
+
"step": 350
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.8530805687203792,
|
| 258 |
+
"grad_norm": 15.749005317687988,
|
| 259 |
+
"learning_rate": 3.977172958735733e-05,
|
| 260 |
+
"loss": 1.5343,
|
| 261 |
+
"step": 360
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.8767772511848341,
|
| 265 |
+
"grad_norm": 22.70486831665039,
|
| 266 |
+
"learning_rate": 3.933274802458297e-05,
|
| 267 |
+
"loss": 1.6898,
|
| 268 |
+
"step": 370
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.9004739336492891,
|
| 272 |
+
"grad_norm": 12.152778625488281,
|
| 273 |
+
"learning_rate": 3.889376646180861e-05,
|
| 274 |
+
"loss": 1.7357,
|
| 275 |
+
"step": 380
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.9241706161137441,
|
| 279 |
+
"grad_norm": 14.83531379699707,
|
| 280 |
+
"learning_rate": 3.8454784899034244e-05,
|
| 281 |
+
"loss": 1.6003,
|
| 282 |
+
"step": 390
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.9478672985781991,
|
| 286 |
+
"grad_norm": 21.903425216674805,
|
| 287 |
+
"learning_rate": 3.801580333625988e-05,
|
| 288 |
+
"loss": 1.6476,
|
| 289 |
+
"step": 400
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.9715639810426541,
|
| 293 |
+
"grad_norm": 12.695667266845703,
|
| 294 |
+
"learning_rate": 3.757682177348552e-05,
|
| 295 |
+
"loss": 1.6488,
|
| 296 |
+
"step": 410
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.995260663507109,
|
| 300 |
+
"grad_norm": 18.37169647216797,
|
| 301 |
+
"learning_rate": 3.713784021071115e-05,
|
| 302 |
+
"loss": 1.4709,
|
| 303 |
+
"step": 420
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 1.0,
|
| 307 |
+
"eval_accuracy": 0.9806666666666667,
|
| 308 |
+
"eval_loss": 0.06099523976445198,
|
| 309 |
+
"eval_runtime": 35.6417,
|
| 310 |
+
"eval_samples_per_second": 168.342,
|
| 311 |
+
"eval_steps_per_second": 5.275,
|
| 312 |
+
"step": 422
|
| 313 |
+
},
|
| 314 |
+
{
|
| 315 |
+
"epoch": 1.018957345971564,
|
| 316 |
+
"grad_norm": 14.619269371032715,
|
| 317 |
+
"learning_rate": 3.669885864793679e-05,
|
| 318 |
+
"loss": 1.6787,
|
| 319 |
+
"step": 430
|
| 320 |
+
},
|
| 321 |
+
{
|
| 322 |
+
"epoch": 1.042654028436019,
|
| 323 |
+
"grad_norm": 17.89664077758789,
|
| 324 |
+
"learning_rate": 3.6259877085162426e-05,
|
| 325 |
+
"loss": 1.667,
|
| 326 |
+
"step": 440
|
| 327 |
+
},
|
| 328 |
+
{
|
| 329 |
+
"epoch": 1.066350710900474,
|
| 330 |
+
"grad_norm": 17.098857879638672,
|
| 331 |
+
"learning_rate": 3.582089552238806e-05,
|
| 332 |
+
"loss": 1.5486,
|
| 333 |
+
"step": 450
|
| 334 |
+
},
|
| 335 |
+
{
|
| 336 |
+
"epoch": 1.0900473933649288,
|
| 337 |
+
"grad_norm": 18.231796264648438,
|
| 338 |
+
"learning_rate": 3.53819139596137e-05,
|
| 339 |
+
"loss": 1.5404,
|
| 340 |
+
"step": 460
|
| 341 |
+
},
|
| 342 |
+
{
|
| 343 |
+
"epoch": 1.113744075829384,
|
| 344 |
+
"grad_norm": 18.734926223754883,
|
| 345 |
+
"learning_rate": 3.4942932396839335e-05,
|
| 346 |
+
"loss": 1.6994,
|
| 347 |
+
"step": 470
|
| 348 |
+
},
|
| 349 |
+
{
|
| 350 |
+
"epoch": 1.1374407582938388,
|
| 351 |
+
"grad_norm": 15.583982467651367,
|
| 352 |
+
"learning_rate": 3.450395083406497e-05,
|
| 353 |
+
"loss": 1.5612,
|
| 354 |
+
"step": 480
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 1.161137440758294,
|
| 358 |
+
"grad_norm": 19.229564666748047,
|
| 359 |
+
"learning_rate": 3.406496927129061e-05,
|
| 360 |
+
"loss": 1.4565,
|
| 361 |
+
"step": 490
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"epoch": 1.1848341232227488,
|
| 365 |
+
"grad_norm": 23.661117553710938,
|
| 366 |
+
"learning_rate": 3.3625987708516244e-05,
|
| 367 |
+
"loss": 1.638,
|
| 368 |
+
"step": 500
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 1.2085308056872037,
|
| 372 |
+
"grad_norm": 19.3845272064209,
|
| 373 |
+
"learning_rate": 3.318700614574188e-05,
|
| 374 |
+
"loss": 1.841,
|
| 375 |
+
"step": 510
|
| 376 |
+
},
|
| 377 |
+
{
|
| 378 |
+
"epoch": 1.2322274881516588,
|
| 379 |
+
"grad_norm": 22.151485443115234,
|
| 380 |
+
"learning_rate": 3.274802458296752e-05,
|
| 381 |
+
"loss": 1.8073,
|
| 382 |
+
"step": 520
|
| 383 |
+
},
|
| 384 |
+
{
|
| 385 |
+
"epoch": 1.2559241706161137,
|
| 386 |
+
"grad_norm": 24.08620262145996,
|
| 387 |
+
"learning_rate": 3.230904302019315e-05,
|
| 388 |
+
"loss": 1.523,
|
| 389 |
+
"step": 530
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 1.2796208530805688,
|
| 393 |
+
"grad_norm": 25.005783081054688,
|
| 394 |
+
"learning_rate": 3.187006145741879e-05,
|
| 395 |
+
"loss": 1.5896,
|
| 396 |
+
"step": 540
|
| 397 |
+
},
|
| 398 |
+
{
|
| 399 |
+
"epoch": 1.3033175355450237,
|
| 400 |
+
"grad_norm": 22.422975540161133,
|
| 401 |
+
"learning_rate": 3.1431079894644426e-05,
|
| 402 |
+
"loss": 1.599,
|
| 403 |
+
"step": 550
|
| 404 |
+
},
|
| 405 |
+
{
|
| 406 |
+
"epoch": 1.3270142180094786,
|
| 407 |
+
"grad_norm": 18.511837005615234,
|
| 408 |
+
"learning_rate": 3.099209833187006e-05,
|
| 409 |
+
"loss": 1.621,
|
| 410 |
+
"step": 560
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 1.3507109004739337,
|
| 414 |
+
"grad_norm": 20.40373420715332,
|
| 415 |
+
"learning_rate": 3.0553116769095705e-05,
|
| 416 |
+
"loss": 1.573,
|
| 417 |
+
"step": 570
|
| 418 |
+
},
|
| 419 |
+
{
|
| 420 |
+
"epoch": 1.3744075829383886,
|
| 421 |
+
"grad_norm": 19.02882957458496,
|
| 422 |
+
"learning_rate": 3.0114135206321338e-05,
|
| 423 |
+
"loss": 1.464,
|
| 424 |
+
"step": 580
|
| 425 |
+
},
|
| 426 |
+
{
|
| 427 |
+
"epoch": 1.3981042654028437,
|
| 428 |
+
"grad_norm": 18.40727996826172,
|
| 429 |
+
"learning_rate": 2.967515364354697e-05,
|
| 430 |
+
"loss": 1.5742,
|
| 431 |
+
"step": 590
|
| 432 |
+
},
|
| 433 |
+
{
|
| 434 |
+
"epoch": 1.4218009478672986,
|
| 435 |
+
"grad_norm": 21.194843292236328,
|
| 436 |
+
"learning_rate": 2.923617208077261e-05,
|
| 437 |
+
"loss": 1.334,
|
| 438 |
+
"step": 600
|
| 439 |
+
},
|
| 440 |
+
{
|
| 441 |
+
"epoch": 1.4454976303317535,
|
| 442 |
+
"grad_norm": 18.509262084960938,
|
| 443 |
+
"learning_rate": 2.8797190517998247e-05,
|
| 444 |
+
"loss": 1.5017,
|
| 445 |
+
"step": 610
|
| 446 |
+
},
|
| 447 |
+
{
|
| 448 |
+
"epoch": 1.4691943127962086,
|
| 449 |
+
"grad_norm": 20.656261444091797,
|
| 450 |
+
"learning_rate": 2.835820895522388e-05,
|
| 451 |
+
"loss": 1.6425,
|
| 452 |
+
"step": 620
|
| 453 |
+
},
|
| 454 |
+
{
|
| 455 |
+
"epoch": 1.4928909952606635,
|
| 456 |
+
"grad_norm": 16.49464988708496,
|
| 457 |
+
"learning_rate": 2.791922739244952e-05,
|
| 458 |
+
"loss": 1.4641,
|
| 459 |
+
"step": 630
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 1.5165876777251186,
|
| 463 |
+
"grad_norm": 29.223224639892578,
|
| 464 |
+
"learning_rate": 2.7480245829675156e-05,
|
| 465 |
+
"loss": 1.4755,
|
| 466 |
+
"step": 640
|
| 467 |
+
},
|
| 468 |
+
{
|
| 469 |
+
"epoch": 1.5402843601895735,
|
| 470 |
+
"grad_norm": 22.57363510131836,
|
| 471 |
+
"learning_rate": 2.704126426690079e-05,
|
| 472 |
+
"loss": 1.418,
|
| 473 |
+
"step": 650
|
| 474 |
+
},
|
| 475 |
+
{
|
| 476 |
+
"epoch": 1.5639810426540284,
|
| 477 |
+
"grad_norm": 20.45244026184082,
|
| 478 |
+
"learning_rate": 2.660228270412643e-05,
|
| 479 |
+
"loss": 1.5061,
|
| 480 |
+
"step": 660
|
| 481 |
+
},
|
| 482 |
+
{
|
| 483 |
+
"epoch": 1.5876777251184833,
|
| 484 |
+
"grad_norm": 18.2165470123291,
|
| 485 |
+
"learning_rate": 2.6163301141352066e-05,
|
| 486 |
+
"loss": 1.5323,
|
| 487 |
+
"step": 670
|
| 488 |
+
},
|
| 489 |
+
{
|
| 490 |
+
"epoch": 1.6113744075829384,
|
| 491 |
+
"grad_norm": 15.586666107177734,
|
| 492 |
+
"learning_rate": 2.57243195785777e-05,
|
| 493 |
+
"loss": 1.4886,
|
| 494 |
+
"step": 680
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 1.6350710900473935,
|
| 498 |
+
"grad_norm": 21.27082633972168,
|
| 499 |
+
"learning_rate": 2.5285338015803338e-05,
|
| 500 |
+
"loss": 1.5164,
|
| 501 |
+
"step": 690
|
| 502 |
+
},
|
| 503 |
+
{
|
| 504 |
+
"epoch": 1.6587677725118484,
|
| 505 |
+
"grad_norm": 18.144695281982422,
|
| 506 |
+
"learning_rate": 2.4846356453028975e-05,
|
| 507 |
+
"loss": 1.4823,
|
| 508 |
+
"step": 700
|
| 509 |
+
},
|
| 510 |
+
{
|
| 511 |
+
"epoch": 1.6824644549763033,
|
| 512 |
+
"grad_norm": 18.286558151245117,
|
| 513 |
+
"learning_rate": 2.440737489025461e-05,
|
| 514 |
+
"loss": 1.5464,
|
| 515 |
+
"step": 710
|
| 516 |
+
},
|
| 517 |
+
{
|
| 518 |
+
"epoch": 1.7061611374407581,
|
| 519 |
+
"grad_norm": 16.35245704650879,
|
| 520 |
+
"learning_rate": 2.3968393327480247e-05,
|
| 521 |
+
"loss": 1.4798,
|
| 522 |
+
"step": 720
|
| 523 |
+
},
|
| 524 |
+
{
|
| 525 |
+
"epoch": 1.7298578199052133,
|
| 526 |
+
"grad_norm": 18.115129470825195,
|
| 527 |
+
"learning_rate": 2.3529411764705884e-05,
|
| 528 |
+
"loss": 1.4282,
|
| 529 |
+
"step": 730
|
| 530 |
+
},
|
| 531 |
+
{
|
| 532 |
+
"epoch": 1.7535545023696684,
|
| 533 |
+
"grad_norm": 10.362217903137207,
|
| 534 |
+
"learning_rate": 2.309043020193152e-05,
|
| 535 |
+
"loss": 1.5111,
|
| 536 |
+
"step": 740
|
| 537 |
+
},
|
| 538 |
+
{
|
| 539 |
+
"epoch": 1.7772511848341233,
|
| 540 |
+
"grad_norm": 14.301712989807129,
|
| 541 |
+
"learning_rate": 2.2651448639157156e-05,
|
| 542 |
+
"loss": 1.5408,
|
| 543 |
+
"step": 750
|
| 544 |
+
},
|
| 545 |
+
{
|
| 546 |
+
"epoch": 1.8009478672985781,
|
| 547 |
+
"grad_norm": 11.159360885620117,
|
| 548 |
+
"learning_rate": 2.2212467076382793e-05,
|
| 549 |
+
"loss": 1.3557,
|
| 550 |
+
"step": 760
|
| 551 |
+
},
|
| 552 |
+
{
|
| 553 |
+
"epoch": 1.824644549763033,
|
| 554 |
+
"grad_norm": 22.02513885498047,
|
| 555 |
+
"learning_rate": 2.177348551360843e-05,
|
| 556 |
+
"loss": 1.4304,
|
| 557 |
+
"step": 770
|
| 558 |
+
},
|
| 559 |
+
{
|
| 560 |
+
"epoch": 1.8483412322274881,
|
| 561 |
+
"grad_norm": 15.173885345458984,
|
| 562 |
+
"learning_rate": 2.1334503950834065e-05,
|
| 563 |
+
"loss": 1.5921,
|
| 564 |
+
"step": 780
|
| 565 |
+
},
|
| 566 |
+
{
|
| 567 |
+
"epoch": 1.8720379146919433,
|
| 568 |
+
"grad_norm": 16.821144104003906,
|
| 569 |
+
"learning_rate": 2.0895522388059702e-05,
|
| 570 |
+
"loss": 1.4525,
|
| 571 |
+
"step": 790
|
| 572 |
+
},
|
| 573 |
+
{
|
| 574 |
+
"epoch": 1.8957345971563981,
|
| 575 |
+
"grad_norm": 18.36412811279297,
|
| 576 |
+
"learning_rate": 2.0456540825285338e-05,
|
| 577 |
+
"loss": 1.3811,
|
| 578 |
+
"step": 800
|
| 579 |
+
},
|
| 580 |
+
{
|
| 581 |
+
"epoch": 1.919431279620853,
|
| 582 |
+
"grad_norm": 18.73419189453125,
|
| 583 |
+
"learning_rate": 2.0017559262510978e-05,
|
| 584 |
+
"loss": 1.5829,
|
| 585 |
+
"step": 810
|
| 586 |
+
},
|
| 587 |
+
{
|
| 588 |
+
"epoch": 1.943127962085308,
|
| 589 |
+
"grad_norm": 15.17447566986084,
|
| 590 |
+
"learning_rate": 1.957857769973661e-05,
|
| 591 |
+
"loss": 1.317,
|
| 592 |
+
"step": 820
|
| 593 |
+
},
|
| 594 |
+
{
|
| 595 |
+
"epoch": 1.966824644549763,
|
| 596 |
+
"grad_norm": 18.464017868041992,
|
| 597 |
+
"learning_rate": 1.9139596136962247e-05,
|
| 598 |
+
"loss": 1.5998,
|
| 599 |
+
"step": 830
|
| 600 |
+
},
|
| 601 |
+
{
|
| 602 |
+
"epoch": 1.9905213270142181,
|
| 603 |
+
"grad_norm": 16.639902114868164,
|
| 604 |
+
"learning_rate": 1.8700614574187887e-05,
|
| 605 |
+
"loss": 1.5976,
|
| 606 |
+
"step": 840
|
| 607 |
+
},
|
| 608 |
+
{
|
| 609 |
+
"epoch": 2.0,
|
| 610 |
+
"eval_accuracy": 0.9868333333333333,
|
| 611 |
+
"eval_loss": 0.04062141478061676,
|
| 612 |
+
"eval_runtime": 35.7487,
|
| 613 |
+
"eval_samples_per_second": 167.838,
|
| 614 |
+
"eval_steps_per_second": 5.259,
|
| 615 |
+
"step": 844
|
| 616 |
+
},
|
| 617 |
+
{
|
| 618 |
+
"epoch": 2.014218009478673,
|
| 619 |
+
"grad_norm": 14.642781257629395,
|
| 620 |
+
"learning_rate": 1.826163301141352e-05,
|
| 621 |
+
"loss": 1.4676,
|
| 622 |
+
"step": 850
|
| 623 |
+
},
|
| 624 |
+
{
|
| 625 |
+
"epoch": 2.037914691943128,
|
| 626 |
+
"grad_norm": 18.012798309326172,
|
| 627 |
+
"learning_rate": 1.7822651448639156e-05,
|
| 628 |
+
"loss": 1.5122,
|
| 629 |
+
"step": 860
|
| 630 |
+
},
|
| 631 |
+
{
|
| 632 |
+
"epoch": 2.061611374407583,
|
| 633 |
+
"grad_norm": 19.37873077392578,
|
| 634 |
+
"learning_rate": 1.7383669885864796e-05,
|
| 635 |
+
"loss": 1.3235,
|
| 636 |
+
"step": 870
|
| 637 |
+
},
|
| 638 |
+
{
|
| 639 |
+
"epoch": 2.085308056872038,
|
| 640 |
+
"grad_norm": 16.801088333129883,
|
| 641 |
+
"learning_rate": 1.694468832309043e-05,
|
| 642 |
+
"loss": 1.359,
|
| 643 |
+
"step": 880
|
| 644 |
+
},
|
| 645 |
+
{
|
| 646 |
+
"epoch": 2.109004739336493,
|
| 647 |
+
"grad_norm": 18.08489418029785,
|
| 648 |
+
"learning_rate": 1.6505706760316065e-05,
|
| 649 |
+
"loss": 1.3441,
|
| 650 |
+
"step": 890
|
| 651 |
+
},
|
| 652 |
+
{
|
| 653 |
+
"epoch": 2.132701421800948,
|
| 654 |
+
"grad_norm": 14.385065078735352,
|
| 655 |
+
"learning_rate": 1.6066725197541705e-05,
|
| 656 |
+
"loss": 1.3717,
|
| 657 |
+
"step": 900
|
| 658 |
+
},
|
| 659 |
+
{
|
| 660 |
+
"epoch": 2.156398104265403,
|
| 661 |
+
"grad_norm": 14.971292495727539,
|
| 662 |
+
"learning_rate": 1.562774363476734e-05,
|
| 663 |
+
"loss": 1.4493,
|
| 664 |
+
"step": 910
|
| 665 |
+
},
|
| 666 |
+
{
|
| 667 |
+
"epoch": 2.1800947867298577,
|
| 668 |
+
"grad_norm": 15.784900665283203,
|
| 669 |
+
"learning_rate": 1.5188762071992976e-05,
|
| 670 |
+
"loss": 1.3515,
|
| 671 |
+
"step": 920
|
| 672 |
+
},
|
| 673 |
+
{
|
| 674 |
+
"epoch": 2.2037914691943126,
|
| 675 |
+
"grad_norm": 18.48076820373535,
|
| 676 |
+
"learning_rate": 1.4749780509218614e-05,
|
| 677 |
+
"loss": 1.5577,
|
| 678 |
+
"step": 930
|
| 679 |
+
},
|
| 680 |
+
{
|
| 681 |
+
"epoch": 2.227488151658768,
|
| 682 |
+
"grad_norm": 17.10760498046875,
|
| 683 |
+
"learning_rate": 1.431079894644425e-05,
|
| 684 |
+
"loss": 1.4396,
|
| 685 |
+
"step": 940
|
| 686 |
+
},
|
| 687 |
+
{
|
| 688 |
+
"epoch": 2.251184834123223,
|
| 689 |
+
"grad_norm": 25.413055419921875,
|
| 690 |
+
"learning_rate": 1.3871817383669885e-05,
|
| 691 |
+
"loss": 1.36,
|
| 692 |
+
"step": 950
|
| 693 |
+
},
|
| 694 |
+
{
|
| 695 |
+
"epoch": 2.2748815165876777,
|
| 696 |
+
"grad_norm": 25.242835998535156,
|
| 697 |
+
"learning_rate": 1.3432835820895523e-05,
|
| 698 |
+
"loss": 1.4158,
|
| 699 |
+
"step": 960
|
| 700 |
+
},
|
| 701 |
+
{
|
| 702 |
+
"epoch": 2.2985781990521326,
|
| 703 |
+
"grad_norm": 13.354720115661621,
|
| 704 |
+
"learning_rate": 1.299385425812116e-05,
|
| 705 |
+
"loss": 1.2912,
|
| 706 |
+
"step": 970
|
| 707 |
+
},
|
| 708 |
+
{
|
| 709 |
+
"epoch": 2.322274881516588,
|
| 710 |
+
"grad_norm": 19.005489349365234,
|
| 711 |
+
"learning_rate": 1.2554872695346794e-05,
|
| 712 |
+
"loss": 1.3714,
|
| 713 |
+
"step": 980
|
| 714 |
+
},
|
| 715 |
+
{
|
| 716 |
+
"epoch": 2.345971563981043,
|
| 717 |
+
"grad_norm": 14.166816711425781,
|
| 718 |
+
"learning_rate": 1.2115891132572432e-05,
|
| 719 |
+
"loss": 1.5125,
|
| 720 |
+
"step": 990
|
| 721 |
+
},
|
| 722 |
+
{
|
| 723 |
+
"epoch": 2.3696682464454977,
|
| 724 |
+
"grad_norm": 15.295882225036621,
|
| 725 |
+
"learning_rate": 1.1676909569798069e-05,
|
| 726 |
+
"loss": 1.2683,
|
| 727 |
+
"step": 1000
|
| 728 |
+
},
|
| 729 |
+
{
|
| 730 |
+
"epoch": 2.3933649289099526,
|
| 731 |
+
"grad_norm": 15.587239265441895,
|
| 732 |
+
"learning_rate": 1.1237928007023705e-05,
|
| 733 |
+
"loss": 1.4069,
|
| 734 |
+
"step": 1010
|
| 735 |
+
},
|
| 736 |
+
{
|
| 737 |
+
"epoch": 2.4170616113744074,
|
| 738 |
+
"grad_norm": 14.5875883102417,
|
| 739 |
+
"learning_rate": 1.0798946444249341e-05,
|
| 740 |
+
"loss": 1.2272,
|
| 741 |
+
"step": 1020
|
| 742 |
+
},
|
| 743 |
+
{
|
| 744 |
+
"epoch": 2.4407582938388623,
|
| 745 |
+
"grad_norm": 18.715368270874023,
|
| 746 |
+
"learning_rate": 1.035996488147498e-05,
|
| 747 |
+
"loss": 1.2901,
|
| 748 |
+
"step": 1030
|
| 749 |
+
},
|
| 750 |
+
{
|
| 751 |
+
"epoch": 2.4644549763033177,
|
| 752 |
+
"grad_norm": 16.098590850830078,
|
| 753 |
+
"learning_rate": 9.920983318700614e-06,
|
| 754 |
+
"loss": 1.3823,
|
| 755 |
+
"step": 1040
|
| 756 |
+
},
|
| 757 |
+
{
|
| 758 |
+
"epoch": 2.4881516587677726,
|
| 759 |
+
"grad_norm": 13.056846618652344,
|
| 760 |
+
"learning_rate": 9.482001755926252e-06,
|
| 761 |
+
"loss": 1.377,
|
| 762 |
+
"step": 1050
|
| 763 |
+
},
|
| 764 |
+
{
|
| 765 |
+
"epoch": 2.5118483412322274,
|
| 766 |
+
"grad_norm": 20.628856658935547,
|
| 767 |
+
"learning_rate": 9.043020193151889e-06,
|
| 768 |
+
"loss": 1.2071,
|
| 769 |
+
"step": 1060
|
| 770 |
+
},
|
| 771 |
+
{
|
| 772 |
+
"epoch": 2.5355450236966823,
|
| 773 |
+
"grad_norm": 16.558698654174805,
|
| 774 |
+
"learning_rate": 8.604038630377525e-06,
|
| 775 |
+
"loss": 1.431,
|
| 776 |
+
"step": 1070
|
| 777 |
+
},
|
| 778 |
+
{
|
| 779 |
+
"epoch": 2.5592417061611377,
|
| 780 |
+
"grad_norm": 15.537813186645508,
|
| 781 |
+
"learning_rate": 8.165057067603161e-06,
|
| 782 |
+
"loss": 1.2182,
|
| 783 |
+
"step": 1080
|
| 784 |
+
},
|
| 785 |
+
{
|
| 786 |
+
"epoch": 2.5829383886255926,
|
| 787 |
+
"grad_norm": 17.30754280090332,
|
| 788 |
+
"learning_rate": 7.726075504828798e-06,
|
| 789 |
+
"loss": 1.3382,
|
| 790 |
+
"step": 1090
|
| 791 |
+
},
|
| 792 |
+
{
|
| 793 |
+
"epoch": 2.6066350710900474,
|
| 794 |
+
"grad_norm": 14.758044242858887,
|
| 795 |
+
"learning_rate": 7.287093942054433e-06,
|
| 796 |
+
"loss": 1.3603,
|
| 797 |
+
"step": 1100
|
| 798 |
+
},
|
| 799 |
+
{
|
| 800 |
+
"epoch": 2.6303317535545023,
|
| 801 |
+
"grad_norm": 17.097063064575195,
|
| 802 |
+
"learning_rate": 6.84811237928007e-06,
|
| 803 |
+
"loss": 1.3597,
|
| 804 |
+
"step": 1110
|
| 805 |
+
},
|
| 806 |
+
{
|
| 807 |
+
"epoch": 2.654028436018957,
|
| 808 |
+
"grad_norm": 15.043681144714355,
|
| 809 |
+
"learning_rate": 6.409130816505707e-06,
|
| 810 |
+
"loss": 1.2835,
|
| 811 |
+
"step": 1120
|
| 812 |
+
},
|
| 813 |
+
{
|
| 814 |
+
"epoch": 2.677725118483412,
|
| 815 |
+
"grad_norm": 17.108274459838867,
|
| 816 |
+
"learning_rate": 5.970149253731343e-06,
|
| 817 |
+
"loss": 1.2077,
|
| 818 |
+
"step": 1130
|
| 819 |
+
},
|
| 820 |
+
{
|
| 821 |
+
"epoch": 2.7014218009478674,
|
| 822 |
+
"grad_norm": 15.565308570861816,
|
| 823 |
+
"learning_rate": 5.53116769095698e-06,
|
| 824 |
+
"loss": 1.1986,
|
| 825 |
+
"step": 1140
|
| 826 |
+
},
|
| 827 |
+
{
|
| 828 |
+
"epoch": 2.7251184834123223,
|
| 829 |
+
"grad_norm": 17.27542495727539,
|
| 830 |
+
"learning_rate": 5.092186128182617e-06,
|
| 831 |
+
"loss": 1.3716,
|
| 832 |
+
"step": 1150
|
| 833 |
+
},
|
| 834 |
+
{
|
| 835 |
+
"epoch": 2.748815165876777,
|
| 836 |
+
"grad_norm": 18.529542922973633,
|
| 837 |
+
"learning_rate": 4.653204565408253e-06,
|
| 838 |
+
"loss": 1.2624,
|
| 839 |
+
"step": 1160
|
| 840 |
+
},
|
| 841 |
+
{
|
| 842 |
+
"epoch": 2.772511848341232,
|
| 843 |
+
"grad_norm": 13.805538177490234,
|
| 844 |
+
"learning_rate": 4.2142230026338894e-06,
|
| 845 |
+
"loss": 1.3426,
|
| 846 |
+
"step": 1170
|
| 847 |
+
},
|
| 848 |
+
{
|
| 849 |
+
"epoch": 2.7962085308056874,
|
| 850 |
+
"grad_norm": 17.240198135375977,
|
| 851 |
+
"learning_rate": 3.775241439859526e-06,
|
| 852 |
+
"loss": 1.3494,
|
| 853 |
+
"step": 1180
|
| 854 |
+
},
|
| 855 |
+
{
|
| 856 |
+
"epoch": 2.8199052132701423,
|
| 857 |
+
"grad_norm": 14.451141357421875,
|
| 858 |
+
"learning_rate": 3.3362598770851626e-06,
|
| 859 |
+
"loss": 1.3952,
|
| 860 |
+
"step": 1190
|
| 861 |
+
},
|
| 862 |
+
{
|
| 863 |
+
"epoch": 2.843601895734597,
|
| 864 |
+
"grad_norm": 13.656115531921387,
|
| 865 |
+
"learning_rate": 2.897278314310799e-06,
|
| 866 |
+
"loss": 1.3488,
|
| 867 |
+
"step": 1200
|
| 868 |
+
},
|
| 869 |
+
{
|
| 870 |
+
"epoch": 2.867298578199052,
|
| 871 |
+
"grad_norm": 13.846714973449707,
|
| 872 |
+
"learning_rate": 2.4582967515364357e-06,
|
| 873 |
+
"loss": 1.2827,
|
| 874 |
+
"step": 1210
|
| 875 |
+
},
|
| 876 |
+
{
|
| 877 |
+
"epoch": 2.890995260663507,
|
| 878 |
+
"grad_norm": 22.159263610839844,
|
| 879 |
+
"learning_rate": 2.019315188762072e-06,
|
| 880 |
+
"loss": 1.434,
|
| 881 |
+
"step": 1220
|
| 882 |
+
},
|
| 883 |
+
{
|
| 884 |
+
"epoch": 2.914691943127962,
|
| 885 |
+
"grad_norm": 20.175447463989258,
|
| 886 |
+
"learning_rate": 1.5803336259877086e-06,
|
| 887 |
+
"loss": 1.2202,
|
| 888 |
+
"step": 1230
|
| 889 |
+
},
|
| 890 |
+
{
|
| 891 |
+
"epoch": 2.938388625592417,
|
| 892 |
+
"grad_norm": 17.73303985595703,
|
| 893 |
+
"learning_rate": 1.141352063213345e-06,
|
| 894 |
+
"loss": 1.4187,
|
| 895 |
+
"step": 1240
|
| 896 |
+
},
|
| 897 |
+
{
|
| 898 |
+
"epoch": 2.962085308056872,
|
| 899 |
+
"grad_norm": 17.480005264282227,
|
| 900 |
+
"learning_rate": 7.023705004389816e-07,
|
| 901 |
+
"loss": 1.2624,
|
| 902 |
+
"step": 1250
|
| 903 |
+
},
|
| 904 |
+
{
|
| 905 |
+
"epoch": 2.985781990521327,
|
| 906 |
+
"grad_norm": 12.864943504333496,
|
| 907 |
+
"learning_rate": 2.633889376646181e-07,
|
| 908 |
+
"loss": 1.1605,
|
| 909 |
+
"step": 1260
|
| 910 |
+
},
|
| 911 |
+
{
|
| 912 |
+
"epoch": 3.0,
|
| 913 |
+
"eval_accuracy": 0.9886666666666667,
|
| 914 |
+
"eval_loss": 0.0382872000336647,
|
| 915 |
+
"eval_runtime": 36.235,
|
| 916 |
+
"eval_samples_per_second": 165.586,
|
| 917 |
+
"eval_steps_per_second": 5.188,
|
| 918 |
+
"step": 1266
|
| 919 |
+
},
|
| 920 |
+
{
|
| 921 |
+
"epoch": 3.0,
|
| 922 |
+
"step": 1266,
|
| 923 |
+
"total_flos": 4.027570334613504e+18,
|
| 924 |
+
"train_loss": 1.9390247602598363,
|
| 925 |
+
"train_runtime": 2519.9488,
|
| 926 |
+
"train_samples_per_second": 64.287,
|
| 927 |
+
"train_steps_per_second": 0.502
|
| 928 |
+
}
|
| 929 |
+
],
|
| 930 |
+
"logging_steps": 10,
|
| 931 |
+
"max_steps": 1266,
|
| 932 |
+
"num_input_tokens_seen": 0,
|
| 933 |
+
"num_train_epochs": 3,
|
| 934 |
+
"save_steps": 500,
|
| 935 |
+
"stateful_callbacks": {
|
| 936 |
+
"TrainerControl": {
|
| 937 |
+
"args": {
|
| 938 |
+
"should_epoch_stop": false,
|
| 939 |
+
"should_evaluate": false,
|
| 940 |
+
"should_log": false,
|
| 941 |
+
"should_save": true,
|
| 942 |
+
"should_training_stop": true
|
| 943 |
+
},
|
| 944 |
+
"attributes": {}
|
| 945 |
+
}
|
| 946 |
+
},
|
| 947 |
+
"total_flos": 4.027570334613504e+18,
|
| 948 |
+
"train_batch_size": 32,
|
| 949 |
+
"trial_name": null,
|
| 950 |
+
"trial_params": null
|
| 951 |
+
}
|