End of training
Browse files
README.md
CHANGED
|
@@ -22,7 +22,7 @@ model-index:
|
|
| 22 |
metrics:
|
| 23 |
- name: Accuracy
|
| 24 |
type: accuracy
|
| 25 |
-
value: 0.
|
| 26 |
---
|
| 27 |
|
| 28 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
@@ -32,8 +32,8 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 32 |
|
| 33 |
This model is a fine-tuned version of [microsoft/swin-small-patch4-window7-224](https://huggingface.co/microsoft/swin-small-patch4-window7-224) on the imagefolder dataset.
|
| 34 |
It achieves the following results on the evaluation set:
|
| 35 |
-
- Loss:
|
| 36 |
-
- Accuracy: 0.
|
| 37 |
|
| 38 |
## Model description
|
| 39 |
|
|
|
|
| 22 |
metrics:
|
| 23 |
- name: Accuracy
|
| 24 |
type: accuracy
|
| 25 |
+
value: 0.5454545454545454
|
| 26 |
---
|
| 27 |
|
| 28 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
|
|
| 32 |
|
| 33 |
This model is a fine-tuned version of [microsoft/swin-small-patch4-window7-224](https://huggingface.co/microsoft/swin-small-patch4-window7-224) on the imagefolder dataset.
|
| 34 |
It achieves the following results on the evaluation set:
|
| 35 |
+
- Loss: 1.9417
|
| 36 |
+
- Accuracy: 0.5455
|
| 37 |
|
| 38 |
## Model description
|
| 39 |
|
all_results.json
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 29.387755102040817,
|
| 3 |
+
"eval_accuracy": 0.5454545454545454,
|
| 4 |
+
"eval_loss": 1.9417484998703003,
|
| 5 |
+
"eval_runtime": 4.7953,
|
| 6 |
+
"eval_samples_per_second": 4.588,
|
| 7 |
+
"eval_steps_per_second": 2.294,
|
| 8 |
+
"total_flos": 2.5282019639994163e+17,
|
| 9 |
+
"train_loss": 0.5089964731906851,
|
| 10 |
+
"train_runtime": 1181.0861,
|
| 11 |
+
"train_samples_per_second": 4.953,
|
| 12 |
+
"train_steps_per_second": 0.61
|
| 13 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 29.387755102040817,
|
| 3 |
+
"eval_accuracy": 0.5454545454545454,
|
| 4 |
+
"eval_loss": 1.9417484998703003,
|
| 5 |
+
"eval_runtime": 4.7953,
|
| 6 |
+
"eval_samples_per_second": 4.588,
|
| 7 |
+
"eval_steps_per_second": 2.294
|
| 8 |
+
}
|
runs/May31_00-12-13_ip-10-110-161-167.ap-southeast-2.compute.internal/events.out.tfevents.1717115547.ip-10-110-161-167.ap-southeast-2.compute.internal.9572.1
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c56b38579ab9e9cb2d80d84d5400df6a5aede41c17e68e213b9ccd149699322b
|
| 3 |
+
size 411
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 29.387755102040817,
|
| 3 |
+
"total_flos": 2.5282019639994163e+17,
|
| 4 |
+
"train_loss": 0.5089964731906851,
|
| 5 |
+
"train_runtime": 1181.0861,
|
| 6 |
+
"train_samples_per_second": 4.953,
|
| 7 |
+
"train_steps_per_second": 0.61
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,816 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.5454545454545454,
|
| 3 |
+
"best_model_checkpoint": "swin-small-patch4-window7-224-finetuned-isic217/checkpoint-269",
|
| 4 |
+
"epoch": 29.387755102040817,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 720,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.40816326530612246,
|
| 13 |
+
"grad_norm": 15.3554048538208,
|
| 14 |
+
"learning_rate": 6.944444444444445e-06,
|
| 15 |
+
"loss": 2.1982,
|
| 16 |
+
"step": 10
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.8163265306122449,
|
| 20 |
+
"grad_norm": 12.736021041870117,
|
| 21 |
+
"learning_rate": 1.388888888888889e-05,
|
| 22 |
+
"loss": 2.1844,
|
| 23 |
+
"step": 20
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.9795918367346939,
|
| 27 |
+
"eval_accuracy": 0.13636363636363635,
|
| 28 |
+
"eval_loss": 2.1102521419525146,
|
| 29 |
+
"eval_runtime": 4.2332,
|
| 30 |
+
"eval_samples_per_second": 5.197,
|
| 31 |
+
"eval_steps_per_second": 2.599,
|
| 32 |
+
"step": 24
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 1.2244897959183674,
|
| 36 |
+
"grad_norm": 20.908735275268555,
|
| 37 |
+
"learning_rate": 2.0833333333333336e-05,
|
| 38 |
+
"loss": 2.0984,
|
| 39 |
+
"step": 30
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 1.6326530612244898,
|
| 43 |
+
"grad_norm": 17.49510955810547,
|
| 44 |
+
"learning_rate": 2.777777777777778e-05,
|
| 45 |
+
"loss": 2.0018,
|
| 46 |
+
"step": 40
|
| 47 |
+
},
|
| 48 |
+
{
|
| 49 |
+
"epoch": 2.0,
|
| 50 |
+
"eval_accuracy": 0.2727272727272727,
|
| 51 |
+
"eval_loss": 1.8736993074417114,
|
| 52 |
+
"eval_runtime": 4.0949,
|
| 53 |
+
"eval_samples_per_second": 5.373,
|
| 54 |
+
"eval_steps_per_second": 2.686,
|
| 55 |
+
"step": 49
|
| 56 |
+
},
|
| 57 |
+
{
|
| 58 |
+
"epoch": 2.0408163265306123,
|
| 59 |
+
"grad_norm": 16.44957733154297,
|
| 60 |
+
"learning_rate": 3.472222222222222e-05,
|
| 61 |
+
"loss": 1.9104,
|
| 62 |
+
"step": 50
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 2.4489795918367347,
|
| 66 |
+
"grad_norm": 25.78542137145996,
|
| 67 |
+
"learning_rate": 4.166666666666667e-05,
|
| 68 |
+
"loss": 1.6759,
|
| 69 |
+
"step": 60
|
| 70 |
+
},
|
| 71 |
+
{
|
| 72 |
+
"epoch": 2.857142857142857,
|
| 73 |
+
"grad_norm": 35.257904052734375,
|
| 74 |
+
"learning_rate": 4.8611111111111115e-05,
|
| 75 |
+
"loss": 1.6474,
|
| 76 |
+
"step": 70
|
| 77 |
+
},
|
| 78 |
+
{
|
| 79 |
+
"epoch": 2.979591836734694,
|
| 80 |
+
"eval_accuracy": 0.2727272727272727,
|
| 81 |
+
"eval_loss": 1.901879906654358,
|
| 82 |
+
"eval_runtime": 3.9033,
|
| 83 |
+
"eval_samples_per_second": 5.636,
|
| 84 |
+
"eval_steps_per_second": 2.818,
|
| 85 |
+
"step": 73
|
| 86 |
+
},
|
| 87 |
+
{
|
| 88 |
+
"epoch": 3.2653061224489797,
|
| 89 |
+
"grad_norm": 24.751829147338867,
|
| 90 |
+
"learning_rate": 4.938271604938271e-05,
|
| 91 |
+
"loss": 1.5587,
|
| 92 |
+
"step": 80
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 3.673469387755102,
|
| 96 |
+
"grad_norm": 29.13814353942871,
|
| 97 |
+
"learning_rate": 4.8611111111111115e-05,
|
| 98 |
+
"loss": 1.3757,
|
| 99 |
+
"step": 90
|
| 100 |
+
},
|
| 101 |
+
{
|
| 102 |
+
"epoch": 4.0,
|
| 103 |
+
"eval_accuracy": 0.36363636363636365,
|
| 104 |
+
"eval_loss": 1.7486611604690552,
|
| 105 |
+
"eval_runtime": 4.8083,
|
| 106 |
+
"eval_samples_per_second": 4.575,
|
| 107 |
+
"eval_steps_per_second": 2.288,
|
| 108 |
+
"step": 98
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 4.081632653061225,
|
| 112 |
+
"grad_norm": 24.104280471801758,
|
| 113 |
+
"learning_rate": 4.783950617283951e-05,
|
| 114 |
+
"loss": 1.3791,
|
| 115 |
+
"step": 100
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 4.489795918367347,
|
| 119 |
+
"grad_norm": 24.011398315429688,
|
| 120 |
+
"learning_rate": 4.70679012345679e-05,
|
| 121 |
+
"loss": 1.0546,
|
| 122 |
+
"step": 110
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 4.8979591836734695,
|
| 126 |
+
"grad_norm": 18.644390106201172,
|
| 127 |
+
"learning_rate": 4.62962962962963e-05,
|
| 128 |
+
"loss": 1.1526,
|
| 129 |
+
"step": 120
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 4.979591836734694,
|
| 133 |
+
"eval_accuracy": 0.4090909090909091,
|
| 134 |
+
"eval_loss": 1.7576110363006592,
|
| 135 |
+
"eval_runtime": 3.8964,
|
| 136 |
+
"eval_samples_per_second": 5.646,
|
| 137 |
+
"eval_steps_per_second": 2.823,
|
| 138 |
+
"step": 122
|
| 139 |
+
},
|
| 140 |
+
{
|
| 141 |
+
"epoch": 5.3061224489795915,
|
| 142 |
+
"grad_norm": 19.769208908081055,
|
| 143 |
+
"learning_rate": 4.5524691358024696e-05,
|
| 144 |
+
"loss": 0.7525,
|
| 145 |
+
"step": 130
|
| 146 |
+
},
|
| 147 |
+
{
|
| 148 |
+
"epoch": 5.714285714285714,
|
| 149 |
+
"grad_norm": 28.662109375,
|
| 150 |
+
"learning_rate": 4.4753086419753084e-05,
|
| 151 |
+
"loss": 0.9161,
|
| 152 |
+
"step": 140
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 6.0,
|
| 156 |
+
"eval_accuracy": 0.5,
|
| 157 |
+
"eval_loss": 1.5886327028274536,
|
| 158 |
+
"eval_runtime": 4.047,
|
| 159 |
+
"eval_samples_per_second": 5.436,
|
| 160 |
+
"eval_steps_per_second": 2.718,
|
| 161 |
+
"step": 147
|
| 162 |
+
},
|
| 163 |
+
{
|
| 164 |
+
"epoch": 6.122448979591836,
|
| 165 |
+
"grad_norm": 24.809764862060547,
|
| 166 |
+
"learning_rate": 4.3981481481481486e-05,
|
| 167 |
+
"loss": 0.8674,
|
| 168 |
+
"step": 150
|
| 169 |
+
},
|
| 170 |
+
{
|
| 171 |
+
"epoch": 6.530612244897959,
|
| 172 |
+
"grad_norm": 41.678409576416016,
|
| 173 |
+
"learning_rate": 4.3209876543209875e-05,
|
| 174 |
+
"loss": 0.8796,
|
| 175 |
+
"step": 160
|
| 176 |
+
},
|
| 177 |
+
{
|
| 178 |
+
"epoch": 6.938775510204081,
|
| 179 |
+
"grad_norm": 37.33673095703125,
|
| 180 |
+
"learning_rate": 4.243827160493827e-05,
|
| 181 |
+
"loss": 0.7568,
|
| 182 |
+
"step": 170
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 6.979591836734694,
|
| 186 |
+
"eval_accuracy": 0.45454545454545453,
|
| 187 |
+
"eval_loss": 1.8934894800186157,
|
| 188 |
+
"eval_runtime": 3.9506,
|
| 189 |
+
"eval_samples_per_second": 5.569,
|
| 190 |
+
"eval_steps_per_second": 2.784,
|
| 191 |
+
"step": 171
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 7.346938775510204,
|
| 195 |
+
"grad_norm": 22.92353057861328,
|
| 196 |
+
"learning_rate": 4.166666666666667e-05,
|
| 197 |
+
"loss": 0.6483,
|
| 198 |
+
"step": 180
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 7.755102040816326,
|
| 202 |
+
"grad_norm": 16.21958351135254,
|
| 203 |
+
"learning_rate": 4.089506172839506e-05,
|
| 204 |
+
"loss": 0.4024,
|
| 205 |
+
"step": 190
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 8.0,
|
| 209 |
+
"eval_accuracy": 0.45454545454545453,
|
| 210 |
+
"eval_loss": 1.6766648292541504,
|
| 211 |
+
"eval_runtime": 4.0618,
|
| 212 |
+
"eval_samples_per_second": 5.416,
|
| 213 |
+
"eval_steps_per_second": 2.708,
|
| 214 |
+
"step": 196
|
| 215 |
+
},
|
| 216 |
+
{
|
| 217 |
+
"epoch": 8.16326530612245,
|
| 218 |
+
"grad_norm": 22.85525894165039,
|
| 219 |
+
"learning_rate": 4.012345679012346e-05,
|
| 220 |
+
"loss": 0.6058,
|
| 221 |
+
"step": 200
|
| 222 |
+
},
|
| 223 |
+
{
|
| 224 |
+
"epoch": 8.571428571428571,
|
| 225 |
+
"grad_norm": 13.849242210388184,
|
| 226 |
+
"learning_rate": 3.935185185185186e-05,
|
| 227 |
+
"loss": 0.4311,
|
| 228 |
+
"step": 210
|
| 229 |
+
},
|
| 230 |
+
{
|
| 231 |
+
"epoch": 8.979591836734693,
|
| 232 |
+
"grad_norm": 23.329771041870117,
|
| 233 |
+
"learning_rate": 3.8580246913580246e-05,
|
| 234 |
+
"loss": 0.814,
|
| 235 |
+
"step": 220
|
| 236 |
+
},
|
| 237 |
+
{
|
| 238 |
+
"epoch": 8.979591836734693,
|
| 239 |
+
"eval_accuracy": 0.36363636363636365,
|
| 240 |
+
"eval_loss": 1.7112053632736206,
|
| 241 |
+
"eval_runtime": 3.9427,
|
| 242 |
+
"eval_samples_per_second": 5.58,
|
| 243 |
+
"eval_steps_per_second": 2.79,
|
| 244 |
+
"step": 220
|
| 245 |
+
},
|
| 246 |
+
{
|
| 247 |
+
"epoch": 9.387755102040817,
|
| 248 |
+
"grad_norm": 24.695878982543945,
|
| 249 |
+
"learning_rate": 3.780864197530865e-05,
|
| 250 |
+
"loss": 0.561,
|
| 251 |
+
"step": 230
|
| 252 |
+
},
|
| 253 |
+
{
|
| 254 |
+
"epoch": 9.795918367346939,
|
| 255 |
+
"grad_norm": 21.95734405517578,
|
| 256 |
+
"learning_rate": 3.7037037037037037e-05,
|
| 257 |
+
"loss": 0.4346,
|
| 258 |
+
"step": 240
|
| 259 |
+
},
|
| 260 |
+
{
|
| 261 |
+
"epoch": 10.0,
|
| 262 |
+
"eval_accuracy": 0.4090909090909091,
|
| 263 |
+
"eval_loss": 1.9364336729049683,
|
| 264 |
+
"eval_runtime": 4.1094,
|
| 265 |
+
"eval_samples_per_second": 5.354,
|
| 266 |
+
"eval_steps_per_second": 2.677,
|
| 267 |
+
"step": 245
|
| 268 |
+
},
|
| 269 |
+
{
|
| 270 |
+
"epoch": 10.204081632653061,
|
| 271 |
+
"grad_norm": 9.583707809448242,
|
| 272 |
+
"learning_rate": 3.626543209876543e-05,
|
| 273 |
+
"loss": 0.316,
|
| 274 |
+
"step": 250
|
| 275 |
+
},
|
| 276 |
+
{
|
| 277 |
+
"epoch": 10.612244897959183,
|
| 278 |
+
"grad_norm": 3.692790985107422,
|
| 279 |
+
"learning_rate": 3.5493827160493834e-05,
|
| 280 |
+
"loss": 0.3456,
|
| 281 |
+
"step": 260
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 10.979591836734693,
|
| 285 |
+
"eval_accuracy": 0.5454545454545454,
|
| 286 |
+
"eval_loss": 1.9417484998703003,
|
| 287 |
+
"eval_runtime": 3.9243,
|
| 288 |
+
"eval_samples_per_second": 5.606,
|
| 289 |
+
"eval_steps_per_second": 2.803,
|
| 290 |
+
"step": 269
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 11.020408163265307,
|
| 294 |
+
"grad_norm": 20.5589599609375,
|
| 295 |
+
"learning_rate": 3.472222222222222e-05,
|
| 296 |
+
"loss": 0.4472,
|
| 297 |
+
"step": 270
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 11.428571428571429,
|
| 301 |
+
"grad_norm": 25.139612197875977,
|
| 302 |
+
"learning_rate": 3.395061728395062e-05,
|
| 303 |
+
"loss": 0.5478,
|
| 304 |
+
"step": 280
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 11.83673469387755,
|
| 308 |
+
"grad_norm": 21.445762634277344,
|
| 309 |
+
"learning_rate": 3.317901234567901e-05,
|
| 310 |
+
"loss": 0.228,
|
| 311 |
+
"step": 290
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 12.0,
|
| 315 |
+
"eval_accuracy": 0.4090909090909091,
|
| 316 |
+
"eval_loss": 2.156919240951538,
|
| 317 |
+
"eval_runtime": 4.1142,
|
| 318 |
+
"eval_samples_per_second": 5.347,
|
| 319 |
+
"eval_steps_per_second": 2.674,
|
| 320 |
+
"step": 294
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 12.244897959183673,
|
| 324 |
+
"grad_norm": 0.8465752601623535,
|
| 325 |
+
"learning_rate": 3.240740740740741e-05,
|
| 326 |
+
"loss": 0.1284,
|
| 327 |
+
"step": 300
|
| 328 |
+
},
|
| 329 |
+
{
|
| 330 |
+
"epoch": 12.653061224489797,
|
| 331 |
+
"grad_norm": 1.3978017568588257,
|
| 332 |
+
"learning_rate": 3.16358024691358e-05,
|
| 333 |
+
"loss": 0.1681,
|
| 334 |
+
"step": 310
|
| 335 |
+
},
|
| 336 |
+
{
|
| 337 |
+
"epoch": 12.979591836734693,
|
| 338 |
+
"eval_accuracy": 0.45454545454545453,
|
| 339 |
+
"eval_loss": 2.056508779525757,
|
| 340 |
+
"eval_runtime": 3.9483,
|
| 341 |
+
"eval_samples_per_second": 5.572,
|
| 342 |
+
"eval_steps_per_second": 2.786,
|
| 343 |
+
"step": 318
|
| 344 |
+
},
|
| 345 |
+
{
|
| 346 |
+
"epoch": 13.061224489795919,
|
| 347 |
+
"grad_norm": 0.8498027920722961,
|
| 348 |
+
"learning_rate": 3.08641975308642e-05,
|
| 349 |
+
"loss": 0.396,
|
| 350 |
+
"step": 320
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 13.46938775510204,
|
| 354 |
+
"grad_norm": 15.11242961883545,
|
| 355 |
+
"learning_rate": 3.0092592592592593e-05,
|
| 356 |
+
"loss": 0.2593,
|
| 357 |
+
"step": 330
|
| 358 |
+
},
|
| 359 |
+
{
|
| 360 |
+
"epoch": 13.877551020408163,
|
| 361 |
+
"grad_norm": 9.556538581848145,
|
| 362 |
+
"learning_rate": 2.9320987654320992e-05,
|
| 363 |
+
"loss": 0.1498,
|
| 364 |
+
"step": 340
|
| 365 |
+
},
|
| 366 |
+
{
|
| 367 |
+
"epoch": 14.0,
|
| 368 |
+
"eval_accuracy": 0.36363636363636365,
|
| 369 |
+
"eval_loss": 2.070053815841675,
|
| 370 |
+
"eval_runtime": 4.0317,
|
| 371 |
+
"eval_samples_per_second": 5.457,
|
| 372 |
+
"eval_steps_per_second": 2.728,
|
| 373 |
+
"step": 343
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 14.285714285714286,
|
| 377 |
+
"grad_norm": 5.47396993637085,
|
| 378 |
+
"learning_rate": 2.8549382716049384e-05,
|
| 379 |
+
"loss": 0.1807,
|
| 380 |
+
"step": 350
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 14.693877551020408,
|
| 384 |
+
"grad_norm": 21.753263473510742,
|
| 385 |
+
"learning_rate": 2.777777777777778e-05,
|
| 386 |
+
"loss": 0.1599,
|
| 387 |
+
"step": 360
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 14.979591836734693,
|
| 391 |
+
"eval_accuracy": 0.5,
|
| 392 |
+
"eval_loss": 2.4973413944244385,
|
| 393 |
+
"eval_runtime": 4.128,
|
| 394 |
+
"eval_samples_per_second": 5.329,
|
| 395 |
+
"eval_steps_per_second": 2.665,
|
| 396 |
+
"step": 367
|
| 397 |
+
},
|
| 398 |
+
{
|
| 399 |
+
"epoch": 15.10204081632653,
|
| 400 |
+
"grad_norm": 38.01487350463867,
|
| 401 |
+
"learning_rate": 2.700617283950617e-05,
|
| 402 |
+
"loss": 0.2408,
|
| 403 |
+
"step": 370
|
| 404 |
+
},
|
| 405 |
+
{
|
| 406 |
+
"epoch": 15.510204081632653,
|
| 407 |
+
"grad_norm": 13.578157424926758,
|
| 408 |
+
"learning_rate": 2.623456790123457e-05,
|
| 409 |
+
"loss": 0.2606,
|
| 410 |
+
"step": 380
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 15.918367346938776,
|
| 414 |
+
"grad_norm": 7.735710620880127,
|
| 415 |
+
"learning_rate": 2.5462962962962965e-05,
|
| 416 |
+
"loss": 0.3856,
|
| 417 |
+
"step": 390
|
| 418 |
+
},
|
| 419 |
+
{
|
| 420 |
+
"epoch": 16.0,
|
| 421 |
+
"eval_accuracy": 0.45454545454545453,
|
| 422 |
+
"eval_loss": 2.247347831726074,
|
| 423 |
+
"eval_runtime": 4.1251,
|
| 424 |
+
"eval_samples_per_second": 5.333,
|
| 425 |
+
"eval_steps_per_second": 2.667,
|
| 426 |
+
"step": 392
|
| 427 |
+
},
|
| 428 |
+
{
|
| 429 |
+
"epoch": 16.3265306122449,
|
| 430 |
+
"grad_norm": 6.533689975738525,
|
| 431 |
+
"learning_rate": 2.4691358024691357e-05,
|
| 432 |
+
"loss": 0.1284,
|
| 433 |
+
"step": 400
|
| 434 |
+
},
|
| 435 |
+
{
|
| 436 |
+
"epoch": 16.73469387755102,
|
| 437 |
+
"grad_norm": 22.71470069885254,
|
| 438 |
+
"learning_rate": 2.3919753086419755e-05,
|
| 439 |
+
"loss": 0.2529,
|
| 440 |
+
"step": 410
|
| 441 |
+
},
|
| 442 |
+
{
|
| 443 |
+
"epoch": 16.979591836734695,
|
| 444 |
+
"eval_accuracy": 0.45454545454545453,
|
| 445 |
+
"eval_loss": 2.0918023586273193,
|
| 446 |
+
"eval_runtime": 3.917,
|
| 447 |
+
"eval_samples_per_second": 5.617,
|
| 448 |
+
"eval_steps_per_second": 2.808,
|
| 449 |
+
"step": 416
|
| 450 |
+
},
|
| 451 |
+
{
|
| 452 |
+
"epoch": 17.142857142857142,
|
| 453 |
+
"grad_norm": 16.790376663208008,
|
| 454 |
+
"learning_rate": 2.314814814814815e-05,
|
| 455 |
+
"loss": 0.3629,
|
| 456 |
+
"step": 420
|
| 457 |
+
},
|
| 458 |
+
{
|
| 459 |
+
"epoch": 17.551020408163264,
|
| 460 |
+
"grad_norm": 14.542320251464844,
|
| 461 |
+
"learning_rate": 2.2376543209876542e-05,
|
| 462 |
+
"loss": 0.15,
|
| 463 |
+
"step": 430
|
| 464 |
+
},
|
| 465 |
+
{
|
| 466 |
+
"epoch": 17.959183673469386,
|
| 467 |
+
"grad_norm": 2.7366743087768555,
|
| 468 |
+
"learning_rate": 2.1604938271604937e-05,
|
| 469 |
+
"loss": 0.0557,
|
| 470 |
+
"step": 440
|
| 471 |
+
},
|
| 472 |
+
{
|
| 473 |
+
"epoch": 18.0,
|
| 474 |
+
"eval_accuracy": 0.5454545454545454,
|
| 475 |
+
"eval_loss": 1.9595649242401123,
|
| 476 |
+
"eval_runtime": 4.2837,
|
| 477 |
+
"eval_samples_per_second": 5.136,
|
| 478 |
+
"eval_steps_per_second": 2.568,
|
| 479 |
+
"step": 441
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 18.367346938775512,
|
| 483 |
+
"grad_norm": 51.391292572021484,
|
| 484 |
+
"learning_rate": 2.0833333333333336e-05,
|
| 485 |
+
"loss": 0.2554,
|
| 486 |
+
"step": 450
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 18.775510204081634,
|
| 490 |
+
"grad_norm": 18.502050399780273,
|
| 491 |
+
"learning_rate": 2.006172839506173e-05,
|
| 492 |
+
"loss": 0.0895,
|
| 493 |
+
"step": 460
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 18.979591836734695,
|
| 497 |
+
"eval_accuracy": 0.45454545454545453,
|
| 498 |
+
"eval_loss": 2.5522444248199463,
|
| 499 |
+
"eval_runtime": 4.1985,
|
| 500 |
+
"eval_samples_per_second": 5.24,
|
| 501 |
+
"eval_steps_per_second": 2.62,
|
| 502 |
+
"step": 465
|
| 503 |
+
},
|
| 504 |
+
{
|
| 505 |
+
"epoch": 19.183673469387756,
|
| 506 |
+
"grad_norm": 0.6463891267776489,
|
| 507 |
+
"learning_rate": 1.9290123456790123e-05,
|
| 508 |
+
"loss": 0.1512,
|
| 509 |
+
"step": 470
|
| 510 |
+
},
|
| 511 |
+
{
|
| 512 |
+
"epoch": 19.591836734693878,
|
| 513 |
+
"grad_norm": 2.839521646499634,
|
| 514 |
+
"learning_rate": 1.8518518518518518e-05,
|
| 515 |
+
"loss": 0.1938,
|
| 516 |
+
"step": 480
|
| 517 |
+
},
|
| 518 |
+
{
|
| 519 |
+
"epoch": 20.0,
|
| 520 |
+
"grad_norm": 0.6800824403762817,
|
| 521 |
+
"learning_rate": 1.7746913580246917e-05,
|
| 522 |
+
"loss": 0.0719,
|
| 523 |
+
"step": 490
|
| 524 |
+
},
|
| 525 |
+
{
|
| 526 |
+
"epoch": 20.0,
|
| 527 |
+
"eval_accuracy": 0.5,
|
| 528 |
+
"eval_loss": 2.2938432693481445,
|
| 529 |
+
"eval_runtime": 4.0864,
|
| 530 |
+
"eval_samples_per_second": 5.384,
|
| 531 |
+
"eval_steps_per_second": 2.692,
|
| 532 |
+
"step": 490
|
| 533 |
+
},
|
| 534 |
+
{
|
| 535 |
+
"epoch": 20.408163265306122,
|
| 536 |
+
"grad_norm": 14.73177719116211,
|
| 537 |
+
"learning_rate": 1.697530864197531e-05,
|
| 538 |
+
"loss": 0.2486,
|
| 539 |
+
"step": 500
|
| 540 |
+
},
|
| 541 |
+
{
|
| 542 |
+
"epoch": 20.816326530612244,
|
| 543 |
+
"grad_norm": 1.2800395488739014,
|
| 544 |
+
"learning_rate": 1.6203703703703704e-05,
|
| 545 |
+
"loss": 0.0764,
|
| 546 |
+
"step": 510
|
| 547 |
+
},
|
| 548 |
+
{
|
| 549 |
+
"epoch": 20.979591836734695,
|
| 550 |
+
"eval_accuracy": 0.45454545454545453,
|
| 551 |
+
"eval_loss": 2.6754062175750732,
|
| 552 |
+
"eval_runtime": 3.9516,
|
| 553 |
+
"eval_samples_per_second": 5.567,
|
| 554 |
+
"eval_steps_per_second": 2.784,
|
| 555 |
+
"step": 514
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 21.224489795918366,
|
| 559 |
+
"grad_norm": 33.9787483215332,
|
| 560 |
+
"learning_rate": 1.54320987654321e-05,
|
| 561 |
+
"loss": 0.2982,
|
| 562 |
+
"step": 520
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 21.632653061224488,
|
| 566 |
+
"grad_norm": 0.02819265052676201,
|
| 567 |
+
"learning_rate": 1.4660493827160496e-05,
|
| 568 |
+
"loss": 0.1301,
|
| 569 |
+
"step": 530
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 22.0,
|
| 573 |
+
"eval_accuracy": 0.45454545454545453,
|
| 574 |
+
"eval_loss": 2.5286881923675537,
|
| 575 |
+
"eval_runtime": 4.1246,
|
| 576 |
+
"eval_samples_per_second": 5.334,
|
| 577 |
+
"eval_steps_per_second": 2.667,
|
| 578 |
+
"step": 539
|
| 579 |
+
},
|
| 580 |
+
{
|
| 581 |
+
"epoch": 22.040816326530614,
|
| 582 |
+
"grad_norm": 21.637840270996094,
|
| 583 |
+
"learning_rate": 1.388888888888889e-05,
|
| 584 |
+
"loss": 0.0917,
|
| 585 |
+
"step": 540
|
| 586 |
+
},
|
| 587 |
+
{
|
| 588 |
+
"epoch": 22.448979591836736,
|
| 589 |
+
"grad_norm": 0.41712021827697754,
|
| 590 |
+
"learning_rate": 1.3117283950617285e-05,
|
| 591 |
+
"loss": 0.1209,
|
| 592 |
+
"step": 550
|
| 593 |
+
},
|
| 594 |
+
{
|
| 595 |
+
"epoch": 22.857142857142858,
|
| 596 |
+
"grad_norm": 26.01407241821289,
|
| 597 |
+
"learning_rate": 1.2345679012345678e-05,
|
| 598 |
+
"loss": 0.1205,
|
| 599 |
+
"step": 560
|
| 600 |
+
},
|
| 601 |
+
{
|
| 602 |
+
"epoch": 22.979591836734695,
|
| 603 |
+
"eval_accuracy": 0.4090909090909091,
|
| 604 |
+
"eval_loss": 2.753232717514038,
|
| 605 |
+
"eval_runtime": 3.9124,
|
| 606 |
+
"eval_samples_per_second": 5.623,
|
| 607 |
+
"eval_steps_per_second": 2.812,
|
| 608 |
+
"step": 563
|
| 609 |
+
},
|
| 610 |
+
{
|
| 611 |
+
"epoch": 23.26530612244898,
|
| 612 |
+
"grad_norm": 11.909204483032227,
|
| 613 |
+
"learning_rate": 1.1574074074074075e-05,
|
| 614 |
+
"loss": 0.1367,
|
| 615 |
+
"step": 570
|
| 616 |
+
},
|
| 617 |
+
{
|
| 618 |
+
"epoch": 23.6734693877551,
|
| 619 |
+
"grad_norm": 0.861128032207489,
|
| 620 |
+
"learning_rate": 1.0802469135802469e-05,
|
| 621 |
+
"loss": 0.1013,
|
| 622 |
+
"step": 580
|
| 623 |
+
},
|
| 624 |
+
{
|
| 625 |
+
"epoch": 24.0,
|
| 626 |
+
"eval_accuracy": 0.45454545454545453,
|
| 627 |
+
"eval_loss": 2.6988065242767334,
|
| 628 |
+
"eval_runtime": 4.1089,
|
| 629 |
+
"eval_samples_per_second": 5.354,
|
| 630 |
+
"eval_steps_per_second": 2.677,
|
| 631 |
+
"step": 588
|
| 632 |
+
},
|
| 633 |
+
{
|
| 634 |
+
"epoch": 24.081632653061224,
|
| 635 |
+
"grad_norm": 0.21600103378295898,
|
| 636 |
+
"learning_rate": 1.0030864197530866e-05,
|
| 637 |
+
"loss": 0.1176,
|
| 638 |
+
"step": 590
|
| 639 |
+
},
|
| 640 |
+
{
|
| 641 |
+
"epoch": 24.489795918367346,
|
| 642 |
+
"grad_norm": 39.00993347167969,
|
| 643 |
+
"learning_rate": 9.259259259259259e-06,
|
| 644 |
+
"loss": 0.1305,
|
| 645 |
+
"step": 600
|
| 646 |
+
},
|
| 647 |
+
{
|
| 648 |
+
"epoch": 24.897959183673468,
|
| 649 |
+
"grad_norm": 23.242216110229492,
|
| 650 |
+
"learning_rate": 8.487654320987654e-06,
|
| 651 |
+
"loss": 0.0777,
|
| 652 |
+
"step": 610
|
| 653 |
+
},
|
| 654 |
+
{
|
| 655 |
+
"epoch": 24.979591836734695,
|
| 656 |
+
"eval_accuracy": 0.4090909090909091,
|
| 657 |
+
"eval_loss": 2.934515953063965,
|
| 658 |
+
"eval_runtime": 3.9343,
|
| 659 |
+
"eval_samples_per_second": 5.592,
|
| 660 |
+
"eval_steps_per_second": 2.796,
|
| 661 |
+
"step": 612
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 25.306122448979593,
|
| 665 |
+
"grad_norm": 0.09061746299266815,
|
| 666 |
+
"learning_rate": 7.71604938271605e-06,
|
| 667 |
+
"loss": 0.0301,
|
| 668 |
+
"step": 620
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 25.714285714285715,
|
| 672 |
+
"grad_norm": 0.2142326533794403,
|
| 673 |
+
"learning_rate": 6.944444444444445e-06,
|
| 674 |
+
"loss": 0.1807,
|
| 675 |
+
"step": 630
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 26.0,
|
| 679 |
+
"eval_accuracy": 0.45454545454545453,
|
| 680 |
+
"eval_loss": 2.998149871826172,
|
| 681 |
+
"eval_runtime": 4.0792,
|
| 682 |
+
"eval_samples_per_second": 5.393,
|
| 683 |
+
"eval_steps_per_second": 2.697,
|
| 684 |
+
"step": 637
|
| 685 |
+
},
|
| 686 |
+
{
|
| 687 |
+
"epoch": 26.122448979591837,
|
| 688 |
+
"grad_norm": 0.43653419613838196,
|
| 689 |
+
"learning_rate": 6.172839506172839e-06,
|
| 690 |
+
"loss": 0.0358,
|
| 691 |
+
"step": 640
|
| 692 |
+
},
|
| 693 |
+
{
|
| 694 |
+
"epoch": 26.53061224489796,
|
| 695 |
+
"grad_norm": 0.5854562520980835,
|
| 696 |
+
"learning_rate": 5.401234567901234e-06,
|
| 697 |
+
"loss": 0.0422,
|
| 698 |
+
"step": 650
|
| 699 |
+
},
|
| 700 |
+
{
|
| 701 |
+
"epoch": 26.93877551020408,
|
| 702 |
+
"grad_norm": 0.39157870411872864,
|
| 703 |
+
"learning_rate": 4.6296296296296296e-06,
|
| 704 |
+
"loss": 0.0298,
|
| 705 |
+
"step": 660
|
| 706 |
+
},
|
| 707 |
+
{
|
| 708 |
+
"epoch": 26.979591836734695,
|
| 709 |
+
"eval_accuracy": 0.45454545454545453,
|
| 710 |
+
"eval_loss": 2.854871988296509,
|
| 711 |
+
"eval_runtime": 3.9772,
|
| 712 |
+
"eval_samples_per_second": 5.532,
|
| 713 |
+
"eval_steps_per_second": 2.766,
|
| 714 |
+
"step": 661
|
| 715 |
+
},
|
| 716 |
+
{
|
| 717 |
+
"epoch": 27.346938775510203,
|
| 718 |
+
"grad_norm": 0.041275933384895325,
|
| 719 |
+
"learning_rate": 3.858024691358025e-06,
|
| 720 |
+
"loss": 0.0633,
|
| 721 |
+
"step": 670
|
| 722 |
+
},
|
| 723 |
+
{
|
| 724 |
+
"epoch": 27.755102040816325,
|
| 725 |
+
"grad_norm": 18.42709732055664,
|
| 726 |
+
"learning_rate": 3.0864197530864196e-06,
|
| 727 |
+
"loss": 0.0589,
|
| 728 |
+
"step": 680
|
| 729 |
+
},
|
| 730 |
+
{
|
| 731 |
+
"epoch": 28.0,
|
| 732 |
+
"eval_accuracy": 0.45454545454545453,
|
| 733 |
+
"eval_loss": 2.6966726779937744,
|
| 734 |
+
"eval_runtime": 4.1347,
|
| 735 |
+
"eval_samples_per_second": 5.321,
|
| 736 |
+
"eval_steps_per_second": 2.66,
|
| 737 |
+
"step": 686
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 28.163265306122447,
|
| 741 |
+
"grad_norm": 27.545507431030273,
|
| 742 |
+
"learning_rate": 2.3148148148148148e-06,
|
| 743 |
+
"loss": 0.1035,
|
| 744 |
+
"step": 690
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 28.571428571428573,
|
| 748 |
+
"grad_norm": 14.813511848449707,
|
| 749 |
+
"learning_rate": 1.5432098765432098e-06,
|
| 750 |
+
"loss": 0.1094,
|
| 751 |
+
"step": 700
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 28.979591836734695,
|
| 755 |
+
"grad_norm": 0.47487640380859375,
|
| 756 |
+
"learning_rate": 7.716049382716049e-07,
|
| 757 |
+
"loss": 0.0896,
|
| 758 |
+
"step": 710
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 28.979591836734695,
|
| 762 |
+
"eval_accuracy": 0.45454545454545453,
|
| 763 |
+
"eval_loss": 2.6902878284454346,
|
| 764 |
+
"eval_runtime": 3.9324,
|
| 765 |
+
"eval_samples_per_second": 5.595,
|
| 766 |
+
"eval_steps_per_second": 2.797,
|
| 767 |
+
"step": 710
|
| 768 |
+
},
|
| 769 |
+
{
|
| 770 |
+
"epoch": 29.387755102040817,
|
| 771 |
+
"grad_norm": 0.2644538879394531,
|
| 772 |
+
"learning_rate": 0.0,
|
| 773 |
+
"loss": 0.0218,
|
| 774 |
+
"step": 720
|
| 775 |
+
},
|
| 776 |
+
{
|
| 777 |
+
"epoch": 29.387755102040817,
|
| 778 |
+
"eval_accuracy": 0.45454545454545453,
|
| 779 |
+
"eval_loss": 2.690171480178833,
|
| 780 |
+
"eval_runtime": 4.1046,
|
| 781 |
+
"eval_samples_per_second": 5.36,
|
| 782 |
+
"eval_steps_per_second": 2.68,
|
| 783 |
+
"step": 720
|
| 784 |
+
},
|
| 785 |
+
{
|
| 786 |
+
"epoch": 29.387755102040817,
|
| 787 |
+
"step": 720,
|
| 788 |
+
"total_flos": 2.5282019639994163e+17,
|
| 789 |
+
"train_loss": 0.5089964731906851,
|
| 790 |
+
"train_runtime": 1181.0861,
|
| 791 |
+
"train_samples_per_second": 4.953,
|
| 792 |
+
"train_steps_per_second": 0.61
|
| 793 |
+
}
|
| 794 |
+
],
|
| 795 |
+
"logging_steps": 10,
|
| 796 |
+
"max_steps": 720,
|
| 797 |
+
"num_input_tokens_seen": 0,
|
| 798 |
+
"num_train_epochs": 30,
|
| 799 |
+
"save_steps": 500,
|
| 800 |
+
"stateful_callbacks": {
|
| 801 |
+
"TrainerControl": {
|
| 802 |
+
"args": {
|
| 803 |
+
"should_epoch_stop": false,
|
| 804 |
+
"should_evaluate": false,
|
| 805 |
+
"should_log": false,
|
| 806 |
+
"should_save": true,
|
| 807 |
+
"should_training_stop": true
|
| 808 |
+
},
|
| 809 |
+
"attributes": {}
|
| 810 |
+
}
|
| 811 |
+
},
|
| 812 |
+
"total_flos": 2.5282019639994163e+17,
|
| 813 |
+
"train_batch_size": 2,
|
| 814 |
+
"trial_name": null,
|
| 815 |
+
"trial_params": null
|
| 816 |
+
}
|