End of training
Browse files- README.md +3 -3
- all_results.json +13 -0
- eval_results.json +8 -0
- train_results.json +8 -0
- trainer_state.json +1110 -0
README.md
CHANGED
|
@@ -23,7 +23,7 @@ model-index:
|
|
| 23 |
metrics:
|
| 24 |
- name: Accuracy
|
| 25 |
type: accuracy
|
| 26 |
-
value: 0.
|
| 27 |
---
|
| 28 |
|
| 29 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
@@ -33,8 +33,8 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 33 |
|
| 34 |
This model is a fine-tuned version of [google/vit-base-patch16-224-in21k](https://huggingface.co/google/vit-base-patch16-224-in21k) on the imagefolder dataset.
|
| 35 |
It achieves the following results on the evaluation set:
|
| 36 |
-
- Loss: 0.
|
| 37 |
-
- Accuracy: 0.
|
| 38 |
|
| 39 |
## Model description
|
| 40 |
|
|
|
|
| 23 |
metrics:
|
| 24 |
- name: Accuracy
|
| 25 |
type: accuracy
|
| 26 |
+
value: 0.9411764705882353
|
| 27 |
---
|
| 28 |
|
| 29 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
|
|
| 33 |
|
| 34 |
This model is a fine-tuned version of [google/vit-base-patch16-224-in21k](https://huggingface.co/google/vit-base-patch16-224-in21k) on the imagefolder dataset.
|
| 35 |
It achieves the following results on the evaluation set:
|
| 36 |
+
- Loss: 0.2523
|
| 37 |
+
- Accuracy: 0.9412
|
| 38 |
|
| 39 |
## Model description
|
| 40 |
|
all_results.json
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 29.80392156862745,
|
| 3 |
+
"eval_accuracy": 0.9411764705882353,
|
| 4 |
+
"eval_loss": 0.2523283064365387,
|
| 5 |
+
"eval_runtime": 49.5509,
|
| 6 |
+
"eval_samples_per_second": 2.745,
|
| 7 |
+
"eval_steps_per_second": 0.343,
|
| 8 |
+
"total_flos": 2.827009127861453e+18,
|
| 9 |
+
"train_loss": 0.3311853929046999,
|
| 10 |
+
"train_runtime": 10131.822,
|
| 11 |
+
"train_samples_per_second": 3.624,
|
| 12 |
+
"train_steps_per_second": 0.113
|
| 13 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 29.80392156862745,
|
| 3 |
+
"eval_accuracy": 0.9411764705882353,
|
| 4 |
+
"eval_loss": 0.2523283064365387,
|
| 5 |
+
"eval_runtime": 49.5509,
|
| 6 |
+
"eval_samples_per_second": 2.745,
|
| 7 |
+
"eval_steps_per_second": 0.343
|
| 8 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 29.80392156862745,
|
| 3 |
+
"total_flos": 2.827009127861453e+18,
|
| 4 |
+
"train_loss": 0.3311853929046999,
|
| 5 |
+
"train_runtime": 10131.822,
|
| 6 |
+
"train_samples_per_second": 3.624,
|
| 7 |
+
"train_steps_per_second": 0.113
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1110 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.9411764705882353,
|
| 3 |
+
"best_model_checkpoint": "vit-base-patch16-224-in21k-finetuned-papsmear/checkpoint-765",
|
| 4 |
+
"epoch": 29.80392156862745,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 1140,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.26143790849673204,
|
| 13 |
+
"grad_norm": 1.524962306022644,
|
| 14 |
+
"learning_rate": 4.3859649122807014e-06,
|
| 15 |
+
"loss": 1.8303,
|
| 16 |
+
"step": 10
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.5228758169934641,
|
| 20 |
+
"grad_norm": 1.3496487140655518,
|
| 21 |
+
"learning_rate": 8.771929824561403e-06,
|
| 22 |
+
"loss": 1.7834,
|
| 23 |
+
"step": 20
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.7843137254901961,
|
| 27 |
+
"grad_norm": 1.857043981552124,
|
| 28 |
+
"learning_rate": 1.3157894736842106e-05,
|
| 29 |
+
"loss": 1.6954,
|
| 30 |
+
"step": 30
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.9934640522875817,
|
| 34 |
+
"eval_accuracy": 0.34558823529411764,
|
| 35 |
+
"eval_loss": 1.610603928565979,
|
| 36 |
+
"eval_runtime": 33.3958,
|
| 37 |
+
"eval_samples_per_second": 4.072,
|
| 38 |
+
"eval_steps_per_second": 0.509,
|
| 39 |
+
"step": 38
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 1.0457516339869282,
|
| 43 |
+
"grad_norm": 1.5567628145217896,
|
| 44 |
+
"learning_rate": 1.7543859649122806e-05,
|
| 45 |
+
"loss": 1.5872,
|
| 46 |
+
"step": 40
|
| 47 |
+
},
|
| 48 |
+
{
|
| 49 |
+
"epoch": 1.3071895424836601,
|
| 50 |
+
"grad_norm": 1.5087599754333496,
|
| 51 |
+
"learning_rate": 2.1929824561403507e-05,
|
| 52 |
+
"loss": 1.4855,
|
| 53 |
+
"step": 50
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 1.5686274509803921,
|
| 57 |
+
"grad_norm": 1.6773402690887451,
|
| 58 |
+
"learning_rate": 2.6315789473684212e-05,
|
| 59 |
+
"loss": 1.3823,
|
| 60 |
+
"step": 60
|
| 61 |
+
},
|
| 62 |
+
{
|
| 63 |
+
"epoch": 1.8300653594771243,
|
| 64 |
+
"grad_norm": 1.4024033546447754,
|
| 65 |
+
"learning_rate": 3.0701754385964913e-05,
|
| 66 |
+
"loss": 1.2818,
|
| 67 |
+
"step": 70
|
| 68 |
+
},
|
| 69 |
+
{
|
| 70 |
+
"epoch": 1.9869281045751634,
|
| 71 |
+
"eval_accuracy": 0.5735294117647058,
|
| 72 |
+
"eval_loss": 1.2412389516830444,
|
| 73 |
+
"eval_runtime": 32.7103,
|
| 74 |
+
"eval_samples_per_second": 4.158,
|
| 75 |
+
"eval_steps_per_second": 0.52,
|
| 76 |
+
"step": 76
|
| 77 |
+
},
|
| 78 |
+
{
|
| 79 |
+
"epoch": 2.0915032679738563,
|
| 80 |
+
"grad_norm": 1.6061407327651978,
|
| 81 |
+
"learning_rate": 3.508771929824561e-05,
|
| 82 |
+
"loss": 1.2088,
|
| 83 |
+
"step": 80
|
| 84 |
+
},
|
| 85 |
+
{
|
| 86 |
+
"epoch": 2.3529411764705883,
|
| 87 |
+
"grad_norm": 3.8964905738830566,
|
| 88 |
+
"learning_rate": 3.9473684210526316e-05,
|
| 89 |
+
"loss": 1.0983,
|
| 90 |
+
"step": 90
|
| 91 |
+
},
|
| 92 |
+
{
|
| 93 |
+
"epoch": 2.6143790849673203,
|
| 94 |
+
"grad_norm": 3.24979305267334,
|
| 95 |
+
"learning_rate": 4.3859649122807014e-05,
|
| 96 |
+
"loss": 1.0632,
|
| 97 |
+
"step": 100
|
| 98 |
+
},
|
| 99 |
+
{
|
| 100 |
+
"epoch": 2.8758169934640523,
|
| 101 |
+
"grad_norm": 2.1930079460144043,
|
| 102 |
+
"learning_rate": 4.824561403508772e-05,
|
| 103 |
+
"loss": 1.0023,
|
| 104 |
+
"step": 110
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 2.980392156862745,
|
| 108 |
+
"eval_accuracy": 0.7132352941176471,
|
| 109 |
+
"eval_loss": 0.9875355958938599,
|
| 110 |
+
"eval_runtime": 32.8626,
|
| 111 |
+
"eval_samples_per_second": 4.138,
|
| 112 |
+
"eval_steps_per_second": 0.517,
|
| 113 |
+
"step": 114
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"epoch": 3.1372549019607843,
|
| 117 |
+
"grad_norm": 2.084902048110962,
|
| 118 |
+
"learning_rate": 4.970760233918128e-05,
|
| 119 |
+
"loss": 0.823,
|
| 120 |
+
"step": 120
|
| 121 |
+
},
|
| 122 |
+
{
|
| 123 |
+
"epoch": 3.3986928104575163,
|
| 124 |
+
"grad_norm": 1.8761796951293945,
|
| 125 |
+
"learning_rate": 4.9220272904483435e-05,
|
| 126 |
+
"loss": 0.785,
|
| 127 |
+
"step": 130
|
| 128 |
+
},
|
| 129 |
+
{
|
| 130 |
+
"epoch": 3.6601307189542482,
|
| 131 |
+
"grad_norm": 3.031520128250122,
|
| 132 |
+
"learning_rate": 4.8732943469785574e-05,
|
| 133 |
+
"loss": 0.7775,
|
| 134 |
+
"step": 140
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 3.9215686274509802,
|
| 138 |
+
"grad_norm": 2.1056406497955322,
|
| 139 |
+
"learning_rate": 4.824561403508772e-05,
|
| 140 |
+
"loss": 0.7163,
|
| 141 |
+
"step": 150
|
| 142 |
+
},
|
| 143 |
+
{
|
| 144 |
+
"epoch": 4.0,
|
| 145 |
+
"eval_accuracy": 0.6911764705882353,
|
| 146 |
+
"eval_loss": 0.8399065136909485,
|
| 147 |
+
"eval_runtime": 32.9831,
|
| 148 |
+
"eval_samples_per_second": 4.123,
|
| 149 |
+
"eval_steps_per_second": 0.515,
|
| 150 |
+
"step": 153
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 4.183006535947713,
|
| 154 |
+
"grad_norm": 3.3159000873565674,
|
| 155 |
+
"learning_rate": 4.7758284600389865e-05,
|
| 156 |
+
"loss": 0.6677,
|
| 157 |
+
"step": 160
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 4.444444444444445,
|
| 161 |
+
"grad_norm": 2.1595349311828613,
|
| 162 |
+
"learning_rate": 4.727095516569201e-05,
|
| 163 |
+
"loss": 0.6366,
|
| 164 |
+
"step": 170
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 4.705882352941177,
|
| 168 |
+
"grad_norm": 2.2182297706604004,
|
| 169 |
+
"learning_rate": 4.678362573099415e-05,
|
| 170 |
+
"loss": 0.5876,
|
| 171 |
+
"step": 180
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 4.967320261437909,
|
| 175 |
+
"grad_norm": 5.272632122039795,
|
| 176 |
+
"learning_rate": 4.62962962962963e-05,
|
| 177 |
+
"loss": 0.5173,
|
| 178 |
+
"step": 190
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 4.993464052287582,
|
| 182 |
+
"eval_accuracy": 0.8161764705882353,
|
| 183 |
+
"eval_loss": 0.6545882821083069,
|
| 184 |
+
"eval_runtime": 31.9706,
|
| 185 |
+
"eval_samples_per_second": 4.254,
|
| 186 |
+
"eval_steps_per_second": 0.532,
|
| 187 |
+
"step": 191
|
| 188 |
+
},
|
| 189 |
+
{
|
| 190 |
+
"epoch": 5.228758169934641,
|
| 191 |
+
"grad_norm": 5.355340003967285,
|
| 192 |
+
"learning_rate": 4.580896686159844e-05,
|
| 193 |
+
"loss": 0.5057,
|
| 194 |
+
"step": 200
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 5.490196078431373,
|
| 198 |
+
"grad_norm": 2.3517274856567383,
|
| 199 |
+
"learning_rate": 4.5321637426900585e-05,
|
| 200 |
+
"loss": 0.4577,
|
| 201 |
+
"step": 210
|
| 202 |
+
},
|
| 203 |
+
{
|
| 204 |
+
"epoch": 5.751633986928105,
|
| 205 |
+
"grad_norm": 7.760730266571045,
|
| 206 |
+
"learning_rate": 4.483430799220273e-05,
|
| 207 |
+
"loss": 0.5057,
|
| 208 |
+
"step": 220
|
| 209 |
+
},
|
| 210 |
+
{
|
| 211 |
+
"epoch": 5.9869281045751634,
|
| 212 |
+
"eval_accuracy": 0.8308823529411765,
|
| 213 |
+
"eval_loss": 0.6250559091567993,
|
| 214 |
+
"eval_runtime": 32.3916,
|
| 215 |
+
"eval_samples_per_second": 4.199,
|
| 216 |
+
"eval_steps_per_second": 0.525,
|
| 217 |
+
"step": 229
|
| 218 |
+
},
|
| 219 |
+
{
|
| 220 |
+
"epoch": 6.0130718954248366,
|
| 221 |
+
"grad_norm": 3.0986075401306152,
|
| 222 |
+
"learning_rate": 4.4346978557504876e-05,
|
| 223 |
+
"loss": 0.4856,
|
| 224 |
+
"step": 230
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 6.2745098039215685,
|
| 228 |
+
"grad_norm": 7.31836462020874,
|
| 229 |
+
"learning_rate": 4.3859649122807014e-05,
|
| 230 |
+
"loss": 0.5036,
|
| 231 |
+
"step": 240
|
| 232 |
+
},
|
| 233 |
+
{
|
| 234 |
+
"epoch": 6.5359477124183005,
|
| 235 |
+
"grad_norm": 2.5667953491210938,
|
| 236 |
+
"learning_rate": 4.3372319688109166e-05,
|
| 237 |
+
"loss": 0.4063,
|
| 238 |
+
"step": 250
|
| 239 |
+
},
|
| 240 |
+
{
|
| 241 |
+
"epoch": 6.7973856209150325,
|
| 242 |
+
"grad_norm": 2.488330602645874,
|
| 243 |
+
"learning_rate": 4.2884990253411305e-05,
|
| 244 |
+
"loss": 0.4313,
|
| 245 |
+
"step": 260
|
| 246 |
+
},
|
| 247 |
+
{
|
| 248 |
+
"epoch": 6.980392156862745,
|
| 249 |
+
"eval_accuracy": 0.8308823529411765,
|
| 250 |
+
"eval_loss": 0.5696418881416321,
|
| 251 |
+
"eval_runtime": 32.7903,
|
| 252 |
+
"eval_samples_per_second": 4.148,
|
| 253 |
+
"eval_steps_per_second": 0.518,
|
| 254 |
+
"step": 267
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 7.0588235294117645,
|
| 258 |
+
"grad_norm": 4.5028228759765625,
|
| 259 |
+
"learning_rate": 4.239766081871345e-05,
|
| 260 |
+
"loss": 0.3409,
|
| 261 |
+
"step": 270
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 7.3202614379084965,
|
| 265 |
+
"grad_norm": 3.8160030841827393,
|
| 266 |
+
"learning_rate": 4.1910331384015596e-05,
|
| 267 |
+
"loss": 0.4005,
|
| 268 |
+
"step": 280
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 7.5816993464052285,
|
| 272 |
+
"grad_norm": 7.039416790008545,
|
| 273 |
+
"learning_rate": 4.142300194931774e-05,
|
| 274 |
+
"loss": 0.4047,
|
| 275 |
+
"step": 290
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 7.8431372549019605,
|
| 279 |
+
"grad_norm": 2.3010506629943848,
|
| 280 |
+
"learning_rate": 4.093567251461988e-05,
|
| 281 |
+
"loss": 0.325,
|
| 282 |
+
"step": 300
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 8.0,
|
| 286 |
+
"eval_accuracy": 0.8308823529411765,
|
| 287 |
+
"eval_loss": 0.5506752133369446,
|
| 288 |
+
"eval_runtime": 33.2532,
|
| 289 |
+
"eval_samples_per_second": 4.09,
|
| 290 |
+
"eval_steps_per_second": 0.511,
|
| 291 |
+
"step": 306
|
| 292 |
+
},
|
| 293 |
+
{
|
| 294 |
+
"epoch": 8.104575163398692,
|
| 295 |
+
"grad_norm": 2.194329023361206,
|
| 296 |
+
"learning_rate": 4.044834307992203e-05,
|
| 297 |
+
"loss": 0.3016,
|
| 298 |
+
"step": 310
|
| 299 |
+
},
|
| 300 |
+
{
|
| 301 |
+
"epoch": 8.366013071895425,
|
| 302 |
+
"grad_norm": 4.172213554382324,
|
| 303 |
+
"learning_rate": 3.996101364522417e-05,
|
| 304 |
+
"loss": 0.3194,
|
| 305 |
+
"step": 320
|
| 306 |
+
},
|
| 307 |
+
{
|
| 308 |
+
"epoch": 8.627450980392156,
|
| 309 |
+
"grad_norm": 3.775691270828247,
|
| 310 |
+
"learning_rate": 3.9473684210526316e-05,
|
| 311 |
+
"loss": 0.3115,
|
| 312 |
+
"step": 330
|
| 313 |
+
},
|
| 314 |
+
{
|
| 315 |
+
"epoch": 8.88888888888889,
|
| 316 |
+
"grad_norm": 3.285763740539551,
|
| 317 |
+
"learning_rate": 3.898635477582846e-05,
|
| 318 |
+
"loss": 0.3811,
|
| 319 |
+
"step": 340
|
| 320 |
+
},
|
| 321 |
+
{
|
| 322 |
+
"epoch": 8.993464052287582,
|
| 323 |
+
"eval_accuracy": 0.8676470588235294,
|
| 324 |
+
"eval_loss": 0.4429008662700653,
|
| 325 |
+
"eval_runtime": 32.2721,
|
| 326 |
+
"eval_samples_per_second": 4.214,
|
| 327 |
+
"eval_steps_per_second": 0.527,
|
| 328 |
+
"step": 344
|
| 329 |
+
},
|
| 330 |
+
{
|
| 331 |
+
"epoch": 9.15032679738562,
|
| 332 |
+
"grad_norm": 2.8171193599700928,
|
| 333 |
+
"learning_rate": 3.849902534113061e-05,
|
| 334 |
+
"loss": 0.3617,
|
| 335 |
+
"step": 350
|
| 336 |
+
},
|
| 337 |
+
{
|
| 338 |
+
"epoch": 9.411764705882353,
|
| 339 |
+
"grad_norm": 2.766669273376465,
|
| 340 |
+
"learning_rate": 3.8011695906432746e-05,
|
| 341 |
+
"loss": 0.2679,
|
| 342 |
+
"step": 360
|
| 343 |
+
},
|
| 344 |
+
{
|
| 345 |
+
"epoch": 9.673202614379084,
|
| 346 |
+
"grad_norm": 5.8598833084106445,
|
| 347 |
+
"learning_rate": 3.75243664717349e-05,
|
| 348 |
+
"loss": 0.2909,
|
| 349 |
+
"step": 370
|
| 350 |
+
},
|
| 351 |
+
{
|
| 352 |
+
"epoch": 9.934640522875817,
|
| 353 |
+
"grad_norm": 4.5529608726501465,
|
| 354 |
+
"learning_rate": 3.7037037037037037e-05,
|
| 355 |
+
"loss": 0.2341,
|
| 356 |
+
"step": 380
|
| 357 |
+
},
|
| 358 |
+
{
|
| 359 |
+
"epoch": 9.986928104575163,
|
| 360 |
+
"eval_accuracy": 0.875,
|
| 361 |
+
"eval_loss": 0.42220985889434814,
|
| 362 |
+
"eval_runtime": 32.0969,
|
| 363 |
+
"eval_samples_per_second": 4.237,
|
| 364 |
+
"eval_steps_per_second": 0.53,
|
| 365 |
+
"step": 382
|
| 366 |
+
},
|
| 367 |
+
{
|
| 368 |
+
"epoch": 10.196078431372548,
|
| 369 |
+
"grad_norm": 3.98348069190979,
|
| 370 |
+
"learning_rate": 3.654970760233918e-05,
|
| 371 |
+
"loss": 0.2612,
|
| 372 |
+
"step": 390
|
| 373 |
+
},
|
| 374 |
+
{
|
| 375 |
+
"epoch": 10.457516339869281,
|
| 376 |
+
"grad_norm": 5.887331485748291,
|
| 377 |
+
"learning_rate": 3.606237816764133e-05,
|
| 378 |
+
"loss": 0.2518,
|
| 379 |
+
"step": 400
|
| 380 |
+
},
|
| 381 |
+
{
|
| 382 |
+
"epoch": 10.718954248366012,
|
| 383 |
+
"grad_norm": 3.759489059448242,
|
| 384 |
+
"learning_rate": 3.557504873294347e-05,
|
| 385 |
+
"loss": 0.2361,
|
| 386 |
+
"step": 410
|
| 387 |
+
},
|
| 388 |
+
{
|
| 389 |
+
"epoch": 10.980392156862745,
|
| 390 |
+
"grad_norm": 6.444194793701172,
|
| 391 |
+
"learning_rate": 3.508771929824561e-05,
|
| 392 |
+
"loss": 0.3082,
|
| 393 |
+
"step": 420
|
| 394 |
+
},
|
| 395 |
+
{
|
| 396 |
+
"epoch": 10.980392156862745,
|
| 397 |
+
"eval_accuracy": 0.7720588235294118,
|
| 398 |
+
"eval_loss": 0.6572562456130981,
|
| 399 |
+
"eval_runtime": 32.1021,
|
| 400 |
+
"eval_samples_per_second": 4.236,
|
| 401 |
+
"eval_steps_per_second": 0.53,
|
| 402 |
+
"step": 420
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 11.241830065359476,
|
| 406 |
+
"grad_norm": 2.5144402980804443,
|
| 407 |
+
"learning_rate": 3.4600389863547764e-05,
|
| 408 |
+
"loss": 0.2284,
|
| 409 |
+
"step": 430
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 11.50326797385621,
|
| 413 |
+
"grad_norm": 5.127187728881836,
|
| 414 |
+
"learning_rate": 3.41130604288499e-05,
|
| 415 |
+
"loss": 0.2353,
|
| 416 |
+
"step": 440
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 11.764705882352942,
|
| 420 |
+
"grad_norm": 3.2781078815460205,
|
| 421 |
+
"learning_rate": 3.362573099415205e-05,
|
| 422 |
+
"loss": 0.2571,
|
| 423 |
+
"step": 450
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 12.0,
|
| 427 |
+
"eval_accuracy": 0.8897058823529411,
|
| 428 |
+
"eval_loss": 0.4229148328304291,
|
| 429 |
+
"eval_runtime": 31.8733,
|
| 430 |
+
"eval_samples_per_second": 4.267,
|
| 431 |
+
"eval_steps_per_second": 0.533,
|
| 432 |
+
"step": 459
|
| 433 |
+
},
|
| 434 |
+
{
|
| 435 |
+
"epoch": 12.026143790849673,
|
| 436 |
+
"grad_norm": 1.4445286989212036,
|
| 437 |
+
"learning_rate": 3.313840155945419e-05,
|
| 438 |
+
"loss": 0.2537,
|
| 439 |
+
"step": 460
|
| 440 |
+
},
|
| 441 |
+
{
|
| 442 |
+
"epoch": 12.287581699346406,
|
| 443 |
+
"grad_norm": 3.5303471088409424,
|
| 444 |
+
"learning_rate": 3.265107212475634e-05,
|
| 445 |
+
"loss": 0.23,
|
| 446 |
+
"step": 470
|
| 447 |
+
},
|
| 448 |
+
{
|
| 449 |
+
"epoch": 12.549019607843137,
|
| 450 |
+
"grad_norm": 4.927082538604736,
|
| 451 |
+
"learning_rate": 3.216374269005848e-05,
|
| 452 |
+
"loss": 0.1693,
|
| 453 |
+
"step": 480
|
| 454 |
+
},
|
| 455 |
+
{
|
| 456 |
+
"epoch": 12.81045751633987,
|
| 457 |
+
"grad_norm": 1.5047080516815186,
|
| 458 |
+
"learning_rate": 3.167641325536063e-05,
|
| 459 |
+
"loss": 0.2374,
|
| 460 |
+
"step": 490
|
| 461 |
+
},
|
| 462 |
+
{
|
| 463 |
+
"epoch": 12.993464052287582,
|
| 464 |
+
"eval_accuracy": 0.875,
|
| 465 |
+
"eval_loss": 0.42333418130874634,
|
| 466 |
+
"eval_runtime": 32.071,
|
| 467 |
+
"eval_samples_per_second": 4.241,
|
| 468 |
+
"eval_steps_per_second": 0.53,
|
| 469 |
+
"step": 497
|
| 470 |
+
},
|
| 471 |
+
{
|
| 472 |
+
"epoch": 13.071895424836601,
|
| 473 |
+
"grad_norm": 1.6010866165161133,
|
| 474 |
+
"learning_rate": 3.118908382066277e-05,
|
| 475 |
+
"loss": 0.2091,
|
| 476 |
+
"step": 500
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 13.333333333333334,
|
| 480 |
+
"grad_norm": 1.0056742429733276,
|
| 481 |
+
"learning_rate": 3.0701754385964913e-05,
|
| 482 |
+
"loss": 0.2203,
|
| 483 |
+
"step": 510
|
| 484 |
+
},
|
| 485 |
+
{
|
| 486 |
+
"epoch": 13.594771241830065,
|
| 487 |
+
"grad_norm": 0.8992928266525269,
|
| 488 |
+
"learning_rate": 3.0214424951267055e-05,
|
| 489 |
+
"loss": 0.1784,
|
| 490 |
+
"step": 520
|
| 491 |
+
},
|
| 492 |
+
{
|
| 493 |
+
"epoch": 13.856209150326798,
|
| 494 |
+
"grad_norm": 1.4664764404296875,
|
| 495 |
+
"learning_rate": 2.9727095516569204e-05,
|
| 496 |
+
"loss": 0.128,
|
| 497 |
+
"step": 530
|
| 498 |
+
},
|
| 499 |
+
{
|
| 500 |
+
"epoch": 13.986928104575163,
|
| 501 |
+
"eval_accuracy": 0.8970588235294118,
|
| 502 |
+
"eval_loss": 0.367097944021225,
|
| 503 |
+
"eval_runtime": 31.5031,
|
| 504 |
+
"eval_samples_per_second": 4.317,
|
| 505 |
+
"eval_steps_per_second": 0.54,
|
| 506 |
+
"step": 535
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 14.117647058823529,
|
| 510 |
+
"grad_norm": 2.8378169536590576,
|
| 511 |
+
"learning_rate": 2.9239766081871346e-05,
|
| 512 |
+
"loss": 0.1436,
|
| 513 |
+
"step": 540
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 14.379084967320262,
|
| 517 |
+
"grad_norm": 1.6675957441329956,
|
| 518 |
+
"learning_rate": 2.875243664717349e-05,
|
| 519 |
+
"loss": 0.1986,
|
| 520 |
+
"step": 550
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 14.640522875816993,
|
| 524 |
+
"grad_norm": 8.45456314086914,
|
| 525 |
+
"learning_rate": 2.8265107212475634e-05,
|
| 526 |
+
"loss": 0.1711,
|
| 527 |
+
"step": 560
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 14.901960784313726,
|
| 531 |
+
"grad_norm": 6.542934417724609,
|
| 532 |
+
"learning_rate": 2.777777777777778e-05,
|
| 533 |
+
"loss": 0.1718,
|
| 534 |
+
"step": 570
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 14.980392156862745,
|
| 538 |
+
"eval_accuracy": 0.8970588235294118,
|
| 539 |
+
"eval_loss": 0.34296926856040955,
|
| 540 |
+
"eval_runtime": 31.8312,
|
| 541 |
+
"eval_samples_per_second": 4.273,
|
| 542 |
+
"eval_steps_per_second": 0.534,
|
| 543 |
+
"step": 573
|
| 544 |
+
},
|
| 545 |
+
{
|
| 546 |
+
"epoch": 15.163398692810457,
|
| 547 |
+
"grad_norm": 4.025688171386719,
|
| 548 |
+
"learning_rate": 2.729044834307992e-05,
|
| 549 |
+
"loss": 0.1681,
|
| 550 |
+
"step": 580
|
| 551 |
+
},
|
| 552 |
+
{
|
| 553 |
+
"epoch": 15.42483660130719,
|
| 554 |
+
"grad_norm": 0.5761296153068542,
|
| 555 |
+
"learning_rate": 2.680311890838207e-05,
|
| 556 |
+
"loss": 0.1558,
|
| 557 |
+
"step": 590
|
| 558 |
+
},
|
| 559 |
+
{
|
| 560 |
+
"epoch": 15.686274509803921,
|
| 561 |
+
"grad_norm": 3.5217578411102295,
|
| 562 |
+
"learning_rate": 2.6315789473684212e-05,
|
| 563 |
+
"loss": 0.1413,
|
| 564 |
+
"step": 600
|
| 565 |
+
},
|
| 566 |
+
{
|
| 567 |
+
"epoch": 15.947712418300654,
|
| 568 |
+
"grad_norm": 5.222166538238525,
|
| 569 |
+
"learning_rate": 2.5828460038986357e-05,
|
| 570 |
+
"loss": 0.16,
|
| 571 |
+
"step": 610
|
| 572 |
+
},
|
| 573 |
+
{
|
| 574 |
+
"epoch": 16.0,
|
| 575 |
+
"eval_accuracy": 0.875,
|
| 576 |
+
"eval_loss": 0.41041284799575806,
|
| 577 |
+
"eval_runtime": 32.0124,
|
| 578 |
+
"eval_samples_per_second": 4.248,
|
| 579 |
+
"eval_steps_per_second": 0.531,
|
| 580 |
+
"step": 612
|
| 581 |
+
},
|
| 582 |
+
{
|
| 583 |
+
"epoch": 16.209150326797385,
|
| 584 |
+
"grad_norm": 4.111217021942139,
|
| 585 |
+
"learning_rate": 2.53411306042885e-05,
|
| 586 |
+
"loss": 0.136,
|
| 587 |
+
"step": 620
|
| 588 |
+
},
|
| 589 |
+
{
|
| 590 |
+
"epoch": 16.470588235294116,
|
| 591 |
+
"grad_norm": 4.035433769226074,
|
| 592 |
+
"learning_rate": 2.485380116959064e-05,
|
| 593 |
+
"loss": 0.1369,
|
| 594 |
+
"step": 630
|
| 595 |
+
},
|
| 596 |
+
{
|
| 597 |
+
"epoch": 16.73202614379085,
|
| 598 |
+
"grad_norm": 1.081305980682373,
|
| 599 |
+
"learning_rate": 2.4366471734892787e-05,
|
| 600 |
+
"loss": 0.1331,
|
| 601 |
+
"step": 640
|
| 602 |
+
},
|
| 603 |
+
{
|
| 604 |
+
"epoch": 16.99346405228758,
|
| 605 |
+
"grad_norm": 0.21553057432174683,
|
| 606 |
+
"learning_rate": 2.3879142300194932e-05,
|
| 607 |
+
"loss": 0.1096,
|
| 608 |
+
"step": 650
|
| 609 |
+
},
|
| 610 |
+
{
|
| 611 |
+
"epoch": 16.99346405228758,
|
| 612 |
+
"eval_accuracy": 0.9117647058823529,
|
| 613 |
+
"eval_loss": 0.2919999659061432,
|
| 614 |
+
"eval_runtime": 32.5572,
|
| 615 |
+
"eval_samples_per_second": 4.177,
|
| 616 |
+
"eval_steps_per_second": 0.522,
|
| 617 |
+
"step": 650
|
| 618 |
+
},
|
| 619 |
+
{
|
| 620 |
+
"epoch": 17.254901960784313,
|
| 621 |
+
"grad_norm": 5.312492847442627,
|
| 622 |
+
"learning_rate": 2.3391812865497074e-05,
|
| 623 |
+
"loss": 0.0943,
|
| 624 |
+
"step": 660
|
| 625 |
+
},
|
| 626 |
+
{
|
| 627 |
+
"epoch": 17.516339869281047,
|
| 628 |
+
"grad_norm": 0.38019096851348877,
|
| 629 |
+
"learning_rate": 2.290448343079922e-05,
|
| 630 |
+
"loss": 0.1058,
|
| 631 |
+
"step": 670
|
| 632 |
+
},
|
| 633 |
+
{
|
| 634 |
+
"epoch": 17.77777777777778,
|
| 635 |
+
"grad_norm": 9.7896089553833,
|
| 636 |
+
"learning_rate": 2.2417153996101365e-05,
|
| 637 |
+
"loss": 0.1408,
|
| 638 |
+
"step": 680
|
| 639 |
+
},
|
| 640 |
+
{
|
| 641 |
+
"epoch": 17.986928104575163,
|
| 642 |
+
"eval_accuracy": 0.9044117647058824,
|
| 643 |
+
"eval_loss": 0.26298093795776367,
|
| 644 |
+
"eval_runtime": 31.7137,
|
| 645 |
+
"eval_samples_per_second": 4.288,
|
| 646 |
+
"eval_steps_per_second": 0.536,
|
| 647 |
+
"step": 688
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 18.03921568627451,
|
| 651 |
+
"grad_norm": 1.3320598602294922,
|
| 652 |
+
"learning_rate": 2.1929824561403507e-05,
|
| 653 |
+
"loss": 0.1352,
|
| 654 |
+
"step": 690
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 18.30065359477124,
|
| 658 |
+
"grad_norm": 1.0650367736816406,
|
| 659 |
+
"learning_rate": 2.1442495126705653e-05,
|
| 660 |
+
"loss": 0.1149,
|
| 661 |
+
"step": 700
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 18.562091503267975,
|
| 665 |
+
"grad_norm": 0.5184645056724548,
|
| 666 |
+
"learning_rate": 2.0955165692007798e-05,
|
| 667 |
+
"loss": 0.1414,
|
| 668 |
+
"step": 710
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 18.823529411764707,
|
| 672 |
+
"grad_norm": 0.20009098947048187,
|
| 673 |
+
"learning_rate": 2.046783625730994e-05,
|
| 674 |
+
"loss": 0.113,
|
| 675 |
+
"step": 720
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 18.980392156862745,
|
| 679 |
+
"eval_accuracy": 0.8823529411764706,
|
| 680 |
+
"eval_loss": 0.30839478969573975,
|
| 681 |
+
"eval_runtime": 31.8589,
|
| 682 |
+
"eval_samples_per_second": 4.269,
|
| 683 |
+
"eval_steps_per_second": 0.534,
|
| 684 |
+
"step": 726
|
| 685 |
+
},
|
| 686 |
+
{
|
| 687 |
+
"epoch": 19.084967320261438,
|
| 688 |
+
"grad_norm": 0.23899152874946594,
|
| 689 |
+
"learning_rate": 1.9980506822612085e-05,
|
| 690 |
+
"loss": 0.082,
|
| 691 |
+
"step": 730
|
| 692 |
+
},
|
| 693 |
+
{
|
| 694 |
+
"epoch": 19.34640522875817,
|
| 695 |
+
"grad_norm": 2.6471519470214844,
|
| 696 |
+
"learning_rate": 1.949317738791423e-05,
|
| 697 |
+
"loss": 0.0818,
|
| 698 |
+
"step": 740
|
| 699 |
+
},
|
| 700 |
+
{
|
| 701 |
+
"epoch": 19.607843137254903,
|
| 702 |
+
"grad_norm": 3.0529627799987793,
|
| 703 |
+
"learning_rate": 1.9005847953216373e-05,
|
| 704 |
+
"loss": 0.1271,
|
| 705 |
+
"step": 750
|
| 706 |
+
},
|
| 707 |
+
{
|
| 708 |
+
"epoch": 19.869281045751634,
|
| 709 |
+
"grad_norm": 0.2476678043603897,
|
| 710 |
+
"learning_rate": 1.8518518518518518e-05,
|
| 711 |
+
"loss": 0.1272,
|
| 712 |
+
"step": 760
|
| 713 |
+
},
|
| 714 |
+
{
|
| 715 |
+
"epoch": 20.0,
|
| 716 |
+
"eval_accuracy": 0.9411764705882353,
|
| 717 |
+
"eval_loss": 0.2523283064365387,
|
| 718 |
+
"eval_runtime": 31.804,
|
| 719 |
+
"eval_samples_per_second": 4.276,
|
| 720 |
+
"eval_steps_per_second": 0.535,
|
| 721 |
+
"step": 765
|
| 722 |
+
},
|
| 723 |
+
{
|
| 724 |
+
"epoch": 20.130718954248366,
|
| 725 |
+
"grad_norm": 11.856123924255371,
|
| 726 |
+
"learning_rate": 1.8031189083820664e-05,
|
| 727 |
+
"loss": 0.1052,
|
| 728 |
+
"step": 770
|
| 729 |
+
},
|
| 730 |
+
{
|
| 731 |
+
"epoch": 20.392156862745097,
|
| 732 |
+
"grad_norm": 3.1671411991119385,
|
| 733 |
+
"learning_rate": 1.7543859649122806e-05,
|
| 734 |
+
"loss": 0.1264,
|
| 735 |
+
"step": 780
|
| 736 |
+
},
|
| 737 |
+
{
|
| 738 |
+
"epoch": 20.65359477124183,
|
| 739 |
+
"grad_norm": 0.7181591391563416,
|
| 740 |
+
"learning_rate": 1.705653021442495e-05,
|
| 741 |
+
"loss": 0.1141,
|
| 742 |
+
"step": 790
|
| 743 |
+
},
|
| 744 |
+
{
|
| 745 |
+
"epoch": 20.915032679738562,
|
| 746 |
+
"grad_norm": 3.510709762573242,
|
| 747 |
+
"learning_rate": 1.6569200779727097e-05,
|
| 748 |
+
"loss": 0.119,
|
| 749 |
+
"step": 800
|
| 750 |
+
},
|
| 751 |
+
{
|
| 752 |
+
"epoch": 20.99346405228758,
|
| 753 |
+
"eval_accuracy": 0.8823529411764706,
|
| 754 |
+
"eval_loss": 0.42536023259162903,
|
| 755 |
+
"eval_runtime": 32.0193,
|
| 756 |
+
"eval_samples_per_second": 4.247,
|
| 757 |
+
"eval_steps_per_second": 0.531,
|
| 758 |
+
"step": 803
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 21.176470588235293,
|
| 762 |
+
"grad_norm": 2.389382839202881,
|
| 763 |
+
"learning_rate": 1.608187134502924e-05,
|
| 764 |
+
"loss": 0.1316,
|
| 765 |
+
"step": 810
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 21.437908496732025,
|
| 769 |
+
"grad_norm": 6.068481922149658,
|
| 770 |
+
"learning_rate": 1.5594541910331384e-05,
|
| 771 |
+
"loss": 0.1126,
|
| 772 |
+
"step": 820
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 21.69934640522876,
|
| 776 |
+
"grad_norm": 0.7834749817848206,
|
| 777 |
+
"learning_rate": 1.5107212475633528e-05,
|
| 778 |
+
"loss": 0.0782,
|
| 779 |
+
"step": 830
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 21.96078431372549,
|
| 783 |
+
"grad_norm": 2.6613547801971436,
|
| 784 |
+
"learning_rate": 1.4619883040935673e-05,
|
| 785 |
+
"loss": 0.1068,
|
| 786 |
+
"step": 840
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 21.986928104575163,
|
| 790 |
+
"eval_accuracy": 0.8970588235294118,
|
| 791 |
+
"eval_loss": 0.35191574692726135,
|
| 792 |
+
"eval_runtime": 31.8243,
|
| 793 |
+
"eval_samples_per_second": 4.273,
|
| 794 |
+
"eval_steps_per_second": 0.534,
|
| 795 |
+
"step": 841
|
| 796 |
+
},
|
| 797 |
+
{
|
| 798 |
+
"epoch": 22.22222222222222,
|
| 799 |
+
"grad_norm": 2.7536206245422363,
|
| 800 |
+
"learning_rate": 1.4132553606237817e-05,
|
| 801 |
+
"loss": 0.1058,
|
| 802 |
+
"step": 850
|
| 803 |
+
},
|
| 804 |
+
{
|
| 805 |
+
"epoch": 22.483660130718953,
|
| 806 |
+
"grad_norm": 5.085578441619873,
|
| 807 |
+
"learning_rate": 1.364522417153996e-05,
|
| 808 |
+
"loss": 0.0847,
|
| 809 |
+
"step": 860
|
| 810 |
+
},
|
| 811 |
+
{
|
| 812 |
+
"epoch": 22.745098039215687,
|
| 813 |
+
"grad_norm": 1.596222996711731,
|
| 814 |
+
"learning_rate": 1.3157894736842106e-05,
|
| 815 |
+
"loss": 0.0723,
|
| 816 |
+
"step": 870
|
| 817 |
+
},
|
| 818 |
+
{
|
| 819 |
+
"epoch": 22.980392156862745,
|
| 820 |
+
"eval_accuracy": 0.9191176470588235,
|
| 821 |
+
"eval_loss": 0.32932135462760925,
|
| 822 |
+
"eval_runtime": 33.5379,
|
| 823 |
+
"eval_samples_per_second": 4.055,
|
| 824 |
+
"eval_steps_per_second": 0.507,
|
| 825 |
+
"step": 879
|
| 826 |
+
},
|
| 827 |
+
{
|
| 828 |
+
"epoch": 23.00653594771242,
|
| 829 |
+
"grad_norm": 1.2901531457901,
|
| 830 |
+
"learning_rate": 1.267056530214425e-05,
|
| 831 |
+
"loss": 0.0937,
|
| 832 |
+
"step": 880
|
| 833 |
+
},
|
| 834 |
+
{
|
| 835 |
+
"epoch": 23.26797385620915,
|
| 836 |
+
"grad_norm": 0.6107073426246643,
|
| 837 |
+
"learning_rate": 1.2183235867446393e-05,
|
| 838 |
+
"loss": 0.076,
|
| 839 |
+
"step": 890
|
| 840 |
+
},
|
| 841 |
+
{
|
| 842 |
+
"epoch": 23.529411764705884,
|
| 843 |
+
"grad_norm": 0.1317284107208252,
|
| 844 |
+
"learning_rate": 1.1695906432748537e-05,
|
| 845 |
+
"loss": 0.0742,
|
| 846 |
+
"step": 900
|
| 847 |
+
},
|
| 848 |
+
{
|
| 849 |
+
"epoch": 23.790849673202615,
|
| 850 |
+
"grad_norm": 1.982415795326233,
|
| 851 |
+
"learning_rate": 1.1208576998050683e-05,
|
| 852 |
+
"loss": 0.0769,
|
| 853 |
+
"step": 910
|
| 854 |
+
},
|
| 855 |
+
{
|
| 856 |
+
"epoch": 24.0,
|
| 857 |
+
"eval_accuracy": 0.9264705882352942,
|
| 858 |
+
"eval_loss": 0.2613476812839508,
|
| 859 |
+
"eval_runtime": 32.3164,
|
| 860 |
+
"eval_samples_per_second": 4.208,
|
| 861 |
+
"eval_steps_per_second": 0.526,
|
| 862 |
+
"step": 918
|
| 863 |
+
},
|
| 864 |
+
{
|
| 865 |
+
"epoch": 24.052287581699346,
|
| 866 |
+
"grad_norm": 0.13533662259578705,
|
| 867 |
+
"learning_rate": 1.0721247563352826e-05,
|
| 868 |
+
"loss": 0.0872,
|
| 869 |
+
"step": 920
|
| 870 |
+
},
|
| 871 |
+
{
|
| 872 |
+
"epoch": 24.313725490196077,
|
| 873 |
+
"grad_norm": 0.18768663704395294,
|
| 874 |
+
"learning_rate": 1.023391812865497e-05,
|
| 875 |
+
"loss": 0.0836,
|
| 876 |
+
"step": 930
|
| 877 |
+
},
|
| 878 |
+
{
|
| 879 |
+
"epoch": 24.575163398692812,
|
| 880 |
+
"grad_norm": 1.9041831493377686,
|
| 881 |
+
"learning_rate": 9.746588693957115e-06,
|
| 882 |
+
"loss": 0.0968,
|
| 883 |
+
"step": 940
|
| 884 |
+
},
|
| 885 |
+
{
|
| 886 |
+
"epoch": 24.836601307189543,
|
| 887 |
+
"grad_norm": 1.782492756843567,
|
| 888 |
+
"learning_rate": 9.259259259259259e-06,
|
| 889 |
+
"loss": 0.095,
|
| 890 |
+
"step": 950
|
| 891 |
+
},
|
| 892 |
+
{
|
| 893 |
+
"epoch": 24.99346405228758,
|
| 894 |
+
"eval_accuracy": 0.9411764705882353,
|
| 895 |
+
"eval_loss": 0.26091232895851135,
|
| 896 |
+
"eval_runtime": 32.1423,
|
| 897 |
+
"eval_samples_per_second": 4.231,
|
| 898 |
+
"eval_steps_per_second": 0.529,
|
| 899 |
+
"step": 956
|
| 900 |
+
},
|
| 901 |
+
{
|
| 902 |
+
"epoch": 25.098039215686274,
|
| 903 |
+
"grad_norm": 0.38502180576324463,
|
| 904 |
+
"learning_rate": 8.771929824561403e-06,
|
| 905 |
+
"loss": 0.0871,
|
| 906 |
+
"step": 960
|
| 907 |
+
},
|
| 908 |
+
{
|
| 909 |
+
"epoch": 25.359477124183005,
|
| 910 |
+
"grad_norm": 7.287825584411621,
|
| 911 |
+
"learning_rate": 8.284600389863548e-06,
|
| 912 |
+
"loss": 0.0662,
|
| 913 |
+
"step": 970
|
| 914 |
+
},
|
| 915 |
+
{
|
| 916 |
+
"epoch": 25.62091503267974,
|
| 917 |
+
"grad_norm": 3.129642963409424,
|
| 918 |
+
"learning_rate": 7.797270955165692e-06,
|
| 919 |
+
"loss": 0.0506,
|
| 920 |
+
"step": 980
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 25.88235294117647,
|
| 924 |
+
"grad_norm": 0.24554868042469025,
|
| 925 |
+
"learning_rate": 7.3099415204678366e-06,
|
| 926 |
+
"loss": 0.0863,
|
| 927 |
+
"step": 990
|
| 928 |
+
},
|
| 929 |
+
{
|
| 930 |
+
"epoch": 25.986928104575163,
|
| 931 |
+
"eval_accuracy": 0.9264705882352942,
|
| 932 |
+
"eval_loss": 0.264960914850235,
|
| 933 |
+
"eval_runtime": 32.1045,
|
| 934 |
+
"eval_samples_per_second": 4.236,
|
| 935 |
+
"eval_steps_per_second": 0.53,
|
| 936 |
+
"step": 994
|
| 937 |
+
},
|
| 938 |
+
{
|
| 939 |
+
"epoch": 26.143790849673202,
|
| 940 |
+
"grad_norm": 1.0143533945083618,
|
| 941 |
+
"learning_rate": 6.82261208576998e-06,
|
| 942 |
+
"loss": 0.0604,
|
| 943 |
+
"step": 1000
|
| 944 |
+
},
|
| 945 |
+
{
|
| 946 |
+
"epoch": 26.405228758169933,
|
| 947 |
+
"grad_norm": 0.31393057107925415,
|
| 948 |
+
"learning_rate": 6.335282651072125e-06,
|
| 949 |
+
"loss": 0.0589,
|
| 950 |
+
"step": 1010
|
| 951 |
+
},
|
| 952 |
+
{
|
| 953 |
+
"epoch": 26.666666666666668,
|
| 954 |
+
"grad_norm": 5.615325450897217,
|
| 955 |
+
"learning_rate": 5.8479532163742686e-06,
|
| 956 |
+
"loss": 0.0654,
|
| 957 |
+
"step": 1020
|
| 958 |
+
},
|
| 959 |
+
{
|
| 960 |
+
"epoch": 26.9281045751634,
|
| 961 |
+
"grad_norm": 0.4632560610771179,
|
| 962 |
+
"learning_rate": 5.360623781676413e-06,
|
| 963 |
+
"loss": 0.0795,
|
| 964 |
+
"step": 1030
|
| 965 |
+
},
|
| 966 |
+
{
|
| 967 |
+
"epoch": 26.980392156862745,
|
| 968 |
+
"eval_accuracy": 0.9117647058823529,
|
| 969 |
+
"eval_loss": 0.297785222530365,
|
| 970 |
+
"eval_runtime": 31.5613,
|
| 971 |
+
"eval_samples_per_second": 4.309,
|
| 972 |
+
"eval_steps_per_second": 0.539,
|
| 973 |
+
"step": 1032
|
| 974 |
+
},
|
| 975 |
+
{
|
| 976 |
+
"epoch": 27.18954248366013,
|
| 977 |
+
"grad_norm": 0.2585015594959259,
|
| 978 |
+
"learning_rate": 4.873294346978558e-06,
|
| 979 |
+
"loss": 0.0719,
|
| 980 |
+
"step": 1040
|
| 981 |
+
},
|
| 982 |
+
{
|
| 983 |
+
"epoch": 27.45098039215686,
|
| 984 |
+
"grad_norm": 0.364827960729599,
|
| 985 |
+
"learning_rate": 4.3859649122807014e-06,
|
| 986 |
+
"loss": 0.0837,
|
| 987 |
+
"step": 1050
|
| 988 |
+
},
|
| 989 |
+
{
|
| 990 |
+
"epoch": 27.712418300653596,
|
| 991 |
+
"grad_norm": 0.14460325241088867,
|
| 992 |
+
"learning_rate": 3.898635477582846e-06,
|
| 993 |
+
"loss": 0.0483,
|
| 994 |
+
"step": 1060
|
| 995 |
+
},
|
| 996 |
+
{
|
| 997 |
+
"epoch": 27.973856209150327,
|
| 998 |
+
"grad_norm": 0.43661826848983765,
|
| 999 |
+
"learning_rate": 3.41130604288499e-06,
|
| 1000 |
+
"loss": 0.0564,
|
| 1001 |
+
"step": 1070
|
| 1002 |
+
},
|
| 1003 |
+
{
|
| 1004 |
+
"epoch": 28.0,
|
| 1005 |
+
"eval_accuracy": 0.9191176470588235,
|
| 1006 |
+
"eval_loss": 0.2736634314060211,
|
| 1007 |
+
"eval_runtime": 31.9037,
|
| 1008 |
+
"eval_samples_per_second": 4.263,
|
| 1009 |
+
"eval_steps_per_second": 0.533,
|
| 1010 |
+
"step": 1071
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 28.235294117647058,
|
| 1014 |
+
"grad_norm": 0.2871550917625427,
|
| 1015 |
+
"learning_rate": 2.9239766081871343e-06,
|
| 1016 |
+
"loss": 0.0676,
|
| 1017 |
+
"step": 1080
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 28.49673202614379,
|
| 1021 |
+
"grad_norm": 0.19494622945785522,
|
| 1022 |
+
"learning_rate": 2.436647173489279e-06,
|
| 1023 |
+
"loss": 0.055,
|
| 1024 |
+
"step": 1090
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 28.758169934640524,
|
| 1028 |
+
"grad_norm": 0.46231183409690857,
|
| 1029 |
+
"learning_rate": 1.949317738791423e-06,
|
| 1030 |
+
"loss": 0.0562,
|
| 1031 |
+
"step": 1100
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 28.99346405228758,
|
| 1035 |
+
"eval_accuracy": 0.9191176470588235,
|
| 1036 |
+
"eval_loss": 0.2940830886363983,
|
| 1037 |
+
"eval_runtime": 32.3077,
|
| 1038 |
+
"eval_samples_per_second": 4.21,
|
| 1039 |
+
"eval_steps_per_second": 0.526,
|
| 1040 |
+
"step": 1109
|
| 1041 |
+
},
|
| 1042 |
+
{
|
| 1043 |
+
"epoch": 29.019607843137255,
|
| 1044 |
+
"grad_norm": 0.13441592454910278,
|
| 1045 |
+
"learning_rate": 1.4619883040935671e-06,
|
| 1046 |
+
"loss": 0.049,
|
| 1047 |
+
"step": 1110
|
| 1048 |
+
},
|
| 1049 |
+
{
|
| 1050 |
+
"epoch": 29.281045751633986,
|
| 1051 |
+
"grad_norm": 5.47345495223999,
|
| 1052 |
+
"learning_rate": 9.746588693957115e-07,
|
| 1053 |
+
"loss": 0.0712,
|
| 1054 |
+
"step": 1120
|
| 1055 |
+
},
|
| 1056 |
+
{
|
| 1057 |
+
"epoch": 29.54248366013072,
|
| 1058 |
+
"grad_norm": 0.6587017178535461,
|
| 1059 |
+
"learning_rate": 4.873294346978557e-07,
|
| 1060 |
+
"loss": 0.0679,
|
| 1061 |
+
"step": 1130
|
| 1062 |
+
},
|
| 1063 |
+
{
|
| 1064 |
+
"epoch": 29.80392156862745,
|
| 1065 |
+
"grad_norm": 0.12627951800823212,
|
| 1066 |
+
"learning_rate": 0.0,
|
| 1067 |
+
"loss": 0.0751,
|
| 1068 |
+
"step": 1140
|
| 1069 |
+
},
|
| 1070 |
+
{
|
| 1071 |
+
"epoch": 29.80392156862745,
|
| 1072 |
+
"eval_accuracy": 0.9191176470588235,
|
| 1073 |
+
"eval_loss": 0.3110995590686798,
|
| 1074 |
+
"eval_runtime": 33.2625,
|
| 1075 |
+
"eval_samples_per_second": 4.089,
|
| 1076 |
+
"eval_steps_per_second": 0.511,
|
| 1077 |
+
"step": 1140
|
| 1078 |
+
},
|
| 1079 |
+
{
|
| 1080 |
+
"epoch": 29.80392156862745,
|
| 1081 |
+
"step": 1140,
|
| 1082 |
+
"total_flos": 2.827009127861453e+18,
|
| 1083 |
+
"train_loss": 0.3311853929046999,
|
| 1084 |
+
"train_runtime": 10131.822,
|
| 1085 |
+
"train_samples_per_second": 3.624,
|
| 1086 |
+
"train_steps_per_second": 0.113
|
| 1087 |
+
}
|
| 1088 |
+
],
|
| 1089 |
+
"logging_steps": 10,
|
| 1090 |
+
"max_steps": 1140,
|
| 1091 |
+
"num_input_tokens_seen": 0,
|
| 1092 |
+
"num_train_epochs": 30,
|
| 1093 |
+
"save_steps": 500,
|
| 1094 |
+
"stateful_callbacks": {
|
| 1095 |
+
"TrainerControl": {
|
| 1096 |
+
"args": {
|
| 1097 |
+
"should_epoch_stop": false,
|
| 1098 |
+
"should_evaluate": false,
|
| 1099 |
+
"should_log": false,
|
| 1100 |
+
"should_save": true,
|
| 1101 |
+
"should_training_stop": true
|
| 1102 |
+
},
|
| 1103 |
+
"attributes": {}
|
| 1104 |
+
}
|
| 1105 |
+
},
|
| 1106 |
+
"total_flos": 2.827009127861453e+18,
|
| 1107 |
+
"train_batch_size": 8,
|
| 1108 |
+
"trial_name": null,
|
| 1109 |
+
"trial_params": null
|
| 1110 |
+
}
|