End of training
Browse files- README.md +7 -3
- all_results.json +13 -0
- eval_results.json +8 -0
- train_results.json +8 -0
- trainer_state.json +1139 -0
README.md
CHANGED
|
@@ -1,6 +1,10 @@
|
|
| 1 |
---
|
| 2 |
library_name: transformers
|
|
|
|
|
|
|
| 3 |
tags:
|
|
|
|
|
|
|
| 4 |
- generated_from_trainer
|
| 5 |
metrics:
|
| 6 |
- accuracy
|
|
@@ -14,10 +18,10 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 14 |
|
| 15 |
# mobilevit-small_alpha0.5_temp3.0_t3
|
| 16 |
|
| 17 |
-
This model is a fine-tuned version of [](https://huggingface.co/) on
|
| 18 |
It achieves the following results on the evaluation set:
|
| 19 |
-
- Loss: 0.
|
| 20 |
-
- Accuracy: 0.
|
| 21 |
|
| 22 |
## Model description
|
| 23 |
|
|
|
|
| 1 |
---
|
| 2 |
library_name: transformers
|
| 3 |
+
license: apache-2.0
|
| 4 |
+
base_model: c14kevincardenas/ClimBEiT-t3
|
| 5 |
tags:
|
| 6 |
+
- knowledge_distillation
|
| 7 |
+
- vision
|
| 8 |
- generated_from_trainer
|
| 9 |
metrics:
|
| 10 |
- accuracy
|
|
|
|
| 18 |
|
| 19 |
# mobilevit-small_alpha0.5_temp3.0_t3
|
| 20 |
|
| 21 |
+
This model is a fine-tuned version of [c14kevincardenas/ClimBEiT-t3](https://huggingface.co/c14kevincardenas/ClimBEiT-t3) on the c14kevincardenas/beta_caller_284_person_crop_seq_withlimb dataset.
|
| 22 |
It achieves the following results on the evaluation set:
|
| 23 |
+
- Loss: 0.8562
|
| 24 |
+
- Accuracy: 0.7072
|
| 25 |
|
| 26 |
## Model description
|
| 27 |
|
all_results.json
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 20.0,
|
| 3 |
+
"eval_accuracy": 0.7071583514099783,
|
| 4 |
+
"eval_loss": 0.8562042713165283,
|
| 5 |
+
"eval_runtime": 41.819,
|
| 6 |
+
"eval_samples_per_second": 22.047,
|
| 7 |
+
"eval_steps_per_second": 0.693,
|
| 8 |
+
"total_flos": 0.0,
|
| 9 |
+
"train_loss": 0.5078278935537106,
|
| 10 |
+
"train_runtime": 14435.8806,
|
| 11 |
+
"train_samples_per_second": 7.238,
|
| 12 |
+
"train_steps_per_second": 0.227
|
| 13 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 20.0,
|
| 3 |
+
"eval_accuracy": 0.7071583514099783,
|
| 4 |
+
"eval_loss": 0.8562042713165283,
|
| 5 |
+
"eval_runtime": 41.819,
|
| 6 |
+
"eval_samples_per_second": 22.047,
|
| 7 |
+
"eval_steps_per_second": 0.693
|
| 8 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 20.0,
|
| 3 |
+
"total_flos": 0.0,
|
| 4 |
+
"train_loss": 0.5078278935537106,
|
| 5 |
+
"train_runtime": 14435.8806,
|
| 6 |
+
"train_samples_per_second": 7.238,
|
| 7 |
+
"train_steps_per_second": 0.227
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1139 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.8562042713165283,
|
| 3 |
+
"best_model_checkpoint": "./kd_results/apple/mobilevit-small_alpha0.5_temp3.0_t3/checkpoint-2788",
|
| 4 |
+
"epoch": 20.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 3280,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.1524390243902439,
|
| 13 |
+
"grad_norm": 7.762810707092285,
|
| 14 |
+
"learning_rate": 3.8109756097560976e-06,
|
| 15 |
+
"loss": 0.9874,
|
| 16 |
+
"step": 25
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.3048780487804878,
|
| 20 |
+
"grad_norm": 2.2041540145874023,
|
| 21 |
+
"learning_rate": 7.621951219512195e-06,
|
| 22 |
+
"loss": 0.8291,
|
| 23 |
+
"step": 50
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.4573170731707317,
|
| 27 |
+
"grad_norm": 2.862013101577759,
|
| 28 |
+
"learning_rate": 1.1432926829268294e-05,
|
| 29 |
+
"loss": 0.809,
|
| 30 |
+
"step": 75
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.6097560975609756,
|
| 34 |
+
"grad_norm": 3.8926451206207275,
|
| 35 |
+
"learning_rate": 1.524390243902439e-05,
|
| 36 |
+
"loss": 0.8056,
|
| 37 |
+
"step": 100
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.7621951219512195,
|
| 41 |
+
"grad_norm": 3.9026598930358887,
|
| 42 |
+
"learning_rate": 1.9054878048780488e-05,
|
| 43 |
+
"loss": 0.8081,
|
| 44 |
+
"step": 125
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.9146341463414634,
|
| 48 |
+
"grad_norm": 2.8420770168304443,
|
| 49 |
+
"learning_rate": 2.286585365853659e-05,
|
| 50 |
+
"loss": 0.8034,
|
| 51 |
+
"step": 150
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 1.0,
|
| 55 |
+
"eval_accuracy": 0.28633405639913234,
|
| 56 |
+
"eval_loss": 1.3824383020401,
|
| 57 |
+
"eval_runtime": 40.4477,
|
| 58 |
+
"eval_samples_per_second": 22.795,
|
| 59 |
+
"eval_steps_per_second": 0.717,
|
| 60 |
+
"step": 164
|
| 61 |
+
},
|
| 62 |
+
{
|
| 63 |
+
"epoch": 1.0670731707317074,
|
| 64 |
+
"grad_norm": 2.4602479934692383,
|
| 65 |
+
"learning_rate": 2.6676829268292686e-05,
|
| 66 |
+
"loss": 0.8053,
|
| 67 |
+
"step": 175
|
| 68 |
+
},
|
| 69 |
+
{
|
| 70 |
+
"epoch": 1.2195121951219512,
|
| 71 |
+
"grad_norm": 2.140545606613159,
|
| 72 |
+
"learning_rate": 3.048780487804878e-05,
|
| 73 |
+
"loss": 0.7981,
|
| 74 |
+
"step": 200
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 1.3719512195121952,
|
| 78 |
+
"grad_norm": 4.539191246032715,
|
| 79 |
+
"learning_rate": 3.429878048780488e-05,
|
| 80 |
+
"loss": 0.795,
|
| 81 |
+
"step": 225
|
| 82 |
+
},
|
| 83 |
+
{
|
| 84 |
+
"epoch": 1.524390243902439,
|
| 85 |
+
"grad_norm": 3.07138729095459,
|
| 86 |
+
"learning_rate": 3.8109756097560976e-05,
|
| 87 |
+
"loss": 0.8012,
|
| 88 |
+
"step": 250
|
| 89 |
+
},
|
| 90 |
+
{
|
| 91 |
+
"epoch": 1.6768292682926829,
|
| 92 |
+
"grad_norm": 2.5887339115142822,
|
| 93 |
+
"learning_rate": 4.1920731707317077e-05,
|
| 94 |
+
"loss": 0.7953,
|
| 95 |
+
"step": 275
|
| 96 |
+
},
|
| 97 |
+
{
|
| 98 |
+
"epoch": 1.8292682926829267,
|
| 99 |
+
"grad_norm": 4.393621444702148,
|
| 100 |
+
"learning_rate": 4.573170731707318e-05,
|
| 101 |
+
"loss": 0.772,
|
| 102 |
+
"step": 300
|
| 103 |
+
},
|
| 104 |
+
{
|
| 105 |
+
"epoch": 1.9817073170731707,
|
| 106 |
+
"grad_norm": 1.1688978672027588,
|
| 107 |
+
"learning_rate": 4.954268292682927e-05,
|
| 108 |
+
"loss": 0.7605,
|
| 109 |
+
"step": 325
|
| 110 |
+
},
|
| 111 |
+
{
|
| 112 |
+
"epoch": 2.0,
|
| 113 |
+
"eval_accuracy": 0.2917570498915401,
|
| 114 |
+
"eval_loss": 1.3373875617980957,
|
| 115 |
+
"eval_runtime": 40.6346,
|
| 116 |
+
"eval_samples_per_second": 22.69,
|
| 117 |
+
"eval_steps_per_second": 0.714,
|
| 118 |
+
"step": 328
|
| 119 |
+
},
|
| 120 |
+
{
|
| 121 |
+
"epoch": 2.1341463414634148,
|
| 122 |
+
"grad_norm": 1.0945930480957031,
|
| 123 |
+
"learning_rate": 4.962737127371274e-05,
|
| 124 |
+
"loss": 0.7616,
|
| 125 |
+
"step": 350
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"epoch": 2.2865853658536586,
|
| 129 |
+
"grad_norm": 2.745100259780884,
|
| 130 |
+
"learning_rate": 4.92039295392954e-05,
|
| 131 |
+
"loss": 0.7438,
|
| 132 |
+
"step": 375
|
| 133 |
+
},
|
| 134 |
+
{
|
| 135 |
+
"epoch": 2.4390243902439024,
|
| 136 |
+
"grad_norm": 5.273860454559326,
|
| 137 |
+
"learning_rate": 4.878048780487805e-05,
|
| 138 |
+
"loss": 0.7124,
|
| 139 |
+
"step": 400
|
| 140 |
+
},
|
| 141 |
+
{
|
| 142 |
+
"epoch": 2.591463414634146,
|
| 143 |
+
"grad_norm": 5.696925640106201,
|
| 144 |
+
"learning_rate": 4.8357046070460704e-05,
|
| 145 |
+
"loss": 0.7182,
|
| 146 |
+
"step": 425
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 2.7439024390243905,
|
| 150 |
+
"grad_norm": 5.35153865814209,
|
| 151 |
+
"learning_rate": 4.793360433604336e-05,
|
| 152 |
+
"loss": 0.7061,
|
| 153 |
+
"step": 450
|
| 154 |
+
},
|
| 155 |
+
{
|
| 156 |
+
"epoch": 2.8963414634146343,
|
| 157 |
+
"grad_norm": 4.60999059677124,
|
| 158 |
+
"learning_rate": 4.7510162601626016e-05,
|
| 159 |
+
"loss": 0.6903,
|
| 160 |
+
"step": 475
|
| 161 |
+
},
|
| 162 |
+
{
|
| 163 |
+
"epoch": 3.0,
|
| 164 |
+
"eval_accuracy": 0.5151843817787418,
|
| 165 |
+
"eval_loss": 1.140594720840454,
|
| 166 |
+
"eval_runtime": 40.2587,
|
| 167 |
+
"eval_samples_per_second": 22.902,
|
| 168 |
+
"eval_steps_per_second": 0.72,
|
| 169 |
+
"step": 492
|
| 170 |
+
},
|
| 171 |
+
{
|
| 172 |
+
"epoch": 3.048780487804878,
|
| 173 |
+
"grad_norm": 5.811767578125,
|
| 174 |
+
"learning_rate": 4.7086720867208675e-05,
|
| 175 |
+
"loss": 0.6817,
|
| 176 |
+
"step": 500
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 3.201219512195122,
|
| 180 |
+
"grad_norm": 4.599453449249268,
|
| 181 |
+
"learning_rate": 4.6663279132791334e-05,
|
| 182 |
+
"loss": 0.6679,
|
| 183 |
+
"step": 525
|
| 184 |
+
},
|
| 185 |
+
{
|
| 186 |
+
"epoch": 3.3536585365853657,
|
| 187 |
+
"grad_norm": 6.098621368408203,
|
| 188 |
+
"learning_rate": 4.6239837398373986e-05,
|
| 189 |
+
"loss": 0.6385,
|
| 190 |
+
"step": 550
|
| 191 |
+
},
|
| 192 |
+
{
|
| 193 |
+
"epoch": 3.5060975609756095,
|
| 194 |
+
"grad_norm": 6.299458026885986,
|
| 195 |
+
"learning_rate": 4.581639566395664e-05,
|
| 196 |
+
"loss": 0.6364,
|
| 197 |
+
"step": 575
|
| 198 |
+
},
|
| 199 |
+
{
|
| 200 |
+
"epoch": 3.658536585365854,
|
| 201 |
+
"grad_norm": 3.688998222351074,
|
| 202 |
+
"learning_rate": 4.53929539295393e-05,
|
| 203 |
+
"loss": 0.6396,
|
| 204 |
+
"step": 600
|
| 205 |
+
},
|
| 206 |
+
{
|
| 207 |
+
"epoch": 3.8109756097560976,
|
| 208 |
+
"grad_norm": 5.4493727684021,
|
| 209 |
+
"learning_rate": 4.496951219512196e-05,
|
| 210 |
+
"loss": 0.6175,
|
| 211 |
+
"step": 625
|
| 212 |
+
},
|
| 213 |
+
{
|
| 214 |
+
"epoch": 3.9634146341463414,
|
| 215 |
+
"grad_norm": 6.126657485961914,
|
| 216 |
+
"learning_rate": 4.454607046070461e-05,
|
| 217 |
+
"loss": 0.6045,
|
| 218 |
+
"step": 650
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 4.0,
|
| 222 |
+
"eval_accuracy": 0.5629067245119306,
|
| 223 |
+
"eval_loss": 1.0571489334106445,
|
| 224 |
+
"eval_runtime": 40.0524,
|
| 225 |
+
"eval_samples_per_second": 23.02,
|
| 226 |
+
"eval_steps_per_second": 0.724,
|
| 227 |
+
"step": 656
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 4.115853658536586,
|
| 231 |
+
"grad_norm": 8.532530784606934,
|
| 232 |
+
"learning_rate": 4.412262872628726e-05,
|
| 233 |
+
"loss": 0.6161,
|
| 234 |
+
"step": 675
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 4.2682926829268295,
|
| 238 |
+
"grad_norm": 3.7659671306610107,
|
| 239 |
+
"learning_rate": 4.369918699186992e-05,
|
| 240 |
+
"loss": 0.603,
|
| 241 |
+
"step": 700
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 4.420731707317073,
|
| 245 |
+
"grad_norm": 2.876420736312866,
|
| 246 |
+
"learning_rate": 4.3275745257452574e-05,
|
| 247 |
+
"loss": 0.5798,
|
| 248 |
+
"step": 725
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 4.573170731707317,
|
| 252 |
+
"grad_norm": 7.7819013595581055,
|
| 253 |
+
"learning_rate": 4.285230352303523e-05,
|
| 254 |
+
"loss": 0.5904,
|
| 255 |
+
"step": 750
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 4.725609756097561,
|
| 259 |
+
"grad_norm": 5.8853678703308105,
|
| 260 |
+
"learning_rate": 4.242886178861789e-05,
|
| 261 |
+
"loss": 0.573,
|
| 262 |
+
"step": 775
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 4.878048780487805,
|
| 266 |
+
"grad_norm": 6.983067989349365,
|
| 267 |
+
"learning_rate": 4.2005420054200545e-05,
|
| 268 |
+
"loss": 0.5779,
|
| 269 |
+
"step": 800
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 5.0,
|
| 273 |
+
"eval_accuracy": 0.5813449023861171,
|
| 274 |
+
"eval_loss": 1.022658348083496,
|
| 275 |
+
"eval_runtime": 40.4951,
|
| 276 |
+
"eval_samples_per_second": 22.768,
|
| 277 |
+
"eval_steps_per_second": 0.716,
|
| 278 |
+
"step": 820
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 5.030487804878049,
|
| 282 |
+
"grad_norm": 7.891158103942871,
|
| 283 |
+
"learning_rate": 4.15819783197832e-05,
|
| 284 |
+
"loss": 0.5842,
|
| 285 |
+
"step": 825
|
| 286 |
+
},
|
| 287 |
+
{
|
| 288 |
+
"epoch": 5.182926829268292,
|
| 289 |
+
"grad_norm": 4.526270389556885,
|
| 290 |
+
"learning_rate": 4.1158536585365856e-05,
|
| 291 |
+
"loss": 0.55,
|
| 292 |
+
"step": 850
|
| 293 |
+
},
|
| 294 |
+
{
|
| 295 |
+
"epoch": 5.335365853658536,
|
| 296 |
+
"grad_norm": 3.818694591522217,
|
| 297 |
+
"learning_rate": 4.0735094850948516e-05,
|
| 298 |
+
"loss": 0.5396,
|
| 299 |
+
"step": 875
|
| 300 |
+
},
|
| 301 |
+
{
|
| 302 |
+
"epoch": 5.487804878048781,
|
| 303 |
+
"grad_norm": 2.55497670173645,
|
| 304 |
+
"learning_rate": 4.031165311653117e-05,
|
| 305 |
+
"loss": 0.554,
|
| 306 |
+
"step": 900
|
| 307 |
+
},
|
| 308 |
+
{
|
| 309 |
+
"epoch": 5.640243902439025,
|
| 310 |
+
"grad_norm": 5.239187717437744,
|
| 311 |
+
"learning_rate": 3.988821138211382e-05,
|
| 312 |
+
"loss": 0.5413,
|
| 313 |
+
"step": 925
|
| 314 |
+
},
|
| 315 |
+
{
|
| 316 |
+
"epoch": 5.7926829268292686,
|
| 317 |
+
"grad_norm": 3.5609655380249023,
|
| 318 |
+
"learning_rate": 3.946476964769648e-05,
|
| 319 |
+
"loss": 0.5388,
|
| 320 |
+
"step": 950
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 5.945121951219512,
|
| 324 |
+
"grad_norm": 2.538914918899536,
|
| 325 |
+
"learning_rate": 3.904132791327913e-05,
|
| 326 |
+
"loss": 0.5423,
|
| 327 |
+
"step": 975
|
| 328 |
+
},
|
| 329 |
+
{
|
| 330 |
+
"epoch": 6.0,
|
| 331 |
+
"eval_accuracy": 0.6334056399132321,
|
| 332 |
+
"eval_loss": 0.9594108462333679,
|
| 333 |
+
"eval_runtime": 39.996,
|
| 334 |
+
"eval_samples_per_second": 23.052,
|
| 335 |
+
"eval_steps_per_second": 0.725,
|
| 336 |
+
"step": 984
|
| 337 |
+
},
|
| 338 |
+
{
|
| 339 |
+
"epoch": 6.097560975609756,
|
| 340 |
+
"grad_norm": 9.002790451049805,
|
| 341 |
+
"learning_rate": 3.861788617886179e-05,
|
| 342 |
+
"loss": 0.5333,
|
| 343 |
+
"step": 1000
|
| 344 |
+
},
|
| 345 |
+
{
|
| 346 |
+
"epoch": 6.25,
|
| 347 |
+
"grad_norm": 2.795607805252075,
|
| 348 |
+
"learning_rate": 3.8194444444444444e-05,
|
| 349 |
+
"loss": 0.5304,
|
| 350 |
+
"step": 1025
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 6.402439024390244,
|
| 354 |
+
"grad_norm": 3.0136711597442627,
|
| 355 |
+
"learning_rate": 3.77710027100271e-05,
|
| 356 |
+
"loss": 0.5338,
|
| 357 |
+
"step": 1050
|
| 358 |
+
},
|
| 359 |
+
{
|
| 360 |
+
"epoch": 6.554878048780488,
|
| 361 |
+
"grad_norm": 4.359175205230713,
|
| 362 |
+
"learning_rate": 3.7347560975609755e-05,
|
| 363 |
+
"loss": 0.5213,
|
| 364 |
+
"step": 1075
|
| 365 |
+
},
|
| 366 |
+
{
|
| 367 |
+
"epoch": 6.7073170731707314,
|
| 368 |
+
"grad_norm": 5.616451740264893,
|
| 369 |
+
"learning_rate": 3.6924119241192415e-05,
|
| 370 |
+
"loss": 0.5211,
|
| 371 |
+
"step": 1100
|
| 372 |
+
},
|
| 373 |
+
{
|
| 374 |
+
"epoch": 6.859756097560975,
|
| 375 |
+
"grad_norm": 3.337740659713745,
|
| 376 |
+
"learning_rate": 3.6500677506775074e-05,
|
| 377 |
+
"loss": 0.5154,
|
| 378 |
+
"step": 1125
|
| 379 |
+
},
|
| 380 |
+
{
|
| 381 |
+
"epoch": 7.0,
|
| 382 |
+
"eval_accuracy": 0.6518438177874186,
|
| 383 |
+
"eval_loss": 0.9497526288032532,
|
| 384 |
+
"eval_runtime": 40.4853,
|
| 385 |
+
"eval_samples_per_second": 22.774,
|
| 386 |
+
"eval_steps_per_second": 0.716,
|
| 387 |
+
"step": 1148
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 7.012195121951219,
|
| 391 |
+
"grad_norm": 3.785557746887207,
|
| 392 |
+
"learning_rate": 3.6077235772357726e-05,
|
| 393 |
+
"loss": 0.5004,
|
| 394 |
+
"step": 1150
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 7.164634146341464,
|
| 398 |
+
"grad_norm": 4.928786277770996,
|
| 399 |
+
"learning_rate": 3.565379403794038e-05,
|
| 400 |
+
"loss": 0.5024,
|
| 401 |
+
"step": 1175
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 7.317073170731708,
|
| 405 |
+
"grad_norm": 3.090078353881836,
|
| 406 |
+
"learning_rate": 3.523035230352303e-05,
|
| 407 |
+
"loss": 0.4839,
|
| 408 |
+
"step": 1200
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 7.469512195121951,
|
| 412 |
+
"grad_norm": 6.933969020843506,
|
| 413 |
+
"learning_rate": 3.48069105691057e-05,
|
| 414 |
+
"loss": 0.4764,
|
| 415 |
+
"step": 1225
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 7.621951219512195,
|
| 419 |
+
"grad_norm": 6.955783367156982,
|
| 420 |
+
"learning_rate": 3.438346883468835e-05,
|
| 421 |
+
"loss": 0.5134,
|
| 422 |
+
"step": 1250
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 7.774390243902439,
|
| 426 |
+
"grad_norm": 4.848341941833496,
|
| 427 |
+
"learning_rate": 3.3960027100271e-05,
|
| 428 |
+
"loss": 0.5051,
|
| 429 |
+
"step": 1275
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 7.926829268292683,
|
| 433 |
+
"grad_norm": 3.339418888092041,
|
| 434 |
+
"learning_rate": 3.353658536585366e-05,
|
| 435 |
+
"loss": 0.4984,
|
| 436 |
+
"step": 1300
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 8.0,
|
| 440 |
+
"eval_accuracy": 0.675704989154013,
|
| 441 |
+
"eval_loss": 0.9110548496246338,
|
| 442 |
+
"eval_runtime": 40.3793,
|
| 443 |
+
"eval_samples_per_second": 22.833,
|
| 444 |
+
"eval_steps_per_second": 0.718,
|
| 445 |
+
"step": 1312
|
| 446 |
+
},
|
| 447 |
+
{
|
| 448 |
+
"epoch": 8.079268292682928,
|
| 449 |
+
"grad_norm": 5.299203395843506,
|
| 450 |
+
"learning_rate": 3.3113143631436314e-05,
|
| 451 |
+
"loss": 0.4745,
|
| 452 |
+
"step": 1325
|
| 453 |
+
},
|
| 454 |
+
{
|
| 455 |
+
"epoch": 8.231707317073171,
|
| 456 |
+
"grad_norm": 5.381677627563477,
|
| 457 |
+
"learning_rate": 3.268970189701897e-05,
|
| 458 |
+
"loss": 0.4696,
|
| 459 |
+
"step": 1350
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 8.384146341463415,
|
| 463 |
+
"grad_norm": 6.133314609527588,
|
| 464 |
+
"learning_rate": 3.226626016260163e-05,
|
| 465 |
+
"loss": 0.4876,
|
| 466 |
+
"step": 1375
|
| 467 |
+
},
|
| 468 |
+
{
|
| 469 |
+
"epoch": 8.536585365853659,
|
| 470 |
+
"grad_norm": 3.0064644813537598,
|
| 471 |
+
"learning_rate": 3.1842818428184285e-05,
|
| 472 |
+
"loss": 0.4581,
|
| 473 |
+
"step": 1400
|
| 474 |
+
},
|
| 475 |
+
{
|
| 476 |
+
"epoch": 8.689024390243903,
|
| 477 |
+
"grad_norm": 3.6373887062072754,
|
| 478 |
+
"learning_rate": 3.141937669376694e-05,
|
| 479 |
+
"loss": 0.4813,
|
| 480 |
+
"step": 1425
|
| 481 |
+
},
|
| 482 |
+
{
|
| 483 |
+
"epoch": 8.841463414634147,
|
| 484 |
+
"grad_norm": 2.668551206588745,
|
| 485 |
+
"learning_rate": 3.0995934959349596e-05,
|
| 486 |
+
"loss": 0.4616,
|
| 487 |
+
"step": 1450
|
| 488 |
+
},
|
| 489 |
+
{
|
| 490 |
+
"epoch": 8.99390243902439,
|
| 491 |
+
"grad_norm": 4.62333345413208,
|
| 492 |
+
"learning_rate": 3.0572493224932255e-05,
|
| 493 |
+
"loss": 0.47,
|
| 494 |
+
"step": 1475
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 9.0,
|
| 498 |
+
"eval_accuracy": 0.6518438177874186,
|
| 499 |
+
"eval_loss": 0.9155306816101074,
|
| 500 |
+
"eval_runtime": 39.9567,
|
| 501 |
+
"eval_samples_per_second": 23.075,
|
| 502 |
+
"eval_steps_per_second": 0.726,
|
| 503 |
+
"step": 1476
|
| 504 |
+
},
|
| 505 |
+
{
|
| 506 |
+
"epoch": 9.146341463414634,
|
| 507 |
+
"grad_norm": 4.4853434562683105,
|
| 508 |
+
"learning_rate": 3.0149051490514908e-05,
|
| 509 |
+
"loss": 0.4618,
|
| 510 |
+
"step": 1500
|
| 511 |
+
},
|
| 512 |
+
{
|
| 513 |
+
"epoch": 9.298780487804878,
|
| 514 |
+
"grad_norm": 3.469381332397461,
|
| 515 |
+
"learning_rate": 2.972560975609756e-05,
|
| 516 |
+
"loss": 0.4687,
|
| 517 |
+
"step": 1525
|
| 518 |
+
},
|
| 519 |
+
{
|
| 520 |
+
"epoch": 9.451219512195122,
|
| 521 |
+
"grad_norm": 5.934073448181152,
|
| 522 |
+
"learning_rate": 2.930216802168022e-05,
|
| 523 |
+
"loss": 0.4563,
|
| 524 |
+
"step": 1550
|
| 525 |
+
},
|
| 526 |
+
{
|
| 527 |
+
"epoch": 9.603658536585366,
|
| 528 |
+
"grad_norm": 3.998873472213745,
|
| 529 |
+
"learning_rate": 2.8878726287262875e-05,
|
| 530 |
+
"loss": 0.4502,
|
| 531 |
+
"step": 1575
|
| 532 |
+
},
|
| 533 |
+
{
|
| 534 |
+
"epoch": 9.75609756097561,
|
| 535 |
+
"grad_norm": 5.720112323760986,
|
| 536 |
+
"learning_rate": 2.8455284552845528e-05,
|
| 537 |
+
"loss": 0.4633,
|
| 538 |
+
"step": 1600
|
| 539 |
+
},
|
| 540 |
+
{
|
| 541 |
+
"epoch": 9.908536585365853,
|
| 542 |
+
"grad_norm": 2.881168842315674,
|
| 543 |
+
"learning_rate": 2.8031842818428184e-05,
|
| 544 |
+
"loss": 0.4474,
|
| 545 |
+
"step": 1625
|
| 546 |
+
},
|
| 547 |
+
{
|
| 548 |
+
"epoch": 10.0,
|
| 549 |
+
"eval_accuracy": 0.6518438177874186,
|
| 550 |
+
"eval_loss": 0.9252634644508362,
|
| 551 |
+
"eval_runtime": 39.6703,
|
| 552 |
+
"eval_samples_per_second": 23.242,
|
| 553 |
+
"eval_steps_per_second": 0.731,
|
| 554 |
+
"step": 1640
|
| 555 |
+
},
|
| 556 |
+
{
|
| 557 |
+
"epoch": 10.060975609756097,
|
| 558 |
+
"grad_norm": 4.207843780517578,
|
| 559 |
+
"learning_rate": 2.7608401084010843e-05,
|
| 560 |
+
"loss": 0.4542,
|
| 561 |
+
"step": 1650
|
| 562 |
+
},
|
| 563 |
+
{
|
| 564 |
+
"epoch": 10.213414634146341,
|
| 565 |
+
"grad_norm": 3.44258713722229,
|
| 566 |
+
"learning_rate": 2.71849593495935e-05,
|
| 567 |
+
"loss": 0.4369,
|
| 568 |
+
"step": 1675
|
| 569 |
+
},
|
| 570 |
+
{
|
| 571 |
+
"epoch": 10.365853658536585,
|
| 572 |
+
"grad_norm": 6.224090099334717,
|
| 573 |
+
"learning_rate": 2.676151761517615e-05,
|
| 574 |
+
"loss": 0.4532,
|
| 575 |
+
"step": 1700
|
| 576 |
+
},
|
| 577 |
+
{
|
| 578 |
+
"epoch": 10.518292682926829,
|
| 579 |
+
"grad_norm": 2.638032913208008,
|
| 580 |
+
"learning_rate": 2.633807588075881e-05,
|
| 581 |
+
"loss": 0.4424,
|
| 582 |
+
"step": 1725
|
| 583 |
+
},
|
| 584 |
+
{
|
| 585 |
+
"epoch": 10.670731707317072,
|
| 586 |
+
"grad_norm": 2.48111629486084,
|
| 587 |
+
"learning_rate": 2.5914634146341466e-05,
|
| 588 |
+
"loss": 0.4279,
|
| 589 |
+
"step": 1750
|
| 590 |
+
},
|
| 591 |
+
{
|
| 592 |
+
"epoch": 10.823170731707316,
|
| 593 |
+
"grad_norm": 4.390626907348633,
|
| 594 |
+
"learning_rate": 2.549119241192412e-05,
|
| 595 |
+
"loss": 0.448,
|
| 596 |
+
"step": 1775
|
| 597 |
+
},
|
| 598 |
+
{
|
| 599 |
+
"epoch": 10.975609756097562,
|
| 600 |
+
"grad_norm": 4.277877330780029,
|
| 601 |
+
"learning_rate": 2.5067750677506774e-05,
|
| 602 |
+
"loss": 0.4547,
|
| 603 |
+
"step": 1800
|
| 604 |
+
},
|
| 605 |
+
{
|
| 606 |
+
"epoch": 11.0,
|
| 607 |
+
"eval_accuracy": 0.6561822125813449,
|
| 608 |
+
"eval_loss": 0.9355393648147583,
|
| 609 |
+
"eval_runtime": 40.5434,
|
| 610 |
+
"eval_samples_per_second": 22.741,
|
| 611 |
+
"eval_steps_per_second": 0.715,
|
| 612 |
+
"step": 1804
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 11.128048780487806,
|
| 616 |
+
"grad_norm": 6.439655303955078,
|
| 617 |
+
"learning_rate": 2.464430894308943e-05,
|
| 618 |
+
"loss": 0.4528,
|
| 619 |
+
"step": 1825
|
| 620 |
+
},
|
| 621 |
+
{
|
| 622 |
+
"epoch": 11.28048780487805,
|
| 623 |
+
"grad_norm": 3.8702235221862793,
|
| 624 |
+
"learning_rate": 2.422086720867209e-05,
|
| 625 |
+
"loss": 0.4277,
|
| 626 |
+
"step": 1850
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 11.432926829268293,
|
| 630 |
+
"grad_norm": 6.396579265594482,
|
| 631 |
+
"learning_rate": 2.3797425474254745e-05,
|
| 632 |
+
"loss": 0.4298,
|
| 633 |
+
"step": 1875
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 11.585365853658537,
|
| 637 |
+
"grad_norm": 6.784517765045166,
|
| 638 |
+
"learning_rate": 2.3373983739837398e-05,
|
| 639 |
+
"loss": 0.4293,
|
| 640 |
+
"step": 1900
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 11.737804878048781,
|
| 644 |
+
"grad_norm": 3.7410504817962646,
|
| 645 |
+
"learning_rate": 2.2950542005420057e-05,
|
| 646 |
+
"loss": 0.4379,
|
| 647 |
+
"step": 1925
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 11.890243902439025,
|
| 651 |
+
"grad_norm": 6.854469299316406,
|
| 652 |
+
"learning_rate": 2.252710027100271e-05,
|
| 653 |
+
"loss": 0.4432,
|
| 654 |
+
"step": 1950
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 12.0,
|
| 658 |
+
"eval_accuracy": 0.6648590021691974,
|
| 659 |
+
"eval_loss": 0.8717988729476929,
|
| 660 |
+
"eval_runtime": 40.6359,
|
| 661 |
+
"eval_samples_per_second": 22.689,
|
| 662 |
+
"eval_steps_per_second": 0.714,
|
| 663 |
+
"step": 1968
|
| 664 |
+
},
|
| 665 |
+
{
|
| 666 |
+
"epoch": 12.042682926829269,
|
| 667 |
+
"grad_norm": 4.439425468444824,
|
| 668 |
+
"learning_rate": 2.210365853658537e-05,
|
| 669 |
+
"loss": 0.447,
|
| 670 |
+
"step": 1975
|
| 671 |
+
},
|
| 672 |
+
{
|
| 673 |
+
"epoch": 12.195121951219512,
|
| 674 |
+
"grad_norm": 1.7574331760406494,
|
| 675 |
+
"learning_rate": 2.1680216802168024e-05,
|
| 676 |
+
"loss": 0.4109,
|
| 677 |
+
"step": 2000
|
| 678 |
+
},
|
| 679 |
+
{
|
| 680 |
+
"epoch": 12.347560975609756,
|
| 681 |
+
"grad_norm": 5.339428901672363,
|
| 682 |
+
"learning_rate": 2.1256775067750677e-05,
|
| 683 |
+
"loss": 0.4288,
|
| 684 |
+
"step": 2025
|
| 685 |
+
},
|
| 686 |
+
{
|
| 687 |
+
"epoch": 12.5,
|
| 688 |
+
"grad_norm": 2.4824461936950684,
|
| 689 |
+
"learning_rate": 2.0833333333333336e-05,
|
| 690 |
+
"loss": 0.4298,
|
| 691 |
+
"step": 2050
|
| 692 |
+
},
|
| 693 |
+
{
|
| 694 |
+
"epoch": 12.652439024390244,
|
| 695 |
+
"grad_norm": 2.4411211013793945,
|
| 696 |
+
"learning_rate": 2.040989159891599e-05,
|
| 697 |
+
"loss": 0.4265,
|
| 698 |
+
"step": 2075
|
| 699 |
+
},
|
| 700 |
+
{
|
| 701 |
+
"epoch": 12.804878048780488,
|
| 702 |
+
"grad_norm": 6.373781204223633,
|
| 703 |
+
"learning_rate": 1.9986449864498648e-05,
|
| 704 |
+
"loss": 0.422,
|
| 705 |
+
"step": 2100
|
| 706 |
+
},
|
| 707 |
+
{
|
| 708 |
+
"epoch": 12.957317073170731,
|
| 709 |
+
"grad_norm": 3.202080011367798,
|
| 710 |
+
"learning_rate": 1.95630081300813e-05,
|
| 711 |
+
"loss": 0.4235,
|
| 712 |
+
"step": 2125
|
| 713 |
+
},
|
| 714 |
+
{
|
| 715 |
+
"epoch": 13.0,
|
| 716 |
+
"eval_accuracy": 0.7039045553145337,
|
| 717 |
+
"eval_loss": 0.8762956857681274,
|
| 718 |
+
"eval_runtime": 40.5234,
|
| 719 |
+
"eval_samples_per_second": 22.752,
|
| 720 |
+
"eval_steps_per_second": 0.716,
|
| 721 |
+
"step": 2132
|
| 722 |
+
},
|
| 723 |
+
{
|
| 724 |
+
"epoch": 13.109756097560975,
|
| 725 |
+
"grad_norm": 3.008108615875244,
|
| 726 |
+
"learning_rate": 1.9139566395663956e-05,
|
| 727 |
+
"loss": 0.424,
|
| 728 |
+
"step": 2150
|
| 729 |
+
},
|
| 730 |
+
{
|
| 731 |
+
"epoch": 13.262195121951219,
|
| 732 |
+
"grad_norm": 3.0317177772521973,
|
| 733 |
+
"learning_rate": 1.8716124661246615e-05,
|
| 734 |
+
"loss": 0.4246,
|
| 735 |
+
"step": 2175
|
| 736 |
+
},
|
| 737 |
+
{
|
| 738 |
+
"epoch": 13.414634146341463,
|
| 739 |
+
"grad_norm": 2.7585787773132324,
|
| 740 |
+
"learning_rate": 1.8292682926829268e-05,
|
| 741 |
+
"loss": 0.4132,
|
| 742 |
+
"step": 2200
|
| 743 |
+
},
|
| 744 |
+
{
|
| 745 |
+
"epoch": 13.567073170731707,
|
| 746 |
+
"grad_norm": 3.0039117336273193,
|
| 747 |
+
"learning_rate": 1.7869241192411927e-05,
|
| 748 |
+
"loss": 0.4076,
|
| 749 |
+
"step": 2225
|
| 750 |
+
},
|
| 751 |
+
{
|
| 752 |
+
"epoch": 13.71951219512195,
|
| 753 |
+
"grad_norm": 5.297887325286865,
|
| 754 |
+
"learning_rate": 1.744579945799458e-05,
|
| 755 |
+
"loss": 0.4,
|
| 756 |
+
"step": 2250
|
| 757 |
+
},
|
| 758 |
+
{
|
| 759 |
+
"epoch": 13.871951219512194,
|
| 760 |
+
"grad_norm": 3.2223308086395264,
|
| 761 |
+
"learning_rate": 1.702235772357724e-05,
|
| 762 |
+
"loss": 0.419,
|
| 763 |
+
"step": 2275
|
| 764 |
+
},
|
| 765 |
+
{
|
| 766 |
+
"epoch": 14.0,
|
| 767 |
+
"eval_accuracy": 0.6865509761388287,
|
| 768 |
+
"eval_loss": 0.8877863883972168,
|
| 769 |
+
"eval_runtime": 40.0927,
|
| 770 |
+
"eval_samples_per_second": 22.997,
|
| 771 |
+
"eval_steps_per_second": 0.723,
|
| 772 |
+
"step": 2296
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 14.024390243902438,
|
| 776 |
+
"grad_norm": 5.251517295837402,
|
| 777 |
+
"learning_rate": 1.659891598915989e-05,
|
| 778 |
+
"loss": 0.4151,
|
| 779 |
+
"step": 2300
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 14.176829268292684,
|
| 783 |
+
"grad_norm": 5.841452598571777,
|
| 784 |
+
"learning_rate": 1.6175474254742547e-05,
|
| 785 |
+
"loss": 0.409,
|
| 786 |
+
"step": 2325
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 14.329268292682928,
|
| 790 |
+
"grad_norm": 4.45047664642334,
|
| 791 |
+
"learning_rate": 1.5752032520325206e-05,
|
| 792 |
+
"loss": 0.4073,
|
| 793 |
+
"step": 2350
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 14.481707317073171,
|
| 797 |
+
"grad_norm": 2.3922412395477295,
|
| 798 |
+
"learning_rate": 1.532859078590786e-05,
|
| 799 |
+
"loss": 0.4017,
|
| 800 |
+
"step": 2375
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 14.634146341463415,
|
| 804 |
+
"grad_norm": 4.807758331298828,
|
| 805 |
+
"learning_rate": 1.4905149051490516e-05,
|
| 806 |
+
"loss": 0.4107,
|
| 807 |
+
"step": 2400
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 14.786585365853659,
|
| 811 |
+
"grad_norm": 2.9783949851989746,
|
| 812 |
+
"learning_rate": 1.448170731707317e-05,
|
| 813 |
+
"loss": 0.4122,
|
| 814 |
+
"step": 2425
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 14.939024390243903,
|
| 818 |
+
"grad_norm": 2.8491086959838867,
|
| 819 |
+
"learning_rate": 1.4058265582655828e-05,
|
| 820 |
+
"loss": 0.4083,
|
| 821 |
+
"step": 2450
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 15.0,
|
| 825 |
+
"eval_accuracy": 0.6898047722342733,
|
| 826 |
+
"eval_loss": 0.886920154094696,
|
| 827 |
+
"eval_runtime": 39.9853,
|
| 828 |
+
"eval_samples_per_second": 23.058,
|
| 829 |
+
"eval_steps_per_second": 0.725,
|
| 830 |
+
"step": 2460
|
| 831 |
+
},
|
| 832 |
+
{
|
| 833 |
+
"epoch": 15.091463414634147,
|
| 834 |
+
"grad_norm": 5.6329665184021,
|
| 835 |
+
"learning_rate": 1.3634823848238485e-05,
|
| 836 |
+
"loss": 0.393,
|
| 837 |
+
"step": 2475
|
| 838 |
+
},
|
| 839 |
+
{
|
| 840 |
+
"epoch": 15.24390243902439,
|
| 841 |
+
"grad_norm": 4.589029312133789,
|
| 842 |
+
"learning_rate": 1.321138211382114e-05,
|
| 843 |
+
"loss": 0.4068,
|
| 844 |
+
"step": 2500
|
| 845 |
+
},
|
| 846 |
+
{
|
| 847 |
+
"epoch": 15.396341463414634,
|
| 848 |
+
"grad_norm": 1.9374419450759888,
|
| 849 |
+
"learning_rate": 1.2787940379403795e-05,
|
| 850 |
+
"loss": 0.4077,
|
| 851 |
+
"step": 2525
|
| 852 |
+
},
|
| 853 |
+
{
|
| 854 |
+
"epoch": 15.548780487804878,
|
| 855 |
+
"grad_norm": 3.12813401222229,
|
| 856 |
+
"learning_rate": 1.2364498644986451e-05,
|
| 857 |
+
"loss": 0.3932,
|
| 858 |
+
"step": 2550
|
| 859 |
+
},
|
| 860 |
+
{
|
| 861 |
+
"epoch": 15.701219512195122,
|
| 862 |
+
"grad_norm": 4.135628700256348,
|
| 863 |
+
"learning_rate": 1.1941056910569107e-05,
|
| 864 |
+
"loss": 0.4048,
|
| 865 |
+
"step": 2575
|
| 866 |
+
},
|
| 867 |
+
{
|
| 868 |
+
"epoch": 15.853658536585366,
|
| 869 |
+
"grad_norm": 2.268238067626953,
|
| 870 |
+
"learning_rate": 1.1517615176151763e-05,
|
| 871 |
+
"loss": 0.4043,
|
| 872 |
+
"step": 2600
|
| 873 |
+
},
|
| 874 |
+
{
|
| 875 |
+
"epoch": 16.0,
|
| 876 |
+
"eval_accuracy": 0.6984815618221258,
|
| 877 |
+
"eval_loss": 0.8697348237037659,
|
| 878 |
+
"eval_runtime": 40.2898,
|
| 879 |
+
"eval_samples_per_second": 22.884,
|
| 880 |
+
"eval_steps_per_second": 0.72,
|
| 881 |
+
"step": 2624
|
| 882 |
+
},
|
| 883 |
+
{
|
| 884 |
+
"epoch": 16.00609756097561,
|
| 885 |
+
"grad_norm": 3.1583592891693115,
|
| 886 |
+
"learning_rate": 1.1094173441734418e-05,
|
| 887 |
+
"loss": 0.4083,
|
| 888 |
+
"step": 2625
|
| 889 |
+
},
|
| 890 |
+
{
|
| 891 |
+
"epoch": 16.158536585365855,
|
| 892 |
+
"grad_norm": 3.8587262630462646,
|
| 893 |
+
"learning_rate": 1.0670731707317074e-05,
|
| 894 |
+
"loss": 0.3955,
|
| 895 |
+
"step": 2650
|
| 896 |
+
},
|
| 897 |
+
{
|
| 898 |
+
"epoch": 16.3109756097561,
|
| 899 |
+
"grad_norm": 6.293887615203857,
|
| 900 |
+
"learning_rate": 1.0247289972899728e-05,
|
| 901 |
+
"loss": 0.3932,
|
| 902 |
+
"step": 2675
|
| 903 |
+
},
|
| 904 |
+
{
|
| 905 |
+
"epoch": 16.463414634146343,
|
| 906 |
+
"grad_norm": 4.263759613037109,
|
| 907 |
+
"learning_rate": 9.823848238482384e-06,
|
| 908 |
+
"loss": 0.3929,
|
| 909 |
+
"step": 2700
|
| 910 |
+
},
|
| 911 |
+
{
|
| 912 |
+
"epoch": 16.615853658536587,
|
| 913 |
+
"grad_norm": 2.30064058303833,
|
| 914 |
+
"learning_rate": 9.400406504065042e-06,
|
| 915 |
+
"loss": 0.4003,
|
| 916 |
+
"step": 2725
|
| 917 |
+
},
|
| 918 |
+
{
|
| 919 |
+
"epoch": 16.76829268292683,
|
| 920 |
+
"grad_norm": 4.408405303955078,
|
| 921 |
+
"learning_rate": 8.976964769647697e-06,
|
| 922 |
+
"loss": 0.3919,
|
| 923 |
+
"step": 2750
|
| 924 |
+
},
|
| 925 |
+
{
|
| 926 |
+
"epoch": 16.920731707317074,
|
| 927 |
+
"grad_norm": 4.191670894622803,
|
| 928 |
+
"learning_rate": 8.553523035230353e-06,
|
| 929 |
+
"loss": 0.4061,
|
| 930 |
+
"step": 2775
|
| 931 |
+
},
|
| 932 |
+
{
|
| 933 |
+
"epoch": 17.0,
|
| 934 |
+
"eval_accuracy": 0.7071583514099783,
|
| 935 |
+
"eval_loss": 0.8562042713165283,
|
| 936 |
+
"eval_runtime": 40.0276,
|
| 937 |
+
"eval_samples_per_second": 23.034,
|
| 938 |
+
"eval_steps_per_second": 0.724,
|
| 939 |
+
"step": 2788
|
| 940 |
+
},
|
| 941 |
+
{
|
| 942 |
+
"epoch": 17.073170731707318,
|
| 943 |
+
"grad_norm": 2.7028963565826416,
|
| 944 |
+
"learning_rate": 8.130081300813009e-06,
|
| 945 |
+
"loss": 0.3994,
|
| 946 |
+
"step": 2800
|
| 947 |
+
},
|
| 948 |
+
{
|
| 949 |
+
"epoch": 17.225609756097562,
|
| 950 |
+
"grad_norm": 2.178621530532837,
|
| 951 |
+
"learning_rate": 7.706639566395663e-06,
|
| 952 |
+
"loss": 0.3833,
|
| 953 |
+
"step": 2825
|
| 954 |
+
},
|
| 955 |
+
{
|
| 956 |
+
"epoch": 17.378048780487806,
|
| 957 |
+
"grad_norm": 1.921392560005188,
|
| 958 |
+
"learning_rate": 7.283197831978319e-06,
|
| 959 |
+
"loss": 0.4023,
|
| 960 |
+
"step": 2850
|
| 961 |
+
},
|
| 962 |
+
{
|
| 963 |
+
"epoch": 17.53048780487805,
|
| 964 |
+
"grad_norm": 3.9117579460144043,
|
| 965 |
+
"learning_rate": 6.859756097560977e-06,
|
| 966 |
+
"loss": 0.3991,
|
| 967 |
+
"step": 2875
|
| 968 |
+
},
|
| 969 |
+
{
|
| 970 |
+
"epoch": 17.682926829268293,
|
| 971 |
+
"grad_norm": 1.6043190956115723,
|
| 972 |
+
"learning_rate": 6.4363143631436324e-06,
|
| 973 |
+
"loss": 0.3924,
|
| 974 |
+
"step": 2900
|
| 975 |
+
},
|
| 976 |
+
{
|
| 977 |
+
"epoch": 17.835365853658537,
|
| 978 |
+
"grad_norm": 2.8831610679626465,
|
| 979 |
+
"learning_rate": 6.0128726287262874e-06,
|
| 980 |
+
"loss": 0.3837,
|
| 981 |
+
"step": 2925
|
| 982 |
+
},
|
| 983 |
+
{
|
| 984 |
+
"epoch": 17.98780487804878,
|
| 985 |
+
"grad_norm": 2.4830398559570312,
|
| 986 |
+
"learning_rate": 5.589430894308943e-06,
|
| 987 |
+
"loss": 0.389,
|
| 988 |
+
"step": 2950
|
| 989 |
+
},
|
| 990 |
+
{
|
| 991 |
+
"epoch": 18.0,
|
| 992 |
+
"eval_accuracy": 0.7082429501084598,
|
| 993 |
+
"eval_loss": 0.8595220446586609,
|
| 994 |
+
"eval_runtime": 41.6687,
|
| 995 |
+
"eval_samples_per_second": 22.127,
|
| 996 |
+
"eval_steps_per_second": 0.696,
|
| 997 |
+
"step": 2952
|
| 998 |
+
},
|
| 999 |
+
{
|
| 1000 |
+
"epoch": 18.140243902439025,
|
| 1001 |
+
"grad_norm": 1.9708536863327026,
|
| 1002 |
+
"learning_rate": 5.165989159891599e-06,
|
| 1003 |
+
"loss": 0.3896,
|
| 1004 |
+
"step": 2975
|
| 1005 |
+
},
|
| 1006 |
+
{
|
| 1007 |
+
"epoch": 18.29268292682927,
|
| 1008 |
+
"grad_norm": 4.3860673904418945,
|
| 1009 |
+
"learning_rate": 4.742547425474255e-06,
|
| 1010 |
+
"loss": 0.3939,
|
| 1011 |
+
"step": 3000
|
| 1012 |
+
},
|
| 1013 |
+
{
|
| 1014 |
+
"epoch": 18.445121951219512,
|
| 1015 |
+
"grad_norm": 2.5189573764801025,
|
| 1016 |
+
"learning_rate": 4.319105691056911e-06,
|
| 1017 |
+
"loss": 0.3908,
|
| 1018 |
+
"step": 3025
|
| 1019 |
+
},
|
| 1020 |
+
{
|
| 1021 |
+
"epoch": 18.597560975609756,
|
| 1022 |
+
"grad_norm": 1.8949400186538696,
|
| 1023 |
+
"learning_rate": 3.8956639566395666e-06,
|
| 1024 |
+
"loss": 0.3947,
|
| 1025 |
+
"step": 3050
|
| 1026 |
+
},
|
| 1027 |
+
{
|
| 1028 |
+
"epoch": 18.75,
|
| 1029 |
+
"grad_norm": 2.0496764183044434,
|
| 1030 |
+
"learning_rate": 3.4722222222222224e-06,
|
| 1031 |
+
"loss": 0.3837,
|
| 1032 |
+
"step": 3075
|
| 1033 |
+
},
|
| 1034 |
+
{
|
| 1035 |
+
"epoch": 18.902439024390244,
|
| 1036 |
+
"grad_norm": 3.2844817638397217,
|
| 1037 |
+
"learning_rate": 3.0487804878048782e-06,
|
| 1038 |
+
"loss": 0.3982,
|
| 1039 |
+
"step": 3100
|
| 1040 |
+
},
|
| 1041 |
+
{
|
| 1042 |
+
"epoch": 19.0,
|
| 1043 |
+
"eval_accuracy": 0.6984815618221258,
|
| 1044 |
+
"eval_loss": 0.8660802841186523,
|
| 1045 |
+
"eval_runtime": 41.7631,
|
| 1046 |
+
"eval_samples_per_second": 22.077,
|
| 1047 |
+
"eval_steps_per_second": 0.694,
|
| 1048 |
+
"step": 3116
|
| 1049 |
+
},
|
| 1050 |
+
{
|
| 1051 |
+
"epoch": 19.054878048780488,
|
| 1052 |
+
"grad_norm": 2.776163339614868,
|
| 1053 |
+
"learning_rate": 2.625338753387534e-06,
|
| 1054 |
+
"loss": 0.3863,
|
| 1055 |
+
"step": 3125
|
| 1056 |
+
},
|
| 1057 |
+
{
|
| 1058 |
+
"epoch": 19.20731707317073,
|
| 1059 |
+
"grad_norm": 2.2934467792510986,
|
| 1060 |
+
"learning_rate": 2.2018970189701894e-06,
|
| 1061 |
+
"loss": 0.394,
|
| 1062 |
+
"step": 3150
|
| 1063 |
+
},
|
| 1064 |
+
{
|
| 1065 |
+
"epoch": 19.359756097560975,
|
| 1066 |
+
"grad_norm": 1.4411579370498657,
|
| 1067 |
+
"learning_rate": 1.7784552845528455e-06,
|
| 1068 |
+
"loss": 0.3908,
|
| 1069 |
+
"step": 3175
|
| 1070 |
+
},
|
| 1071 |
+
{
|
| 1072 |
+
"epoch": 19.51219512195122,
|
| 1073 |
+
"grad_norm": 1.6821469068527222,
|
| 1074 |
+
"learning_rate": 1.3550135501355015e-06,
|
| 1075 |
+
"loss": 0.3844,
|
| 1076 |
+
"step": 3200
|
| 1077 |
+
},
|
| 1078 |
+
{
|
| 1079 |
+
"epoch": 19.664634146341463,
|
| 1080 |
+
"grad_norm": 3.2860209941864014,
|
| 1081 |
+
"learning_rate": 9.315718157181572e-07,
|
| 1082 |
+
"loss": 0.3879,
|
| 1083 |
+
"step": 3225
|
| 1084 |
+
},
|
| 1085 |
+
{
|
| 1086 |
+
"epoch": 19.817073170731707,
|
| 1087 |
+
"grad_norm": 4.093799591064453,
|
| 1088 |
+
"learning_rate": 5.081300813008131e-07,
|
| 1089 |
+
"loss": 0.3852,
|
| 1090 |
+
"step": 3250
|
| 1091 |
+
},
|
| 1092 |
+
{
|
| 1093 |
+
"epoch": 19.96951219512195,
|
| 1094 |
+
"grad_norm": 1.6917803287506104,
|
| 1095 |
+
"learning_rate": 8.468834688346885e-08,
|
| 1096 |
+
"loss": 0.3823,
|
| 1097 |
+
"step": 3275
|
| 1098 |
+
},
|
| 1099 |
+
{
|
| 1100 |
+
"epoch": 20.0,
|
| 1101 |
+
"eval_accuracy": 0.7114967462039046,
|
| 1102 |
+
"eval_loss": 0.8688989281654358,
|
| 1103 |
+
"eval_runtime": 41.6761,
|
| 1104 |
+
"eval_samples_per_second": 22.123,
|
| 1105 |
+
"eval_steps_per_second": 0.696,
|
| 1106 |
+
"step": 3280
|
| 1107 |
+
},
|
| 1108 |
+
{
|
| 1109 |
+
"epoch": 20.0,
|
| 1110 |
+
"step": 3280,
|
| 1111 |
+
"total_flos": 0.0,
|
| 1112 |
+
"train_loss": 0.5078278935537106,
|
| 1113 |
+
"train_runtime": 14435.8806,
|
| 1114 |
+
"train_samples_per_second": 7.238,
|
| 1115 |
+
"train_steps_per_second": 0.227
|
| 1116 |
+
}
|
| 1117 |
+
],
|
| 1118 |
+
"logging_steps": 25,
|
| 1119 |
+
"max_steps": 3280,
|
| 1120 |
+
"num_input_tokens_seen": 0,
|
| 1121 |
+
"num_train_epochs": 20,
|
| 1122 |
+
"save_steps": 500,
|
| 1123 |
+
"stateful_callbacks": {
|
| 1124 |
+
"TrainerControl": {
|
| 1125 |
+
"args": {
|
| 1126 |
+
"should_epoch_stop": false,
|
| 1127 |
+
"should_evaluate": false,
|
| 1128 |
+
"should_log": false,
|
| 1129 |
+
"should_save": true,
|
| 1130 |
+
"should_training_stop": true
|
| 1131 |
+
},
|
| 1132 |
+
"attributes": {}
|
| 1133 |
+
}
|
| 1134 |
+
},
|
| 1135 |
+
"total_flos": 0.0,
|
| 1136 |
+
"train_batch_size": 32,
|
| 1137 |
+
"trial_name": null,
|
| 1138 |
+
"trial_params": null
|
| 1139 |
+
}
|