End of training
Browse files- README.md +2 -1
- all_results.json +8 -0
- train_results.json +8 -0
- trainer_state.json +1953 -0
- training_loss.png +0 -0
README.md
CHANGED
|
@@ -4,6 +4,7 @@ license: apache-2.0
|
|
| 4 |
base_model: Qwen/Qwen2.5-32B
|
| 5 |
tags:
|
| 6 |
- llama-factory
|
|
|
|
| 7 |
- generated_from_trainer
|
| 8 |
model-index:
|
| 9 |
- name: pretrain-20250131
|
|
@@ -15,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 15 |
|
| 16 |
# pretrain-20250131
|
| 17 |
|
| 18 |
-
This model is a fine-tuned version of [Qwen/Qwen2.5-32B](https://huggingface.co/Qwen/Qwen2.5-32B) on
|
| 19 |
|
| 20 |
## Model description
|
| 21 |
|
|
|
|
| 4 |
base_model: Qwen/Qwen2.5-32B
|
| 5 |
tags:
|
| 6 |
- llama-factory
|
| 7 |
+
- full
|
| 8 |
- generated_from_trainer
|
| 9 |
model-index:
|
| 10 |
- name: pretrain-20250131
|
|
|
|
| 16 |
|
| 17 |
# pretrain-20250131
|
| 18 |
|
| 19 |
+
This model is a fine-tuned version of [Qwen/Qwen2.5-32B](https://huggingface.co/Qwen/Qwen2.5-32B) on the openalex_references dataset.
|
| 20 |
|
| 21 |
## Model description
|
| 22 |
|
all_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 0.9986282578875172,
|
| 3 |
+
"total_flos": 1921476294868992.0,
|
| 4 |
+
"train_loss": 1.1405032803287436,
|
| 5 |
+
"train_runtime": 59872.6859,
|
| 6 |
+
"train_samples_per_second": 0.584,
|
| 7 |
+
"train_steps_per_second": 0.005
|
| 8 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 0.9986282578875172,
|
| 3 |
+
"total_flos": 1921476294868992.0,
|
| 4 |
+
"train_loss": 1.1405032803287436,
|
| 5 |
+
"train_runtime": 59872.6859,
|
| 6 |
+
"train_samples_per_second": 0.584,
|
| 7 |
+
"train_steps_per_second": 0.005
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1953 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.9986282578875172,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 273,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.003657978966620942,
|
| 13 |
+
"grad_norm": 0.11399450898170471,
|
| 14 |
+
"learning_rate": 3.5714285714285714e-06,
|
| 15 |
+
"loss": 1.3486,
|
| 16 |
+
"step": 1
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.007315957933241884,
|
| 20 |
+
"grad_norm": 0.11220081150531769,
|
| 21 |
+
"learning_rate": 7.142857142857143e-06,
|
| 22 |
+
"loss": 1.359,
|
| 23 |
+
"step": 2
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.010973936899862825,
|
| 27 |
+
"grad_norm": 0.11029545217752457,
|
| 28 |
+
"learning_rate": 1.0714285714285714e-05,
|
| 29 |
+
"loss": 1.3603,
|
| 30 |
+
"step": 3
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.014631915866483767,
|
| 34 |
+
"grad_norm": 0.09306512773036957,
|
| 35 |
+
"learning_rate": 1.4285714285714285e-05,
|
| 36 |
+
"loss": 1.352,
|
| 37 |
+
"step": 4
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.01828989483310471,
|
| 41 |
+
"grad_norm": 0.23787762224674225,
|
| 42 |
+
"learning_rate": 1.785714285714286e-05,
|
| 43 |
+
"loss": 1.3645,
|
| 44 |
+
"step": 5
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.02194787379972565,
|
| 48 |
+
"grad_norm": 0.07387927174568176,
|
| 49 |
+
"learning_rate": 2.1428571428571428e-05,
|
| 50 |
+
"loss": 1.3406,
|
| 51 |
+
"step": 6
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.025605852766346592,
|
| 55 |
+
"grad_norm": 0.08182856440544128,
|
| 56 |
+
"learning_rate": 2.5e-05,
|
| 57 |
+
"loss": 1.3445,
|
| 58 |
+
"step": 7
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.029263831732967534,
|
| 62 |
+
"grad_norm": 0.07252974063158035,
|
| 63 |
+
"learning_rate": 2.857142857142857e-05,
|
| 64 |
+
"loss": 1.3225,
|
| 65 |
+
"step": 8
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.03292181069958848,
|
| 69 |
+
"grad_norm": 0.07590305060148239,
|
| 70 |
+
"learning_rate": 3.2142857142857144e-05,
|
| 71 |
+
"loss": 1.3063,
|
| 72 |
+
"step": 9
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.03657978966620942,
|
| 76 |
+
"grad_norm": 0.10524249821901321,
|
| 77 |
+
"learning_rate": 3.571428571428572e-05,
|
| 78 |
+
"loss": 1.3201,
|
| 79 |
+
"step": 10
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.040237768632830365,
|
| 83 |
+
"grad_norm": 0.04849069193005562,
|
| 84 |
+
"learning_rate": 3.928571428571429e-05,
|
| 85 |
+
"loss": 1.324,
|
| 86 |
+
"step": 11
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.0438957475994513,
|
| 90 |
+
"grad_norm": 0.08398816734552383,
|
| 91 |
+
"learning_rate": 4.2857142857142856e-05,
|
| 92 |
+
"loss": 1.324,
|
| 93 |
+
"step": 12
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.04755372656607224,
|
| 97 |
+
"grad_norm": 0.06089089438319206,
|
| 98 |
+
"learning_rate": 4.642857142857143e-05,
|
| 99 |
+
"loss": 1.3072,
|
| 100 |
+
"step": 13
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.051211705532693184,
|
| 104 |
+
"grad_norm": 0.07415427267551422,
|
| 105 |
+
"learning_rate": 5e-05,
|
| 106 |
+
"loss": 1.3005,
|
| 107 |
+
"step": 14
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.05486968449931413,
|
| 111 |
+
"grad_norm": 0.07027556747198105,
|
| 112 |
+
"learning_rate": 4.999816090185888e-05,
|
| 113 |
+
"loss": 1.3036,
|
| 114 |
+
"step": 15
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.05852766346593507,
|
| 118 |
+
"grad_norm": 0.05990540236234665,
|
| 119 |
+
"learning_rate": 4.9992643878018054e-05,
|
| 120 |
+
"loss": 1.3188,
|
| 121 |
+
"step": 16
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.06218564243255601,
|
| 125 |
+
"grad_norm": 0.07783979922533035,
|
| 126 |
+
"learning_rate": 4.99834497401854e-05,
|
| 127 |
+
"loss": 1.2834,
|
| 128 |
+
"step": 17
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.06584362139917696,
|
| 132 |
+
"grad_norm": 0.05285879224538803,
|
| 133 |
+
"learning_rate": 4.997057984107466e-05,
|
| 134 |
+
"loss": 1.287,
|
| 135 |
+
"step": 18
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.0695016003657979,
|
| 139 |
+
"grad_norm": 0.06284961104393005,
|
| 140 |
+
"learning_rate": 4.995403607420644e-05,
|
| 141 |
+
"loss": 1.2819,
|
| 142 |
+
"step": 19
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.07315957933241884,
|
| 146 |
+
"grad_norm": 0.0440058633685112,
|
| 147 |
+
"learning_rate": 4.993382087362959e-05,
|
| 148 |
+
"loss": 1.2616,
|
| 149 |
+
"step": 20
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.07681755829903979,
|
| 153 |
+
"grad_norm": 0.060883328318595886,
|
| 154 |
+
"learning_rate": 4.9909937213563165e-05,
|
| 155 |
+
"loss": 1.276,
|
| 156 |
+
"step": 21
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.08047553726566073,
|
| 160 |
+
"grad_norm": 0.042049508541822433,
|
| 161 |
+
"learning_rate": 4.988238860795873e-05,
|
| 162 |
+
"loss": 1.2654,
|
| 163 |
+
"step": 22
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.08413351623228166,
|
| 167 |
+
"grad_norm": 0.07169731706380844,
|
| 168 |
+
"learning_rate": 4.985117910998345e-05,
|
| 169 |
+
"loss": 1.2802,
|
| 170 |
+
"step": 23
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.0877914951989026,
|
| 174 |
+
"grad_norm": 0.04764694347977638,
|
| 175 |
+
"learning_rate": 4.981631331142368e-05,
|
| 176 |
+
"loss": 1.267,
|
| 177 |
+
"step": 24
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.09144947416552354,
|
| 181 |
+
"grad_norm": 0.05187158286571503,
|
| 182 |
+
"learning_rate": 4.977779634200947e-05,
|
| 183 |
+
"loss": 1.2659,
|
| 184 |
+
"step": 25
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.09510745313214448,
|
| 188 |
+
"grad_norm": 0.0522884801030159,
|
| 189 |
+
"learning_rate": 4.973563386865974e-05,
|
| 190 |
+
"loss": 1.2462,
|
| 191 |
+
"step": 26
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.09876543209876543,
|
| 195 |
+
"grad_norm": 0.03765333071351051,
|
| 196 |
+
"learning_rate": 4.968983209464863e-05,
|
| 197 |
+
"loss": 1.2546,
|
| 198 |
+
"step": 27
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.10242341106538637,
|
| 202 |
+
"grad_norm": 0.05279753729701042,
|
| 203 |
+
"learning_rate": 4.9640397758692715e-05,
|
| 204 |
+
"loss": 1.2446,
|
| 205 |
+
"step": 28
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.10608139003200731,
|
| 209 |
+
"grad_norm": 0.038350097835063934,
|
| 210 |
+
"learning_rate": 4.9587338133959625e-05,
|
| 211 |
+
"loss": 1.2615,
|
| 212 |
+
"step": 29
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.10973936899862825,
|
| 216 |
+
"grad_norm": 0.05348701775074005,
|
| 217 |
+
"learning_rate": 4.953066102699795e-05,
|
| 218 |
+
"loss": 1.275,
|
| 219 |
+
"step": 30
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.1133973479652492,
|
| 223 |
+
"grad_norm": 0.0379619225859642,
|
| 224 |
+
"learning_rate": 4.947037477658865e-05,
|
| 225 |
+
"loss": 1.2436,
|
| 226 |
+
"step": 31
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.11705532693187014,
|
| 230 |
+
"grad_norm": 0.0401979424059391,
|
| 231 |
+
"learning_rate": 4.9406488252518203e-05,
|
| 232 |
+
"loss": 1.2502,
|
| 233 |
+
"step": 32
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.12071330589849108,
|
| 237 |
+
"grad_norm": 0.035914115607738495,
|
| 238 |
+
"learning_rate": 4.933901085427362e-05,
|
| 239 |
+
"loss": 1.2489,
|
| 240 |
+
"step": 33
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.12437128486511202,
|
| 244 |
+
"grad_norm": 0.040159013122320175,
|
| 245 |
+
"learning_rate": 4.926795250965952e-05,
|
| 246 |
+
"loss": 1.233,
|
| 247 |
+
"step": 34
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.12802926383173296,
|
| 251 |
+
"grad_norm": 0.035101838409900665,
|
| 252 |
+
"learning_rate": 4.9193323673337476e-05,
|
| 253 |
+
"loss": 1.2434,
|
| 254 |
+
"step": 35
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.13168724279835392,
|
| 258 |
+
"grad_norm": 0.03814801946282387,
|
| 259 |
+
"learning_rate": 4.91151353252878e-05,
|
| 260 |
+
"loss": 1.2406,
|
| 261 |
+
"step": 36
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.13534522176497485,
|
| 265 |
+
"grad_norm": 0.03351541608572006,
|
| 266 |
+
"learning_rate": 4.9033398969194145e-05,
|
| 267 |
+
"loss": 1.2213,
|
| 268 |
+
"step": 37
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.1390032007315958,
|
| 272 |
+
"grad_norm": 0.036357831209897995,
|
| 273 |
+
"learning_rate": 4.894812663075095e-05,
|
| 274 |
+
"loss": 1.23,
|
| 275 |
+
"step": 38
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.14266117969821673,
|
| 279 |
+
"grad_norm": 0.03457951918244362,
|
| 280 |
+
"learning_rate": 4.8859330855894156e-05,
|
| 281 |
+
"loss": 1.2455,
|
| 282 |
+
"step": 39
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.1463191586648377,
|
| 286 |
+
"grad_norm": 0.04117417335510254,
|
| 287 |
+
"learning_rate": 4.876702470895531e-05,
|
| 288 |
+
"loss": 1.2272,
|
| 289 |
+
"step": 40
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.14997713763145862,
|
| 293 |
+
"grad_norm": 0.03363555297255516,
|
| 294 |
+
"learning_rate": 4.867122177073948e-05,
|
| 295 |
+
"loss": 1.2272,
|
| 296 |
+
"step": 41
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.15363511659807957,
|
| 300 |
+
"grad_norm": 0.04657221958041191,
|
| 301 |
+
"learning_rate": 4.857193613652711e-05,
|
| 302 |
+
"loss": 1.2417,
|
| 303 |
+
"step": 42
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.1572930955647005,
|
| 307 |
+
"grad_norm": 0.03347886726260185,
|
| 308 |
+
"learning_rate": 4.846918241400022e-05,
|
| 309 |
+
"loss": 1.2409,
|
| 310 |
+
"step": 43
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.16095107453132146,
|
| 314 |
+
"grad_norm": 0.04122474789619446,
|
| 315 |
+
"learning_rate": 4.836297572109323e-05,
|
| 316 |
+
"loss": 1.2237,
|
| 317 |
+
"step": 44
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.1646090534979424,
|
| 321 |
+
"grad_norm": 0.03406165540218353,
|
| 322 |
+
"learning_rate": 4.825333168376864e-05,
|
| 323 |
+
"loss": 1.2558,
|
| 324 |
+
"step": 45
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.16826703246456332,
|
| 328 |
+
"grad_norm": 0.037185415625572205,
|
| 329 |
+
"learning_rate": 4.8140266433718095e-05,
|
| 330 |
+
"loss": 1.2149,
|
| 331 |
+
"step": 46
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.17192501143118427,
|
| 335 |
+
"grad_norm": 0.02867889776825905,
|
| 336 |
+
"learning_rate": 4.8023796605988866e-05,
|
| 337 |
+
"loss": 1.1882,
|
| 338 |
+
"step": 47
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.1755829903978052,
|
| 342 |
+
"grad_norm": 0.03610045835375786,
|
| 343 |
+
"learning_rate": 4.790393933653646e-05,
|
| 344 |
+
"loss": 1.2114,
|
| 345 |
+
"step": 48
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.17924096936442616,
|
| 349 |
+
"grad_norm": 0.034297309815883636,
|
| 350 |
+
"learning_rate": 4.77807122597034e-05,
|
| 351 |
+
"loss": 1.1857,
|
| 352 |
+
"step": 49
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.18289894833104708,
|
| 356 |
+
"grad_norm": 0.03260922431945801,
|
| 357 |
+
"learning_rate": 4.7654133505624695e-05,
|
| 358 |
+
"loss": 1.2132,
|
| 359 |
+
"step": 50
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.18655692729766804,
|
| 363 |
+
"grad_norm": 0.03398128226399422,
|
| 364 |
+
"learning_rate": 4.752422169756048e-05,
|
| 365 |
+
"loss": 1.2046,
|
| 366 |
+
"step": 51
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.19021490626428897,
|
| 370 |
+
"grad_norm": 0.035840485244989395,
|
| 371 |
+
"learning_rate": 4.7390995949155915e-05,
|
| 372 |
+
"loss": 1.2374,
|
| 373 |
+
"step": 52
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.19387288523090992,
|
| 377 |
+
"grad_norm": 0.030534490942955017,
|
| 378 |
+
"learning_rate": 4.725447586162911e-05,
|
| 379 |
+
"loss": 1.2076,
|
| 380 |
+
"step": 53
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.19753086419753085,
|
| 384 |
+
"grad_norm": 0.03663424402475357,
|
| 385 |
+
"learning_rate": 4.711468152088719e-05,
|
| 386 |
+
"loss": 1.2239,
|
| 387 |
+
"step": 54
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.2011888431641518,
|
| 391 |
+
"grad_norm": 0.03044678270816803,
|
| 392 |
+
"learning_rate": 4.697163349457115e-05,
|
| 393 |
+
"loss": 1.2117,
|
| 394 |
+
"step": 55
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.20484682213077274,
|
| 398 |
+
"grad_norm": 0.030734116211533546,
|
| 399 |
+
"learning_rate": 4.6825352829029705e-05,
|
| 400 |
+
"loss": 1.1859,
|
| 401 |
+
"step": 56
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.2085048010973937,
|
| 405 |
+
"grad_norm": 0.03116767294704914,
|
| 406 |
+
"learning_rate": 4.667586104622288e-05,
|
| 407 |
+
"loss": 1.2065,
|
| 408 |
+
"step": 57
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.21216278006401462,
|
| 412 |
+
"grad_norm": 0.032509610056877136,
|
| 413 |
+
"learning_rate": 4.652318014055547e-05,
|
| 414 |
+
"loss": 1.2196,
|
| 415 |
+
"step": 58
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.21582075903063558,
|
| 419 |
+
"grad_norm": 0.030822426080703735,
|
| 420 |
+
"learning_rate": 4.636733257564104e-05,
|
| 421 |
+
"loss": 1.204,
|
| 422 |
+
"step": 59
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.2194787379972565,
|
| 426 |
+
"grad_norm": 0.03399716317653656,
|
| 427 |
+
"learning_rate": 4.620834128099696e-05,
|
| 428 |
+
"loss": 1.1865,
|
| 429 |
+
"step": 60
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.22313671696387746,
|
| 433 |
+
"grad_norm": 0.029582664370536804,
|
| 434 |
+
"learning_rate": 4.604622964867078e-05,
|
| 435 |
+
"loss": 1.195,
|
| 436 |
+
"step": 61
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.2267946959304984,
|
| 440 |
+
"grad_norm": 0.03362514078617096,
|
| 441 |
+
"learning_rate": 4.588102152979863e-05,
|
| 442 |
+
"loss": 1.2179,
|
| 443 |
+
"step": 62
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.23045267489711935,
|
| 447 |
+
"grad_norm": 0.03153723105788231,
|
| 448 |
+
"learning_rate": 4.571274123109606e-05,
|
| 449 |
+
"loss": 1.183,
|
| 450 |
+
"step": 63
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.23411065386374028,
|
| 454 |
+
"grad_norm": 0.03338644653558731,
|
| 455 |
+
"learning_rate": 4.5541413511281826e-05,
|
| 456 |
+
"loss": 1.2011,
|
| 457 |
+
"step": 64
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.23776863283036123,
|
| 461 |
+
"grad_norm": 0.02905767410993576,
|
| 462 |
+
"learning_rate": 4.536706357743522e-05,
|
| 463 |
+
"loss": 1.2092,
|
| 464 |
+
"step": 65
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.24142661179698216,
|
| 468 |
+
"grad_norm": 0.03598478436470032,
|
| 469 |
+
"learning_rate": 4.518971708128737e-05,
|
| 470 |
+
"loss": 1.1903,
|
| 471 |
+
"step": 66
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.24508459076360312,
|
| 475 |
+
"grad_norm": 0.03393587842583656,
|
| 476 |
+
"learning_rate": 4.5009400115447206e-05,
|
| 477 |
+
"loss": 1.1801,
|
| 478 |
+
"step": 67
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.24874256973022404,
|
| 482 |
+
"grad_norm": 0.03872101381421089,
|
| 483 |
+
"learning_rate": 4.4826139209562444e-05,
|
| 484 |
+
"loss": 1.1868,
|
| 485 |
+
"step": 68
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.252400548696845,
|
| 489 |
+
"grad_norm": 0.03533606603741646,
|
| 490 |
+
"learning_rate": 4.463996132641641e-05,
|
| 491 |
+
"loss": 1.1879,
|
| 492 |
+
"step": 69
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 0.25605852766346593,
|
| 496 |
+
"grad_norm": 0.03268212825059891,
|
| 497 |
+
"learning_rate": 4.445089385796099e-05,
|
| 498 |
+
"loss": 1.1986,
|
| 499 |
+
"step": 70
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.2597165066300869,
|
| 503 |
+
"grad_norm": 0.03353170305490494,
|
| 504 |
+
"learning_rate": 4.4258964621286584e-05,
|
| 505 |
+
"loss": 1.1835,
|
| 506 |
+
"step": 71
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.26337448559670784,
|
| 510 |
+
"grad_norm": 0.03047579899430275,
|
| 511 |
+
"learning_rate": 4.4064201854529365e-05,
|
| 512 |
+
"loss": 1.1773,
|
| 513 |
+
"step": 72
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.26703246456332874,
|
| 517 |
+
"grad_norm": 0.034304771572351456,
|
| 518 |
+
"learning_rate": 4.386663421271674e-05,
|
| 519 |
+
"loss": 1.1814,
|
| 520 |
+
"step": 73
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.2706904435299497,
|
| 524 |
+
"grad_norm": 0.03483429551124573,
|
| 525 |
+
"learning_rate": 4.366629076355131e-05,
|
| 526 |
+
"loss": 1.1928,
|
| 527 |
+
"step": 74
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.27434842249657065,
|
| 531 |
+
"grad_norm": 0.027306433767080307,
|
| 532 |
+
"learning_rate": 4.346320098313429e-05,
|
| 533 |
+
"loss": 1.1426,
|
| 534 |
+
"step": 75
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.2780064014631916,
|
| 538 |
+
"grad_norm": 0.03144339472055435,
|
| 539 |
+
"learning_rate": 4.3257394751628686e-05,
|
| 540 |
+
"loss": 1.1639,
|
| 541 |
+
"step": 76
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.2816643804298125,
|
| 545 |
+
"grad_norm": 0.033143457025289536,
|
| 546 |
+
"learning_rate": 4.3048902348863116e-05,
|
| 547 |
+
"loss": 1.1879,
|
| 548 |
+
"step": 77
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.28532235939643347,
|
| 552 |
+
"grad_norm": 0.032496023923158646,
|
| 553 |
+
"learning_rate": 4.283775444987681e-05,
|
| 554 |
+
"loss": 1.218,
|
| 555 |
+
"step": 78
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.2889803383630544,
|
| 559 |
+
"grad_norm": 0.03808918222784996,
|
| 560 |
+
"learning_rate": 4.262398212040646e-05,
|
| 561 |
+
"loss": 1.1989,
|
| 562 |
+
"step": 79
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 0.2926383173296754,
|
| 566 |
+
"grad_norm": 0.036832649260759354,
|
| 567 |
+
"learning_rate": 4.2407616812315554e-05,
|
| 568 |
+
"loss": 1.1594,
|
| 569 |
+
"step": 80
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.2962962962962963,
|
| 573 |
+
"grad_norm": 0.02746155858039856,
|
| 574 |
+
"learning_rate": 4.218869035896697e-05,
|
| 575 |
+
"loss": 1.1876,
|
| 576 |
+
"step": 81
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 0.29995427526291724,
|
| 580 |
+
"grad_norm": 0.04084876552224159,
|
| 581 |
+
"learning_rate": 4.1967234970539384e-05,
|
| 582 |
+
"loss": 1.1746,
|
| 583 |
+
"step": 82
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 0.3036122542295382,
|
| 587 |
+
"grad_norm": 0.030713148415088654,
|
| 588 |
+
"learning_rate": 4.1743283229288245e-05,
|
| 589 |
+
"loss": 1.179,
|
| 590 |
+
"step": 83
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.30727023319615915,
|
| 594 |
+
"grad_norm": 0.034150052815675735,
|
| 595 |
+
"learning_rate": 4.151686808475204e-05,
|
| 596 |
+
"loss": 1.1407,
|
| 597 |
+
"step": 84
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 0.31092821216278005,
|
| 601 |
+
"grad_norm": 0.035240545868873596,
|
| 602 |
+
"learning_rate": 4.1288022848904485e-05,
|
| 603 |
+
"loss": 1.1567,
|
| 604 |
+
"step": 85
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 0.314586191129401,
|
| 608 |
+
"grad_norm": 0.03663535416126251,
|
| 609 |
+
"learning_rate": 4.10567811912534e-05,
|
| 610 |
+
"loss": 1.1337,
|
| 611 |
+
"step": 86
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 0.31824417009602196,
|
| 615 |
+
"grad_norm": 0.03188696503639221,
|
| 616 |
+
"learning_rate": 4.082317713388702e-05,
|
| 617 |
+
"loss": 1.1751,
|
| 618 |
+
"step": 87
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 0.3219021490626429,
|
| 622 |
+
"grad_norm": 0.03437764570116997,
|
| 623 |
+
"learning_rate": 4.058724504646834e-05,
|
| 624 |
+
"loss": 1.1659,
|
| 625 |
+
"step": 88
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.3255601280292638,
|
| 629 |
+
"grad_norm": 0.036784157156944275,
|
| 630 |
+
"learning_rate": 4.0349019641178443e-05,
|
| 631 |
+
"loss": 1.1451,
|
| 632 |
+
"step": 89
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.3292181069958848,
|
| 636 |
+
"grad_norm": 0.03242890164256096,
|
| 637 |
+
"learning_rate": 4.010853596760933e-05,
|
| 638 |
+
"loss": 1.1479,
|
| 639 |
+
"step": 90
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.33287608596250573,
|
| 643 |
+
"grad_norm": 0.030873438343405724,
|
| 644 |
+
"learning_rate": 3.986582940760717e-05,
|
| 645 |
+
"loss": 1.1453,
|
| 646 |
+
"step": 91
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 0.33653406492912663,
|
| 650 |
+
"grad_norm": 0.036423344165086746,
|
| 651 |
+
"learning_rate": 3.962093567006662e-05,
|
| 652 |
+
"loss": 1.1505,
|
| 653 |
+
"step": 92
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 0.3401920438957476,
|
| 657 |
+
"grad_norm": 0.029906025156378746,
|
| 658 |
+
"learning_rate": 3.937389078567708e-05,
|
| 659 |
+
"loss": 1.1718,
|
| 660 |
+
"step": 93
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 0.34385002286236854,
|
| 664 |
+
"grad_norm": 0.03078794851899147,
|
| 665 |
+
"learning_rate": 3.912473110162156e-05,
|
| 666 |
+
"loss": 1.1562,
|
| 667 |
+
"step": 94
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 0.3475080018289895,
|
| 671 |
+
"grad_norm": 0.029547788202762604,
|
| 672 |
+
"learning_rate": 3.887349327622901e-05,
|
| 673 |
+
"loss": 1.1549,
|
| 674 |
+
"step": 95
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 0.3511659807956104,
|
| 678 |
+
"grad_norm": 0.03439987823367119,
|
| 679 |
+
"learning_rate": 3.8620214273580846e-05,
|
| 680 |
+
"loss": 1.1736,
|
| 681 |
+
"step": 96
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 0.35482395976223136,
|
| 685 |
+
"grad_norm": 0.03436443582177162,
|
| 686 |
+
"learning_rate": 3.83649313580725e-05,
|
| 687 |
+
"loss": 1.1466,
|
| 688 |
+
"step": 97
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 0.3584819387288523,
|
| 692 |
+
"grad_norm": 0.03227129206061363,
|
| 693 |
+
"learning_rate": 3.8107682088930794e-05,
|
| 694 |
+
"loss": 1.1491,
|
| 695 |
+
"step": 98
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 0.36213991769547327,
|
| 699 |
+
"grad_norm": 0.029436925426125526,
|
| 700 |
+
"learning_rate": 3.784850431468795e-05,
|
| 701 |
+
"loss": 1.1497,
|
| 702 |
+
"step": 99
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 0.36579789666209417,
|
| 706 |
+
"grad_norm": 0.030241703614592552,
|
| 707 |
+
"learning_rate": 3.7587436167613e-05,
|
| 708 |
+
"loss": 1.1434,
|
| 709 |
+
"step": 100
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 0.3694558756287151,
|
| 713 |
+
"grad_norm": 0.030009973794221878,
|
| 714 |
+
"learning_rate": 3.7324516058101454e-05,
|
| 715 |
+
"loss": 1.129,
|
| 716 |
+
"step": 101
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 0.3731138545953361,
|
| 720 |
+
"grad_norm": 0.026958679780364037,
|
| 721 |
+
"learning_rate": 3.705978266902409e-05,
|
| 722 |
+
"loss": 1.15,
|
| 723 |
+
"step": 102
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 0.37677183356195704,
|
| 727 |
+
"grad_norm": 0.02721358649432659,
|
| 728 |
+
"learning_rate": 3.679327495003561e-05,
|
| 729 |
+
"loss": 1.137,
|
| 730 |
+
"step": 103
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 0.38042981252857794,
|
| 734 |
+
"grad_norm": 0.02925017662346363,
|
| 735 |
+
"learning_rate": 3.6525032111844054e-05,
|
| 736 |
+
"loss": 1.1456,
|
| 737 |
+
"step": 104
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 0.3840877914951989,
|
| 741 |
+
"grad_norm": 0.026657432317733765,
|
| 742 |
+
"learning_rate": 3.6255093620441834e-05,
|
| 743 |
+
"loss": 1.1339,
|
| 744 |
+
"step": 105
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 0.38774577046181985,
|
| 748 |
+
"grad_norm": 0.02823042869567871,
|
| 749 |
+
"learning_rate": 3.598349919129917e-05,
|
| 750 |
+
"loss": 1.1406,
|
| 751 |
+
"step": 106
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 0.3914037494284408,
|
| 755 |
+
"grad_norm": 0.031808383762836456,
|
| 756 |
+
"learning_rate": 3.571028878352084e-05,
|
| 757 |
+
"loss": 1.1561,
|
| 758 |
+
"step": 107
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 0.3950617283950617,
|
| 762 |
+
"grad_norm": 0.02964746206998825,
|
| 763 |
+
"learning_rate": 3.5435502593967104e-05,
|
| 764 |
+
"loss": 1.1377,
|
| 765 |
+
"step": 108
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 0.39871970736168266,
|
| 769 |
+
"grad_norm": 0.03088381327688694,
|
| 770 |
+
"learning_rate": 3.515918105133958e-05,
|
| 771 |
+
"loss": 1.1499,
|
| 772 |
+
"step": 109
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 0.4023776863283036,
|
| 776 |
+
"grad_norm": 0.04066149517893791,
|
| 777 |
+
"learning_rate": 3.4881364810233095e-05,
|
| 778 |
+
"loss": 1.1453,
|
| 779 |
+
"step": 110
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 0.4060356652949246,
|
| 783 |
+
"grad_norm": 0.03428781405091286,
|
| 784 |
+
"learning_rate": 3.460209474515428e-05,
|
| 785 |
+
"loss": 1.1363,
|
| 786 |
+
"step": 111
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 0.4096936442615455,
|
| 790 |
+
"grad_norm": 0.03334755077958107,
|
| 791 |
+
"learning_rate": 3.432141194450772e-05,
|
| 792 |
+
"loss": 1.1414,
|
| 793 |
+
"step": 112
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 0.41335162322816643,
|
| 797 |
+
"grad_norm": 0.03784303367137909,
|
| 798 |
+
"learning_rate": 3.403935770455077e-05,
|
| 799 |
+
"loss": 1.1555,
|
| 800 |
+
"step": 113
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 0.4170096021947874,
|
| 804 |
+
"grad_norm": 0.030056986957788467,
|
| 805 |
+
"learning_rate": 3.375597352331772e-05,
|
| 806 |
+
"loss": 1.1272,
|
| 807 |
+
"step": 114
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 0.42066758116140834,
|
| 811 |
+
"grad_norm": 0.030737608671188354,
|
| 812 |
+
"learning_rate": 3.347130109451422e-05,
|
| 813 |
+
"loss": 1.1271,
|
| 814 |
+
"step": 115
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 0.42432556012802924,
|
| 818 |
+
"grad_norm": 0.030851833522319794,
|
| 819 |
+
"learning_rate": 3.3185382301383064e-05,
|
| 820 |
+
"loss": 1.1368,
|
| 821 |
+
"step": 116
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 0.4279835390946502,
|
| 825 |
+
"grad_norm": 0.027810659259557724,
|
| 826 |
+
"learning_rate": 3.28982592105419e-05,
|
| 827 |
+
"loss": 1.1339,
|
| 828 |
+
"step": 117
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 0.43164151806127116,
|
| 832 |
+
"grad_norm": 0.030422871932387352,
|
| 833 |
+
"learning_rate": 3.260997406579417e-05,
|
| 834 |
+
"loss": 1.1104,
|
| 835 |
+
"step": 118
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 0.4352994970278921,
|
| 839 |
+
"grad_norm": 0.029860053211450577,
|
| 840 |
+
"learning_rate": 3.232056928191376e-05,
|
| 841 |
+
"loss": 1.1138,
|
| 842 |
+
"step": 119
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 0.438957475994513,
|
| 846 |
+
"grad_norm": 0.031170522794127464,
|
| 847 |
+
"learning_rate": 3.2030087438404685e-05,
|
| 848 |
+
"loss": 1.1492,
|
| 849 |
+
"step": 120
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 0.44261545496113397,
|
| 853 |
+
"grad_norm": 0.027890468016266823,
|
| 854 |
+
"learning_rate": 3.173857127323642e-05,
|
| 855 |
+
"loss": 1.1154,
|
| 856 |
+
"step": 121
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 0.4462734339277549,
|
| 860 |
+
"grad_norm": 0.02984224259853363,
|
| 861 |
+
"learning_rate": 3.144606367655595e-05,
|
| 862 |
+
"loss": 1.1299,
|
| 863 |
+
"step": 122
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 0.4499314128943759,
|
| 867 |
+
"grad_norm": 0.029711229726672173,
|
| 868 |
+
"learning_rate": 3.115260768437747e-05,
|
| 869 |
+
"loss": 1.1251,
|
| 870 |
+
"step": 123
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 0.4535893918609968,
|
| 874 |
+
"grad_norm": 0.02858361415565014,
|
| 875 |
+
"learning_rate": 3.085824647225056e-05,
|
| 876 |
+
"loss": 1.1,
|
| 877 |
+
"step": 124
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 0.45724737082761774,
|
| 881 |
+
"grad_norm": 0.02895357273519039,
|
| 882 |
+
"learning_rate": 3.056302334890786e-05,
|
| 883 |
+
"loss": 1.1142,
|
| 884 |
+
"step": 125
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 0.4609053497942387,
|
| 888 |
+
"grad_norm": 0.03063139319419861,
|
| 889 |
+
"learning_rate": 3.0266981749893157e-05,
|
| 890 |
+
"loss": 1.1428,
|
| 891 |
+
"step": 126
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 0.46456332876085965,
|
| 895 |
+
"grad_norm": 0.0273541621863842,
|
| 896 |
+
"learning_rate": 2.997016523117081e-05,
|
| 897 |
+
"loss": 1.1185,
|
| 898 |
+
"step": 127
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 0.46822130772748055,
|
| 902 |
+
"grad_norm": 0.028322231024503708,
|
| 903 |
+
"learning_rate": 2.9672617462717444e-05,
|
| 904 |
+
"loss": 1.1318,
|
| 905 |
+
"step": 128
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 0.4718792866941015,
|
| 909 |
+
"grad_norm": 0.03172619268298149,
|
| 910 |
+
"learning_rate": 2.9374382222096886e-05,
|
| 911 |
+
"loss": 1.136,
|
| 912 |
+
"step": 129
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 0.47553726566072246,
|
| 916 |
+
"grad_norm": 0.02729085460305214,
|
| 917 |
+
"learning_rate": 2.9075503388019272e-05,
|
| 918 |
+
"loss": 1.1297,
|
| 919 |
+
"step": 130
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 0.4791952446273434,
|
| 923 |
+
"grad_norm": 0.02979847602546215,
|
| 924 |
+
"learning_rate": 2.877602493388525e-05,
|
| 925 |
+
"loss": 1.1195,
|
| 926 |
+
"step": 131
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 0.4828532235939643,
|
| 930 |
+
"grad_norm": 0.028623459860682487,
|
| 931 |
+
"learning_rate": 2.8475990921316292e-05,
|
| 932 |
+
"loss": 1.1349,
|
| 933 |
+
"step": 132
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 0.4865112025605853,
|
| 937 |
+
"grad_norm": 0.027000540867447853,
|
| 938 |
+
"learning_rate": 2.8175445493671972e-05,
|
| 939 |
+
"loss": 1.1132,
|
| 940 |
+
"step": 133
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 0.49016918152720623,
|
| 944 |
+
"grad_norm": 0.02691473811864853,
|
| 945 |
+
"learning_rate": 2.787443286955528e-05,
|
| 946 |
+
"loss": 1.1083,
|
| 947 |
+
"step": 134
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 0.49382716049382713,
|
| 951 |
+
"grad_norm": 0.027869021520018578,
|
| 952 |
+
"learning_rate": 2.7572997336306812e-05,
|
| 953 |
+
"loss": 1.1345,
|
| 954 |
+
"step": 135
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 0.4974851394604481,
|
| 958 |
+
"grad_norm": 0.029120417311787605,
|
| 959 |
+
"learning_rate": 2.7271183243488878e-05,
|
| 960 |
+
"loss": 1.1413,
|
| 961 |
+
"step": 136
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 0.501143118427069,
|
| 965 |
+
"grad_norm": 0.02720428816974163,
|
| 966 |
+
"learning_rate": 2.696903499636045e-05,
|
| 967 |
+
"loss": 1.1066,
|
| 968 |
+
"step": 137
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 0.50480109739369,
|
| 972 |
+
"grad_norm": 0.0264853835105896,
|
| 973 |
+
"learning_rate": 2.6666597049343882e-05,
|
| 974 |
+
"loss": 1.1257,
|
| 975 |
+
"step": 138
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 0.508459076360311,
|
| 979 |
+
"grad_norm": 0.02642214484512806,
|
| 980 |
+
"learning_rate": 2.636391389948449e-05,
|
| 981 |
+
"loss": 1.087,
|
| 982 |
+
"step": 139
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 0.5121170553269319,
|
| 986 |
+
"grad_norm": 0.02971823699772358,
|
| 987 |
+
"learning_rate": 2.606103007990371e-05,
|
| 988 |
+
"loss": 1.1263,
|
| 989 |
+
"step": 140
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 0.5157750342935528,
|
| 993 |
+
"grad_norm": 0.026655763387680054,
|
| 994 |
+
"learning_rate": 2.5757990153247124e-05,
|
| 995 |
+
"loss": 1.095,
|
| 996 |
+
"step": 141
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 0.5194330132601738,
|
| 1000 |
+
"grad_norm": 0.03130535036325455,
|
| 1001 |
+
"learning_rate": 2.5454838705127993e-05,
|
| 1002 |
+
"loss": 1.1259,
|
| 1003 |
+
"step": 142
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 0.5230909922267947,
|
| 1007 |
+
"grad_norm": 0.02821686491370201,
|
| 1008 |
+
"learning_rate": 2.5151620337567495e-05,
|
| 1009 |
+
"loss": 1.1159,
|
| 1010 |
+
"step": 143
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 0.5267489711934157,
|
| 1014 |
+
"grad_norm": 0.026856008917093277,
|
| 1015 |
+
"learning_rate": 2.4848379662432515e-05,
|
| 1016 |
+
"loss": 1.0893,
|
| 1017 |
+
"step": 144
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 0.5304069501600366,
|
| 1021 |
+
"grad_norm": 0.026246678084135056,
|
| 1022 |
+
"learning_rate": 2.4545161294872013e-05,
|
| 1023 |
+
"loss": 1.107,
|
| 1024 |
+
"step": 145
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 0.5340649291266575,
|
| 1028 |
+
"grad_norm": 0.03086758591234684,
|
| 1029 |
+
"learning_rate": 2.4242009846752885e-05,
|
| 1030 |
+
"loss": 1.096,
|
| 1031 |
+
"step": 146
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 0.5377229080932785,
|
| 1035 |
+
"grad_norm": 0.03089798055589199,
|
| 1036 |
+
"learning_rate": 2.39389699200963e-05,
|
| 1037 |
+
"loss": 1.1143,
|
| 1038 |
+
"step": 147
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 0.5413808870598994,
|
| 1042 |
+
"grad_norm": 0.02865259163081646,
|
| 1043 |
+
"learning_rate": 2.3636086100515527e-05,
|
| 1044 |
+
"loss": 1.0935,
|
| 1045 |
+
"step": 148
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 0.5450388660265203,
|
| 1049 |
+
"grad_norm": 0.02822408266365528,
|
| 1050 |
+
"learning_rate": 2.3333402950656124e-05,
|
| 1051 |
+
"loss": 1.092,
|
| 1052 |
+
"step": 149
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 0.5486968449931413,
|
| 1056 |
+
"grad_norm": 0.027232255786657333,
|
| 1057 |
+
"learning_rate": 2.3030965003639566e-05,
|
| 1058 |
+
"loss": 1.1075,
|
| 1059 |
+
"step": 150
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 0.5523548239597622,
|
| 1063 |
+
"grad_norm": 0.030684614554047585,
|
| 1064 |
+
"learning_rate": 2.272881675651112e-05,
|
| 1065 |
+
"loss": 1.0915,
|
| 1066 |
+
"step": 151
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 0.5560128029263832,
|
| 1070 |
+
"grad_norm": 0.027584511786699295,
|
| 1071 |
+
"learning_rate": 2.242700266369319e-05,
|
| 1072 |
+
"loss": 1.1051,
|
| 1073 |
+
"step": 152
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 0.5596707818930041,
|
| 1077 |
+
"grad_norm": 0.027764879167079926,
|
| 1078 |
+
"learning_rate": 2.2125567130444724e-05,
|
| 1079 |
+
"loss": 1.1229,
|
| 1080 |
+
"step": 153
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 0.563328760859625,
|
| 1084 |
+
"grad_norm": 0.02984793111681938,
|
| 1085 |
+
"learning_rate": 2.182455450632803e-05,
|
| 1086 |
+
"loss": 1.0987,
|
| 1087 |
+
"step": 154
|
| 1088 |
+
},
|
| 1089 |
+
{
|
| 1090 |
+
"epoch": 0.566986739826246,
|
| 1091 |
+
"grad_norm": 0.03010164201259613,
|
| 1092 |
+
"learning_rate": 2.1524009078683717e-05,
|
| 1093 |
+
"loss": 1.1299,
|
| 1094 |
+
"step": 155
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 0.5706447187928669,
|
| 1098 |
+
"grad_norm": 0.026011740788817406,
|
| 1099 |
+
"learning_rate": 2.1223975066114754e-05,
|
| 1100 |
+
"loss": 1.0861,
|
| 1101 |
+
"step": 156
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 0.5743026977594878,
|
| 1105 |
+
"grad_norm": 0.02605109103024006,
|
| 1106 |
+
"learning_rate": 2.0924496611980734e-05,
|
| 1107 |
+
"loss": 1.1276,
|
| 1108 |
+
"step": 157
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"epoch": 0.5779606767261088,
|
| 1112 |
+
"grad_norm": 0.02799968421459198,
|
| 1113 |
+
"learning_rate": 2.062561777790312e-05,
|
| 1114 |
+
"loss": 1.1116,
|
| 1115 |
+
"step": 158
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 0.5816186556927297,
|
| 1119 |
+
"grad_norm": 0.02790476195514202,
|
| 1120 |
+
"learning_rate": 2.0327382537282562e-05,
|
| 1121 |
+
"loss": 1.0978,
|
| 1122 |
+
"step": 159
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 0.5852766346593508,
|
| 1126 |
+
"grad_norm": 0.028209157288074493,
|
| 1127 |
+
"learning_rate": 2.0029834768829196e-05,
|
| 1128 |
+
"loss": 1.1248,
|
| 1129 |
+
"step": 160
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 0.5889346136259717,
|
| 1133 |
+
"grad_norm": 0.026728255674242973,
|
| 1134 |
+
"learning_rate": 1.973301825010685e-05,
|
| 1135 |
+
"loss": 1.0831,
|
| 1136 |
+
"step": 161
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 0.5925925925925926,
|
| 1140 |
+
"grad_norm": 0.02696623094379902,
|
| 1141 |
+
"learning_rate": 1.9436976651092144e-05,
|
| 1142 |
+
"loss": 1.0861,
|
| 1143 |
+
"step": 162
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 0.5962505715592136,
|
| 1147 |
+
"grad_norm": 0.025825291872024536,
|
| 1148 |
+
"learning_rate": 1.9141753527749443e-05,
|
| 1149 |
+
"loss": 1.1009,
|
| 1150 |
+
"step": 163
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 0.5999085505258345,
|
| 1154 |
+
"grad_norm": 0.02682458609342575,
|
| 1155 |
+
"learning_rate": 1.8847392315622535e-05,
|
| 1156 |
+
"loss": 1.1083,
|
| 1157 |
+
"step": 164
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 0.6035665294924554,
|
| 1161 |
+
"grad_norm": 0.029443522915244102,
|
| 1162 |
+
"learning_rate": 1.8553936323444058e-05,
|
| 1163 |
+
"loss": 1.1143,
|
| 1164 |
+
"step": 165
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 0.6072245084590764,
|
| 1168 |
+
"grad_norm": 0.02744133211672306,
|
| 1169 |
+
"learning_rate": 1.8261428726763584e-05,
|
| 1170 |
+
"loss": 1.0903,
|
| 1171 |
+
"step": 166
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 0.6108824874256973,
|
| 1175 |
+
"grad_norm": 0.026245735585689545,
|
| 1176 |
+
"learning_rate": 1.7969912561595317e-05,
|
| 1177 |
+
"loss": 1.0843,
|
| 1178 |
+
"step": 167
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 0.6145404663923183,
|
| 1182 |
+
"grad_norm": 0.027246467769145966,
|
| 1183 |
+
"learning_rate": 1.7679430718086243e-05,
|
| 1184 |
+
"loss": 1.091,
|
| 1185 |
+
"step": 168
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 0.6181984453589392,
|
| 1189 |
+
"grad_norm": 0.026942765340209007,
|
| 1190 |
+
"learning_rate": 1.7390025934205837e-05,
|
| 1191 |
+
"loss": 1.0844,
|
| 1192 |
+
"step": 169
|
| 1193 |
+
},
|
| 1194 |
+
{
|
| 1195 |
+
"epoch": 0.6218564243255601,
|
| 1196 |
+
"grad_norm": 0.02487928792834282,
|
| 1197 |
+
"learning_rate": 1.7101740789458097e-05,
|
| 1198 |
+
"loss": 1.0941,
|
| 1199 |
+
"step": 170
|
| 1200 |
+
},
|
| 1201 |
+
{
|
| 1202 |
+
"epoch": 0.6255144032921811,
|
| 1203 |
+
"grad_norm": 0.027003785595297813,
|
| 1204 |
+
"learning_rate": 1.6814617698616945e-05,
|
| 1205 |
+
"loss": 1.097,
|
| 1206 |
+
"step": 171
|
| 1207 |
+
},
|
| 1208 |
+
{
|
| 1209 |
+
"epoch": 0.629172382258802,
|
| 1210 |
+
"grad_norm": 0.025280645117163658,
|
| 1211 |
+
"learning_rate": 1.6528698905485784e-05,
|
| 1212 |
+
"loss": 1.0773,
|
| 1213 |
+
"step": 172
|
| 1214 |
+
},
|
| 1215 |
+
{
|
| 1216 |
+
"epoch": 0.6328303612254229,
|
| 1217 |
+
"grad_norm": 0.02396760694682598,
|
| 1218 |
+
"learning_rate": 1.624402647668229e-05,
|
| 1219 |
+
"loss": 1.0755,
|
| 1220 |
+
"step": 173
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 0.6364883401920439,
|
| 1224 |
+
"grad_norm": 0.023587588220834732,
|
| 1225 |
+
"learning_rate": 1.5960642295449228e-05,
|
| 1226 |
+
"loss": 1.0741,
|
| 1227 |
+
"step": 174
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"epoch": 0.6401463191586648,
|
| 1231 |
+
"grad_norm": 0.024295572191476822,
|
| 1232 |
+
"learning_rate": 1.567858805549229e-05,
|
| 1233 |
+
"loss": 1.1111,
|
| 1234 |
+
"step": 175
|
| 1235 |
+
},
|
| 1236 |
+
{
|
| 1237 |
+
"epoch": 0.6438042981252858,
|
| 1238 |
+
"grad_norm": 0.024151455610990524,
|
| 1239 |
+
"learning_rate": 1.539790525484573e-05,
|
| 1240 |
+
"loss": 1.0789,
|
| 1241 |
+
"step": 176
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 0.6474622770919067,
|
| 1245 |
+
"grad_norm": 0.024293672293424606,
|
| 1246 |
+
"learning_rate": 1.511863518976691e-05,
|
| 1247 |
+
"loss": 1.0671,
|
| 1248 |
+
"step": 177
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 0.6511202560585276,
|
| 1252 |
+
"grad_norm": 0.025013579055666924,
|
| 1253 |
+
"learning_rate": 1.4840818948660434e-05,
|
| 1254 |
+
"loss": 1.0899,
|
| 1255 |
+
"step": 178
|
| 1256 |
+
},
|
| 1257 |
+
{
|
| 1258 |
+
"epoch": 0.6547782350251486,
|
| 1259 |
+
"grad_norm": 0.02457435615360737,
|
| 1260 |
+
"learning_rate": 1.456449740603291e-05,
|
| 1261 |
+
"loss": 1.0998,
|
| 1262 |
+
"step": 179
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 0.6584362139917695,
|
| 1266 |
+
"grad_norm": 0.024968327954411507,
|
| 1267 |
+
"learning_rate": 1.4289711216479156e-05,
|
| 1268 |
+
"loss": 1.0846,
|
| 1269 |
+
"step": 180
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 0.6620941929583904,
|
| 1273 |
+
"grad_norm": 0.023626726120710373,
|
| 1274 |
+
"learning_rate": 1.401650080870083e-05,
|
| 1275 |
+
"loss": 1.0774,
|
| 1276 |
+
"step": 181
|
| 1277 |
+
},
|
| 1278 |
+
{
|
| 1279 |
+
"epoch": 0.6657521719250115,
|
| 1280 |
+
"grad_norm": 0.025500517338514328,
|
| 1281 |
+
"learning_rate": 1.3744906379558165e-05,
|
| 1282 |
+
"loss": 1.096,
|
| 1283 |
+
"step": 182
|
| 1284 |
+
},
|
| 1285 |
+
{
|
| 1286 |
+
"epoch": 0.6694101508916324,
|
| 1287 |
+
"grad_norm": 0.02676280215382576,
|
| 1288 |
+
"learning_rate": 1.3474967888155948e-05,
|
| 1289 |
+
"loss": 1.0984,
|
| 1290 |
+
"step": 183
|
| 1291 |
+
},
|
| 1292 |
+
{
|
| 1293 |
+
"epoch": 0.6730681298582533,
|
| 1294 |
+
"grad_norm": 0.024672549217939377,
|
| 1295 |
+
"learning_rate": 1.3206725049964392e-05,
|
| 1296 |
+
"loss": 1.1026,
|
| 1297 |
+
"step": 184
|
| 1298 |
+
},
|
| 1299 |
+
{
|
| 1300 |
+
"epoch": 0.6767261088248743,
|
| 1301 |
+
"grad_norm": 0.02653788961470127,
|
| 1302 |
+
"learning_rate": 1.2940217330975912e-05,
|
| 1303 |
+
"loss": 1.0898,
|
| 1304 |
+
"step": 185
|
| 1305 |
+
},
|
| 1306 |
+
{
|
| 1307 |
+
"epoch": 0.6803840877914952,
|
| 1308 |
+
"grad_norm": 0.024248803034424782,
|
| 1309 |
+
"learning_rate": 1.2675483941898548e-05,
|
| 1310 |
+
"loss": 1.0925,
|
| 1311 |
+
"step": 186
|
| 1312 |
+
},
|
| 1313 |
+
{
|
| 1314 |
+
"epoch": 0.6840420667581162,
|
| 1315 |
+
"grad_norm": 0.02738628350198269,
|
| 1316 |
+
"learning_rate": 1.2412563832387003e-05,
|
| 1317 |
+
"loss": 1.0916,
|
| 1318 |
+
"step": 187
|
| 1319 |
+
},
|
| 1320 |
+
{
|
| 1321 |
+
"epoch": 0.6877000457247371,
|
| 1322 |
+
"grad_norm": 0.023650605231523514,
|
| 1323 |
+
"learning_rate": 1.2151495685312054e-05,
|
| 1324 |
+
"loss": 1.0862,
|
| 1325 |
+
"step": 188
|
| 1326 |
+
},
|
| 1327 |
+
{
|
| 1328 |
+
"epoch": 0.691358024691358,
|
| 1329 |
+
"grad_norm": 0.024786144495010376,
|
| 1330 |
+
"learning_rate": 1.1892317911069212e-05,
|
| 1331 |
+
"loss": 1.0791,
|
| 1332 |
+
"step": 189
|
| 1333 |
+
},
|
| 1334 |
+
{
|
| 1335 |
+
"epoch": 0.695016003657979,
|
| 1336 |
+
"grad_norm": 0.026861809194087982,
|
| 1337 |
+
"learning_rate": 1.163506864192751e-05,
|
| 1338 |
+
"loss": 1.1008,
|
| 1339 |
+
"step": 190
|
| 1340 |
+
},
|
| 1341 |
+
{
|
| 1342 |
+
"epoch": 0.6986739826245999,
|
| 1343 |
+
"grad_norm": 0.024741502478718758,
|
| 1344 |
+
"learning_rate": 1.1379785726419162e-05,
|
| 1345 |
+
"loss": 1.0804,
|
| 1346 |
+
"step": 191
|
| 1347 |
+
},
|
| 1348 |
+
{
|
| 1349 |
+
"epoch": 0.7023319615912208,
|
| 1350 |
+
"grad_norm": 0.023557139560580254,
|
| 1351 |
+
"learning_rate": 1.1126506723770996e-05,
|
| 1352 |
+
"loss": 1.0503,
|
| 1353 |
+
"step": 192
|
| 1354 |
+
},
|
| 1355 |
+
{
|
| 1356 |
+
"epoch": 0.7059899405578418,
|
| 1357 |
+
"grad_norm": 0.023487713187932968,
|
| 1358 |
+
"learning_rate": 1.087526889837845e-05,
|
| 1359 |
+
"loss": 1.0778,
|
| 1360 |
+
"step": 193
|
| 1361 |
+
},
|
| 1362 |
+
{
|
| 1363 |
+
"epoch": 0.7096479195244627,
|
| 1364 |
+
"grad_norm": 0.025643397122621536,
|
| 1365 |
+
"learning_rate": 1.0626109214322923e-05,
|
| 1366 |
+
"loss": 1.081,
|
| 1367 |
+
"step": 194
|
| 1368 |
+
},
|
| 1369 |
+
{
|
| 1370 |
+
"epoch": 0.7133058984910837,
|
| 1371 |
+
"grad_norm": 0.02375604212284088,
|
| 1372 |
+
"learning_rate": 1.0379064329933385e-05,
|
| 1373 |
+
"loss": 1.0659,
|
| 1374 |
+
"step": 195
|
| 1375 |
+
},
|
| 1376 |
+
{
|
| 1377 |
+
"epoch": 0.7169638774577046,
|
| 1378 |
+
"grad_norm": 0.02461252547800541,
|
| 1379 |
+
"learning_rate": 1.0134170592392836e-05,
|
| 1380 |
+
"loss": 1.0785,
|
| 1381 |
+
"step": 196
|
| 1382 |
+
},
|
| 1383 |
+
{
|
| 1384 |
+
"epoch": 0.7206218564243255,
|
| 1385 |
+
"grad_norm": 0.024264885112643242,
|
| 1386 |
+
"learning_rate": 9.89146403239067e-06,
|
| 1387 |
+
"loss": 1.0868,
|
| 1388 |
+
"step": 197
|
| 1389 |
+
},
|
| 1390 |
+
{
|
| 1391 |
+
"epoch": 0.7242798353909465,
|
| 1392 |
+
"grad_norm": 0.02400391362607479,
|
| 1393 |
+
"learning_rate": 9.650980358821559e-06,
|
| 1394 |
+
"loss": 1.0826,
|
| 1395 |
+
"step": 198
|
| 1396 |
+
},
|
| 1397 |
+
{
|
| 1398 |
+
"epoch": 0.7279378143575674,
|
| 1399 |
+
"grad_norm": 0.025086689740419388,
|
| 1400 |
+
"learning_rate": 9.412754953531663e-06,
|
| 1401 |
+
"loss": 1.0726,
|
| 1402 |
+
"step": 199
|
| 1403 |
+
},
|
| 1404 |
+
{
|
| 1405 |
+
"epoch": 0.7315957933241883,
|
| 1406 |
+
"grad_norm": 0.023492420092225075,
|
| 1407 |
+
"learning_rate": 9.176822866112988e-06,
|
| 1408 |
+
"loss": 1.0781,
|
| 1409 |
+
"step": 200
|
| 1410 |
+
},
|
| 1411 |
+
{
|
| 1412 |
+
"epoch": 0.7352537722908093,
|
| 1413 |
+
"grad_norm": 0.022942854091525078,
|
| 1414 |
+
"learning_rate": 8.943218808746603e-06,
|
| 1415 |
+
"loss": 1.0589,
|
| 1416 |
+
"step": 201
|
| 1417 |
+
},
|
| 1418 |
+
{
|
| 1419 |
+
"epoch": 0.7389117512574302,
|
| 1420 |
+
"grad_norm": 0.0239090695977211,
|
| 1421 |
+
"learning_rate": 8.711977151095524e-06,
|
| 1422 |
+
"loss": 1.0627,
|
| 1423 |
+
"step": 202
|
| 1424 |
+
},
|
| 1425 |
+
{
|
| 1426 |
+
"epoch": 0.7425697302240513,
|
| 1427 |
+
"grad_norm": 0.02404557354748249,
|
| 1428 |
+
"learning_rate": 8.483131915247968e-06,
|
| 1429 |
+
"loss": 1.0707,
|
| 1430 |
+
"step": 203
|
| 1431 |
+
},
|
| 1432 |
+
{
|
| 1433 |
+
"epoch": 0.7462277091906722,
|
| 1434 |
+
"grad_norm": 0.02398700639605522,
|
| 1435 |
+
"learning_rate": 8.256716770711764e-06,
|
| 1436 |
+
"loss": 1.0637,
|
| 1437 |
+
"step": 204
|
| 1438 |
+
},
|
| 1439 |
+
{
|
| 1440 |
+
"epoch": 0.7498856881572931,
|
| 1441 |
+
"grad_norm": 0.02359054796397686,
|
| 1442 |
+
"learning_rate": 8.032765029460625e-06,
|
| 1443 |
+
"loss": 1.0806,
|
| 1444 |
+
"step": 205
|
| 1445 |
+
},
|
| 1446 |
+
{
|
| 1447 |
+
"epoch": 0.7535436671239141,
|
| 1448 |
+
"grad_norm": 0.022856276482343674,
|
| 1449 |
+
"learning_rate": 7.811309641033041e-06,
|
| 1450 |
+
"loss": 1.0466,
|
| 1451 |
+
"step": 206
|
| 1452 |
+
},
|
| 1453 |
+
{
|
| 1454 |
+
"epoch": 0.757201646090535,
|
| 1455 |
+
"grad_norm": 0.02421443723142147,
|
| 1456 |
+
"learning_rate": 7.592383187684457e-06,
|
| 1457 |
+
"loss": 1.0805,
|
| 1458 |
+
"step": 207
|
| 1459 |
+
},
|
| 1460 |
+
{
|
| 1461 |
+
"epoch": 0.7608596250571559,
|
| 1462 |
+
"grad_norm": 0.025367150083184242,
|
| 1463 |
+
"learning_rate": 7.376017879593553e-06,
|
| 1464 |
+
"loss": 1.098,
|
| 1465 |
+
"step": 208
|
| 1466 |
+
},
|
| 1467 |
+
{
|
| 1468 |
+
"epoch": 0.7645176040237769,
|
| 1469 |
+
"grad_norm": 0.02458193711936474,
|
| 1470 |
+
"learning_rate": 7.1622455501231925e-06,
|
| 1471 |
+
"loss": 1.0872,
|
| 1472 |
+
"step": 209
|
| 1473 |
+
},
|
| 1474 |
+
{
|
| 1475 |
+
"epoch": 0.7681755829903978,
|
| 1476 |
+
"grad_norm": 0.02363410033285618,
|
| 1477 |
+
"learning_rate": 6.951097651136889e-06,
|
| 1478 |
+
"loss": 1.0892,
|
| 1479 |
+
"step": 210
|
| 1480 |
+
},
|
| 1481 |
+
{
|
| 1482 |
+
"epoch": 0.7718335619570188,
|
| 1483 |
+
"grad_norm": 0.023377366364002228,
|
| 1484 |
+
"learning_rate": 6.74260524837132e-06,
|
| 1485 |
+
"loss": 1.0703,
|
| 1486 |
+
"step": 211
|
| 1487 |
+
},
|
| 1488 |
+
{
|
| 1489 |
+
"epoch": 0.7754915409236397,
|
| 1490 |
+
"grad_norm": 0.02378440462052822,
|
| 1491 |
+
"learning_rate": 6.536799016865713e-06,
|
| 1492 |
+
"loss": 1.0825,
|
| 1493 |
+
"step": 212
|
| 1494 |
+
},
|
| 1495 |
+
{
|
| 1496 |
+
"epoch": 0.7791495198902606,
|
| 1497 |
+
"grad_norm": 0.023137414827942848,
|
| 1498 |
+
"learning_rate": 6.333709236448693e-06,
|
| 1499 |
+
"loss": 1.054,
|
| 1500 |
+
"step": 213
|
| 1501 |
+
},
|
| 1502 |
+
{
|
| 1503 |
+
"epoch": 0.7828074988568816,
|
| 1504 |
+
"grad_norm": 0.024228321388363838,
|
| 1505 |
+
"learning_rate": 6.133365787283268e-06,
|
| 1506 |
+
"loss": 1.0677,
|
| 1507 |
+
"step": 214
|
| 1508 |
+
},
|
| 1509 |
+
{
|
| 1510 |
+
"epoch": 0.7864654778235025,
|
| 1511 |
+
"grad_norm": 0.02322709932923317,
|
| 1512 |
+
"learning_rate": 5.935798145470639e-06,
|
| 1513 |
+
"loss": 1.0539,
|
| 1514 |
+
"step": 215
|
| 1515 |
+
},
|
| 1516 |
+
{
|
| 1517 |
+
"epoch": 0.7901234567901234,
|
| 1518 |
+
"grad_norm": 0.02640087902545929,
|
| 1519 |
+
"learning_rate": 5.741035378713428e-06,
|
| 1520 |
+
"loss": 1.0623,
|
| 1521 |
+
"step": 216
|
| 1522 |
+
},
|
| 1523 |
+
{
|
| 1524 |
+
"epoch": 0.7937814357567444,
|
| 1525 |
+
"grad_norm": 0.024857627227902412,
|
| 1526 |
+
"learning_rate": 5.549106142039018e-06,
|
| 1527 |
+
"loss": 1.0866,
|
| 1528 |
+
"step": 217
|
| 1529 |
+
},
|
| 1530 |
+
{
|
| 1531 |
+
"epoch": 0.7974394147233653,
|
| 1532 |
+
"grad_norm": 0.02291812375187874,
|
| 1533 |
+
"learning_rate": 5.360038673583601e-06,
|
| 1534 |
+
"loss": 1.0728,
|
| 1535 |
+
"step": 218
|
| 1536 |
+
},
|
| 1537 |
+
{
|
| 1538 |
+
"epoch": 0.8010973936899863,
|
| 1539 |
+
"grad_norm": 0.02379731461405754,
|
| 1540 |
+
"learning_rate": 5.173860790437563e-06,
|
| 1541 |
+
"loss": 1.0333,
|
| 1542 |
+
"step": 219
|
| 1543 |
+
},
|
| 1544 |
+
{
|
| 1545 |
+
"epoch": 0.8047553726566072,
|
| 1546 |
+
"grad_norm": 0.02438983879983425,
|
| 1547 |
+
"learning_rate": 4.990599884552802e-06,
|
| 1548 |
+
"loss": 1.0556,
|
| 1549 |
+
"step": 220
|
| 1550 |
+
},
|
| 1551 |
+
{
|
| 1552 |
+
"epoch": 0.8084133516232281,
|
| 1553 |
+
"grad_norm": 0.023418473079800606,
|
| 1554 |
+
"learning_rate": 4.8102829187126355e-06,
|
| 1555 |
+
"loss": 1.0908,
|
| 1556 |
+
"step": 221
|
| 1557 |
+
},
|
| 1558 |
+
{
|
| 1559 |
+
"epoch": 0.8120713305898491,
|
| 1560 |
+
"grad_norm": 0.022404542192816734,
|
| 1561 |
+
"learning_rate": 4.632936422564785e-06,
|
| 1562 |
+
"loss": 1.0702,
|
| 1563 |
+
"step": 222
|
| 1564 |
+
},
|
| 1565 |
+
{
|
| 1566 |
+
"epoch": 0.81572930955647,
|
| 1567 |
+
"grad_norm": 0.024710949510335922,
|
| 1568 |
+
"learning_rate": 4.458586488718178e-06,
|
| 1569 |
+
"loss": 1.0651,
|
| 1570 |
+
"step": 223
|
| 1571 |
+
},
|
| 1572 |
+
{
|
| 1573 |
+
"epoch": 0.819387288523091,
|
| 1574 |
+
"grad_norm": 0.023974299430847168,
|
| 1575 |
+
"learning_rate": 4.2872587689039484e-06,
|
| 1576 |
+
"loss": 1.0798,
|
| 1577 |
+
"step": 224
|
| 1578 |
+
},
|
| 1579 |
+
{
|
| 1580 |
+
"epoch": 0.823045267489712,
|
| 1581 |
+
"grad_norm": 0.022725731134414673,
|
| 1582 |
+
"learning_rate": 4.118978470201379e-06,
|
| 1583 |
+
"loss": 1.0739,
|
| 1584 |
+
"step": 225
|
| 1585 |
+
},
|
| 1586 |
+
{
|
| 1587 |
+
"epoch": 0.8267032464563329,
|
| 1588 |
+
"grad_norm": 0.023963825777173042,
|
| 1589 |
+
"learning_rate": 3.953770351329228e-06,
|
| 1590 |
+
"loss": 1.0608,
|
| 1591 |
+
"step": 226
|
| 1592 |
+
},
|
| 1593 |
+
{
|
| 1594 |
+
"epoch": 0.8303612254229538,
|
| 1595 |
+
"grad_norm": 0.02529250457882881,
|
| 1596 |
+
"learning_rate": 3.7916587190030467e-06,
|
| 1597 |
+
"loss": 1.0827,
|
| 1598 |
+
"step": 227
|
| 1599 |
+
},
|
| 1600 |
+
{
|
| 1601 |
+
"epoch": 0.8340192043895748,
|
| 1602 |
+
"grad_norm": 0.023284699767827988,
|
| 1603 |
+
"learning_rate": 3.6326674243589653e-06,
|
| 1604 |
+
"loss": 1.0715,
|
| 1605 |
+
"step": 228
|
| 1606 |
+
},
|
| 1607 |
+
{
|
| 1608 |
+
"epoch": 0.8376771833561957,
|
| 1609 |
+
"grad_norm": 0.02325253002345562,
|
| 1610 |
+
"learning_rate": 3.476819859444538e-06,
|
| 1611 |
+
"loss": 1.068,
|
| 1612 |
+
"step": 229
|
| 1613 |
+
},
|
| 1614 |
+
{
|
| 1615 |
+
"epoch": 0.8413351623228167,
|
| 1616 |
+
"grad_norm": 0.02358105778694153,
|
| 1617 |
+
"learning_rate": 3.3241389537771174e-06,
|
| 1618 |
+
"loss": 1.0859,
|
| 1619 |
+
"step": 230
|
| 1620 |
+
},
|
| 1621 |
+
{
|
| 1622 |
+
"epoch": 0.8449931412894376,
|
| 1623 |
+
"grad_norm": 0.023099737241864204,
|
| 1624 |
+
"learning_rate": 3.1746471709702964e-06,
|
| 1625 |
+
"loss": 1.0818,
|
| 1626 |
+
"step": 231
|
| 1627 |
+
},
|
| 1628 |
+
{
|
| 1629 |
+
"epoch": 0.8486511202560585,
|
| 1630 |
+
"grad_norm": 0.02324778586626053,
|
| 1631 |
+
"learning_rate": 3.028366505428856e-06,
|
| 1632 |
+
"loss": 1.0371,
|
| 1633 |
+
"step": 232
|
| 1634 |
+
},
|
| 1635 |
+
{
|
| 1636 |
+
"epoch": 0.8523090992226795,
|
| 1637 |
+
"grad_norm": 0.02218066342175007,
|
| 1638 |
+
"learning_rate": 2.8853184791128086e-06,
|
| 1639 |
+
"loss": 1.1001,
|
| 1640 |
+
"step": 233
|
| 1641 |
+
},
|
| 1642 |
+
{
|
| 1643 |
+
"epoch": 0.8559670781893004,
|
| 1644 |
+
"grad_norm": 0.022952284663915634,
|
| 1645 |
+
"learning_rate": 2.745524138370892e-06,
|
| 1646 |
+
"loss": 1.0495,
|
| 1647 |
+
"step": 234
|
| 1648 |
+
},
|
| 1649 |
+
{
|
| 1650 |
+
"epoch": 0.8596250571559213,
|
| 1651 |
+
"grad_norm": 0.022721242159605026,
|
| 1652 |
+
"learning_rate": 2.6090040508440857e-06,
|
| 1653 |
+
"loss": 1.0614,
|
| 1654 |
+
"step": 235
|
| 1655 |
+
},
|
| 1656 |
+
{
|
| 1657 |
+
"epoch": 0.8632830361225423,
|
| 1658 |
+
"grad_norm": 0.02230103127658367,
|
| 1659 |
+
"learning_rate": 2.475778302439524e-06,
|
| 1660 |
+
"loss": 1.0881,
|
| 1661 |
+
"step": 236
|
| 1662 |
+
},
|
| 1663 |
+
{
|
| 1664 |
+
"epoch": 0.8669410150891632,
|
| 1665 |
+
"grad_norm": 0.023484766483306885,
|
| 1666 |
+
"learning_rate": 2.345866494375307e-06,
|
| 1667 |
+
"loss": 1.0753,
|
| 1668 |
+
"step": 237
|
| 1669 |
+
},
|
| 1670 |
+
{
|
| 1671 |
+
"epoch": 0.8705989940557842,
|
| 1672 |
+
"grad_norm": 0.024059921503067017,
|
| 1673 |
+
"learning_rate": 2.219287740296605e-06,
|
| 1674 |
+
"loss": 1.0512,
|
| 1675 |
+
"step": 238
|
| 1676 |
+
},
|
| 1677 |
+
{
|
| 1678 |
+
"epoch": 0.8742569730224051,
|
| 1679 |
+
"grad_norm": 0.02326652780175209,
|
| 1680 |
+
"learning_rate": 2.0960606634635365e-06,
|
| 1681 |
+
"loss": 1.085,
|
| 1682 |
+
"step": 239
|
| 1683 |
+
},
|
| 1684 |
+
{
|
| 1685 |
+
"epoch": 0.877914951989026,
|
| 1686 |
+
"grad_norm": 0.024198243394494057,
|
| 1687 |
+
"learning_rate": 1.9762033940111335e-06,
|
| 1688 |
+
"loss": 1.0602,
|
| 1689 |
+
"step": 240
|
| 1690 |
+
},
|
| 1691 |
+
{
|
| 1692 |
+
"epoch": 0.881572930955647,
|
| 1693 |
+
"grad_norm": 0.02250683680176735,
|
| 1694 |
+
"learning_rate": 1.8597335662819099e-06,
|
| 1695 |
+
"loss": 1.0696,
|
| 1696 |
+
"step": 241
|
| 1697 |
+
},
|
| 1698 |
+
{
|
| 1699 |
+
"epoch": 0.8852309099222679,
|
| 1700 |
+
"grad_norm": 0.02158784680068493,
|
| 1701 |
+
"learning_rate": 1.746668316231359e-06,
|
| 1702 |
+
"loss": 1.0706,
|
| 1703 |
+
"step": 242
|
| 1704 |
+
},
|
| 1705 |
+
{
|
| 1706 |
+
"epoch": 0.8888888888888888,
|
| 1707 |
+
"grad_norm": 0.023256294429302216,
|
| 1708 |
+
"learning_rate": 1.6370242789067764e-06,
|
| 1709 |
+
"loss": 1.103,
|
| 1710 |
+
"step": 243
|
| 1711 |
+
},
|
| 1712 |
+
{
|
| 1713 |
+
"epoch": 0.8925468678555099,
|
| 1714 |
+
"grad_norm": 0.02234894596040249,
|
| 1715 |
+
"learning_rate": 1.530817585999783e-06,
|
| 1716 |
+
"loss": 1.0722,
|
| 1717 |
+
"step": 244
|
| 1718 |
+
},
|
| 1719 |
+
{
|
| 1720 |
+
"epoch": 0.8962048468221308,
|
| 1721 |
+
"grad_norm": 0.02256467379629612,
|
| 1722 |
+
"learning_rate": 1.428063863472895e-06,
|
| 1723 |
+
"loss": 1.0671,
|
| 1724 |
+
"step": 245
|
| 1725 |
+
},
|
| 1726 |
+
{
|
| 1727 |
+
"epoch": 0.8998628257887518,
|
| 1728 |
+
"grad_norm": 0.02275981567800045,
|
| 1729 |
+
"learning_rate": 1.3287782292605245e-06,
|
| 1730 |
+
"loss": 1.0957,
|
| 1731 |
+
"step": 246
|
| 1732 |
+
},
|
| 1733 |
+
{
|
| 1734 |
+
"epoch": 0.9035208047553727,
|
| 1735 |
+
"grad_norm": 0.022567572072148323,
|
| 1736 |
+
"learning_rate": 1.2329752910446913e-06,
|
| 1737 |
+
"loss": 1.0343,
|
| 1738 |
+
"step": 247
|
| 1739 |
+
},
|
| 1740 |
+
{
|
| 1741 |
+
"epoch": 0.9071787837219936,
|
| 1742 |
+
"grad_norm": 0.022027762606739998,
|
| 1743 |
+
"learning_rate": 1.1406691441058493e-06,
|
| 1744 |
+
"loss": 1.0584,
|
| 1745 |
+
"step": 248
|
| 1746 |
+
},
|
| 1747 |
+
{
|
| 1748 |
+
"epoch": 0.9108367626886146,
|
| 1749 |
+
"grad_norm": 0.022433646023273468,
|
| 1750 |
+
"learning_rate": 1.0518733692490513e-06,
|
| 1751 |
+
"loss": 1.0421,
|
| 1752 |
+
"step": 249
|
| 1753 |
+
},
|
| 1754 |
+
{
|
| 1755 |
+
"epoch": 0.9144947416552355,
|
| 1756 |
+
"grad_norm": 0.023060685023665428,
|
| 1757 |
+
"learning_rate": 9.66601030805861e-07,
|
| 1758 |
+
"loss": 1.057,
|
| 1759 |
+
"step": 250
|
| 1760 |
+
},
|
| 1761 |
+
{
|
| 1762 |
+
"epoch": 0.9181527206218564,
|
| 1763 |
+
"grad_norm": 0.021818097680807114,
|
| 1764 |
+
"learning_rate": 8.848646747122075e-07,
|
| 1765 |
+
"loss": 1.0751,
|
| 1766 |
+
"step": 251
|
| 1767 |
+
},
|
| 1768 |
+
{
|
| 1769 |
+
"epoch": 0.9218106995884774,
|
| 1770 |
+
"grad_norm": 0.02274404466152191,
|
| 1771 |
+
"learning_rate": 8.066763266625282e-07,
|
| 1772 |
+
"loss": 1.0533,
|
| 1773 |
+
"step": 252
|
| 1774 |
+
},
|
| 1775 |
+
{
|
| 1776 |
+
"epoch": 0.9254686785550983,
|
| 1777 |
+
"grad_norm": 0.02207763120532036,
|
| 1778 |
+
"learning_rate": 7.320474903404762e-07,
|
| 1779 |
+
"loss": 1.0718,
|
| 1780 |
+
"step": 253
|
| 1781 |
+
},
|
| 1782 |
+
{
|
| 1783 |
+
"epoch": 0.9291266575217193,
|
| 1784 |
+
"grad_norm": 0.021938692778348923,
|
| 1785 |
+
"learning_rate": 6.609891457263817e-07,
|
| 1786 |
+
"loss": 1.0668,
|
| 1787 |
+
"step": 254
|
| 1788 |
+
},
|
| 1789 |
+
{
|
| 1790 |
+
"epoch": 0.9327846364883402,
|
| 1791 |
+
"grad_norm": 0.023405423387885094,
|
| 1792 |
+
"learning_rate": 5.935117474818036e-07,
|
| 1793 |
+
"loss": 1.0688,
|
| 1794 |
+
"step": 255
|
| 1795 |
+
},
|
| 1796 |
+
{
|
| 1797 |
+
"epoch": 0.9364426154549611,
|
| 1798 |
+
"grad_norm": 0.02231016755104065,
|
| 1799 |
+
"learning_rate": 5.296252234113564e-07,
|
| 1800 |
+
"loss": 1.0612,
|
| 1801 |
+
"step": 256
|
| 1802 |
+
},
|
| 1803 |
+
{
|
| 1804 |
+
"epoch": 0.9401005944215821,
|
| 1805 |
+
"grad_norm": 0.022039178758859634,
|
| 1806 |
+
"learning_rate": 4.6933897300205085e-07,
|
| 1807 |
+
"loss": 1.0607,
|
| 1808 |
+
"step": 257
|
| 1809 |
+
},
|
| 1810 |
+
{
|
| 1811 |
+
"epoch": 0.943758573388203,
|
| 1812 |
+
"grad_norm": 0.02199982851743698,
|
| 1813 |
+
"learning_rate": 4.1266186604037307e-07,
|
| 1814 |
+
"loss": 1.0678,
|
| 1815 |
+
"step": 258
|
| 1816 |
+
},
|
| 1817 |
+
{
|
| 1818 |
+
"epoch": 0.9474165523548239,
|
| 1819 |
+
"grad_norm": 0.021750394254922867,
|
| 1820 |
+
"learning_rate": 3.5960224130728857e-07,
|
| 1821 |
+
"loss": 1.0511,
|
| 1822 |
+
"step": 259
|
| 1823 |
+
},
|
| 1824 |
+
{
|
| 1825 |
+
"epoch": 0.9510745313214449,
|
| 1826 |
+
"grad_norm": 0.021463261917233467,
|
| 1827 |
+
"learning_rate": 3.101679053513745e-07,
|
| 1828 |
+
"loss": 1.0586,
|
| 1829 |
+
"step": 260
|
| 1830 |
+
},
|
| 1831 |
+
{
|
| 1832 |
+
"epoch": 0.9547325102880658,
|
| 1833 |
+
"grad_norm": 0.022153394296765327,
|
| 1834 |
+
"learning_rate": 2.643661313402601e-07,
|
| 1835 |
+
"loss": 1.0655,
|
| 1836 |
+
"step": 261
|
| 1837 |
+
},
|
| 1838 |
+
{
|
| 1839 |
+
"epoch": 0.9583904892546868,
|
| 1840 |
+
"grad_norm": 0.022121533751487732,
|
| 1841 |
+
"learning_rate": 2.2220365799053843e-07,
|
| 1842 |
+
"loss": 1.0519,
|
| 1843 |
+
"step": 262
|
| 1844 |
+
},
|
| 1845 |
+
{
|
| 1846 |
+
"epoch": 0.9620484682213077,
|
| 1847 |
+
"grad_norm": 0.022767378017306328,
|
| 1848 |
+
"learning_rate": 1.836866885763233e-07,
|
| 1849 |
+
"loss": 1.097,
|
| 1850 |
+
"step": 263
|
| 1851 |
+
},
|
| 1852 |
+
{
|
| 1853 |
+
"epoch": 0.9657064471879286,
|
| 1854 |
+
"grad_norm": 0.021641606464982033,
|
| 1855 |
+
"learning_rate": 1.488208900165572e-07,
|
| 1856 |
+
"loss": 1.0564,
|
| 1857 |
+
"step": 264
|
| 1858 |
+
},
|
| 1859 |
+
{
|
| 1860 |
+
"epoch": 0.9693644261545497,
|
| 1861 |
+
"grad_norm": 0.022372765466570854,
|
| 1862 |
+
"learning_rate": 1.176113920412697e-07,
|
| 1863 |
+
"loss": 1.0683,
|
| 1864 |
+
"step": 265
|
| 1865 |
+
},
|
| 1866 |
+
{
|
| 1867 |
+
"epoch": 0.9730224051211706,
|
| 1868 |
+
"grad_norm": 0.021753614768385887,
|
| 1869 |
+
"learning_rate": 9.006278643683696e-08,
|
| 1870 |
+
"loss": 1.0648,
|
| 1871 |
+
"step": 266
|
| 1872 |
+
},
|
| 1873 |
+
{
|
| 1874 |
+
"epoch": 0.9766803840877915,
|
| 1875 |
+
"grad_norm": 0.022067097947001457,
|
| 1876 |
+
"learning_rate": 6.617912637040812e-08,
|
| 1877 |
+
"loss": 1.0781,
|
| 1878 |
+
"step": 267
|
| 1879 |
+
},
|
| 1880 |
+
{
|
| 1881 |
+
"epoch": 0.9803383630544125,
|
| 1882 |
+
"grad_norm": 0.021563511341810226,
|
| 1883 |
+
"learning_rate": 4.596392579356845e-08,
|
| 1884 |
+
"loss": 1.0615,
|
| 1885 |
+
"step": 268
|
| 1886 |
+
},
|
| 1887 |
+
{
|
| 1888 |
+
"epoch": 0.9839963420210334,
|
| 1889 |
+
"grad_norm": 0.022357366979122162,
|
| 1890 |
+
"learning_rate": 2.9420158925341777e-08,
|
| 1891 |
+
"loss": 1.0623,
|
| 1892 |
+
"step": 269
|
| 1893 |
+
},
|
| 1894 |
+
{
|
| 1895 |
+
"epoch": 0.9876543209876543,
|
| 1896 |
+
"grad_norm": 0.02167103812098503,
|
| 1897 |
+
"learning_rate": 1.6550259814601676e-08,
|
| 1898 |
+
"loss": 1.0464,
|
| 1899 |
+
"step": 270
|
| 1900 |
+
},
|
| 1901 |
+
{
|
| 1902 |
+
"epoch": 0.9913122999542753,
|
| 1903 |
+
"grad_norm": 0.021607818081974983,
|
| 1904 |
+
"learning_rate": 7.356121981946795e-09,
|
| 1905 |
+
"loss": 1.0607,
|
| 1906 |
+
"step": 271
|
| 1907 |
+
},
|
| 1908 |
+
{
|
| 1909 |
+
"epoch": 0.9949702789208962,
|
| 1910 |
+
"grad_norm": 0.022355254739522934,
|
| 1911 |
+
"learning_rate": 1.8390981411264829e-09,
|
| 1912 |
+
"loss": 1.0465,
|
| 1913 |
+
"step": 272
|
| 1914 |
+
},
|
| 1915 |
+
{
|
| 1916 |
+
"epoch": 0.9986282578875172,
|
| 1917 |
+
"grad_norm": 0.022043729200959206,
|
| 1918 |
+
"learning_rate": 0.0,
|
| 1919 |
+
"loss": 1.0636,
|
| 1920 |
+
"step": 273
|
| 1921 |
+
},
|
| 1922 |
+
{
|
| 1923 |
+
"epoch": 0.9986282578875172,
|
| 1924 |
+
"step": 273,
|
| 1925 |
+
"total_flos": 1921476294868992.0,
|
| 1926 |
+
"train_loss": 1.1405032803287436,
|
| 1927 |
+
"train_runtime": 59872.6859,
|
| 1928 |
+
"train_samples_per_second": 0.584,
|
| 1929 |
+
"train_steps_per_second": 0.005
|
| 1930 |
+
}
|
| 1931 |
+
],
|
| 1932 |
+
"logging_steps": 1,
|
| 1933 |
+
"max_steps": 273,
|
| 1934 |
+
"num_input_tokens_seen": 0,
|
| 1935 |
+
"num_train_epochs": 1,
|
| 1936 |
+
"save_steps": 100,
|
| 1937 |
+
"stateful_callbacks": {
|
| 1938 |
+
"TrainerControl": {
|
| 1939 |
+
"args": {
|
| 1940 |
+
"should_epoch_stop": false,
|
| 1941 |
+
"should_evaluate": false,
|
| 1942 |
+
"should_log": false,
|
| 1943 |
+
"should_save": true,
|
| 1944 |
+
"should_training_stop": true
|
| 1945 |
+
},
|
| 1946 |
+
"attributes": {}
|
| 1947 |
+
}
|
| 1948 |
+
},
|
| 1949 |
+
"total_flos": 1921476294868992.0,
|
| 1950 |
+
"train_batch_size": 2,
|
| 1951 |
+
"trial_name": null,
|
| 1952 |
+
"trial_params": null
|
| 1953 |
+
}
|
training_loss.png
ADDED
|