End of training
Browse files- README.md +2 -1
- all_results.json +8 -0
- train_results.json +8 -0
- trainer_state.json +1197 -0
- training_loss.png +0 -0
README.md
CHANGED
|
@@ -4,6 +4,7 @@ license: apache-2.0
|
|
| 4 |
base_model: Qwen/Qwen2.5-7B
|
| 5 |
tags:
|
| 6 |
- llama-factory
|
|
|
|
| 7 |
- generated_from_trainer
|
| 8 |
model-index:
|
| 9 |
- name: Bespoke-Stratos-17k-v2
|
|
@@ -15,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 15 |
|
| 16 |
# Bespoke-Stratos-17k-v2
|
| 17 |
|
| 18 |
-
This model is a fine-tuned version of [Qwen/Qwen2.5-7B](https://huggingface.co/Qwen/Qwen2.5-7B) on
|
| 19 |
|
| 20 |
## Model description
|
| 21 |
|
|
|
|
| 4 |
base_model: Qwen/Qwen2.5-7B
|
| 5 |
tags:
|
| 6 |
- llama-factory
|
| 7 |
+
- full
|
| 8 |
- generated_from_trainer
|
| 9 |
model-index:
|
| 10 |
- name: Bespoke-Stratos-17k-v2
|
|
|
|
| 16 |
|
| 17 |
# Bespoke-Stratos-17k-v2
|
| 18 |
|
| 19 |
+
This model is a fine-tuned version of [Qwen/Qwen2.5-7B](https://huggingface.co/Qwen/Qwen2.5-7B) on the bespokelabs/Bespoke-Stratos-17k dataset.
|
| 20 |
|
| 21 |
## Model description
|
| 22 |
|
all_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 2.986425339366516,
|
| 3 |
+
"total_flos": 518749773889536.0,
|
| 4 |
+
"train_loss": 0.5635366864276655,
|
| 5 |
+
"train_runtime": 44509.1752,
|
| 6 |
+
"train_samples_per_second": 0.357,
|
| 7 |
+
"train_steps_per_second": 0.004
|
| 8 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 2.986425339366516,
|
| 3 |
+
"total_flos": 518749773889536.0,
|
| 4 |
+
"train_loss": 0.5635366864276655,
|
| 5 |
+
"train_runtime": 44509.1752,
|
| 6 |
+
"train_samples_per_second": 0.357,
|
| 7 |
+
"train_steps_per_second": 0.004
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1197 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 2.986425339366516,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 165,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.01809954751131222,
|
| 13 |
+
"grad_norm": 4.270520950401436,
|
| 14 |
+
"learning_rate": 5.882352941176471e-07,
|
| 15 |
+
"loss": 0.8651,
|
| 16 |
+
"step": 1
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.03619909502262444,
|
| 20 |
+
"grad_norm": 4.08130709590183,
|
| 21 |
+
"learning_rate": 1.1764705882352942e-06,
|
| 22 |
+
"loss": 0.834,
|
| 23 |
+
"step": 2
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.05429864253393665,
|
| 27 |
+
"grad_norm": 6.3327514730586545,
|
| 28 |
+
"learning_rate": 1.7647058823529414e-06,
|
| 29 |
+
"loss": 0.8286,
|
| 30 |
+
"step": 3
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.07239819004524888,
|
| 34 |
+
"grad_norm": 4.619785546975456,
|
| 35 |
+
"learning_rate": 2.3529411764705885e-06,
|
| 36 |
+
"loss": 0.8592,
|
| 37 |
+
"step": 4
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.09049773755656108,
|
| 41 |
+
"grad_norm": 3.2305077950239336,
|
| 42 |
+
"learning_rate": 2.9411764705882355e-06,
|
| 43 |
+
"loss": 0.8714,
|
| 44 |
+
"step": 5
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.1085972850678733,
|
| 48 |
+
"grad_norm": 2.7370616230365483,
|
| 49 |
+
"learning_rate": 3.529411764705883e-06,
|
| 50 |
+
"loss": 0.846,
|
| 51 |
+
"step": 6
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.12669683257918551,
|
| 55 |
+
"grad_norm": 2.0471907108946947,
|
| 56 |
+
"learning_rate": 4.11764705882353e-06,
|
| 57 |
+
"loss": 0.8033,
|
| 58 |
+
"step": 7
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.14479638009049775,
|
| 62 |
+
"grad_norm": 1.7780057810998475,
|
| 63 |
+
"learning_rate": 4.705882352941177e-06,
|
| 64 |
+
"loss": 0.7808,
|
| 65 |
+
"step": 8
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.16289592760180996,
|
| 69 |
+
"grad_norm": 2.3908751757347773,
|
| 70 |
+
"learning_rate": 5.294117647058824e-06,
|
| 71 |
+
"loss": 0.7251,
|
| 72 |
+
"step": 9
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.18099547511312217,
|
| 76 |
+
"grad_norm": 2.25104000061671,
|
| 77 |
+
"learning_rate": 5.882352941176471e-06,
|
| 78 |
+
"loss": 0.756,
|
| 79 |
+
"step": 10
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.19909502262443438,
|
| 83 |
+
"grad_norm": 1.8941739130758755,
|
| 84 |
+
"learning_rate": 6.470588235294119e-06,
|
| 85 |
+
"loss": 0.7291,
|
| 86 |
+
"step": 11
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.2171945701357466,
|
| 90 |
+
"grad_norm": 1.6227195871257627,
|
| 91 |
+
"learning_rate": 7.058823529411766e-06,
|
| 92 |
+
"loss": 0.7187,
|
| 93 |
+
"step": 12
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.23529411764705882,
|
| 97 |
+
"grad_norm": 2.117879729057304,
|
| 98 |
+
"learning_rate": 7.647058823529411e-06,
|
| 99 |
+
"loss": 0.7101,
|
| 100 |
+
"step": 13
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.25339366515837103,
|
| 104 |
+
"grad_norm": 1.6475677575928487,
|
| 105 |
+
"learning_rate": 8.23529411764706e-06,
|
| 106 |
+
"loss": 0.693,
|
| 107 |
+
"step": 14
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.27149321266968324,
|
| 111 |
+
"grad_norm": 1.352033468624444,
|
| 112 |
+
"learning_rate": 8.823529411764707e-06,
|
| 113 |
+
"loss": 0.7142,
|
| 114 |
+
"step": 15
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.2895927601809955,
|
| 118 |
+
"grad_norm": 1.0129577643301186,
|
| 119 |
+
"learning_rate": 9.411764705882354e-06,
|
| 120 |
+
"loss": 0.6941,
|
| 121 |
+
"step": 16
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.3076923076923077,
|
| 125 |
+
"grad_norm": 1.3363060312344113,
|
| 126 |
+
"learning_rate": 1e-05,
|
| 127 |
+
"loss": 0.6946,
|
| 128 |
+
"step": 17
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.3257918552036199,
|
| 132 |
+
"grad_norm": 0.9818113208945979,
|
| 133 |
+
"learning_rate": 9.998873580873848e-06,
|
| 134 |
+
"loss": 0.6559,
|
| 135 |
+
"step": 18
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.3438914027149321,
|
| 139 |
+
"grad_norm": 0.9497100630620063,
|
| 140 |
+
"learning_rate": 9.99549483102341e-06,
|
| 141 |
+
"loss": 0.6293,
|
| 142 |
+
"step": 19
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.36199095022624433,
|
| 146 |
+
"grad_norm": 0.7666994164661116,
|
| 147 |
+
"learning_rate": 9.989865272804064e-06,
|
| 148 |
+
"loss": 0.6363,
|
| 149 |
+
"step": 20
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.38009049773755654,
|
| 153 |
+
"grad_norm": 0.7657473460823022,
|
| 154 |
+
"learning_rate": 9.981987442712634e-06,
|
| 155 |
+
"loss": 0.619,
|
| 156 |
+
"step": 21
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.39819004524886875,
|
| 160 |
+
"grad_norm": 0.7395769869216733,
|
| 161 |
+
"learning_rate": 9.971864890244514e-06,
|
| 162 |
+
"loss": 0.6137,
|
| 163 |
+
"step": 22
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.416289592760181,
|
| 167 |
+
"grad_norm": 0.6202139618434414,
|
| 168 |
+
"learning_rate": 9.959502176294384e-06,
|
| 169 |
+
"loss": 0.6142,
|
| 170 |
+
"step": 23
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.4343891402714932,
|
| 174 |
+
"grad_norm": 0.7479430347437638,
|
| 175 |
+
"learning_rate": 9.944904871101227e-06,
|
| 176 |
+
"loss": 0.618,
|
| 177 |
+
"step": 24
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.45248868778280543,
|
| 181 |
+
"grad_norm": 0.6779505395481803,
|
| 182 |
+
"learning_rate": 9.928079551738542e-06,
|
| 183 |
+
"loss": 0.6177,
|
| 184 |
+
"step": 25
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.47058823529411764,
|
| 188 |
+
"grad_norm": 0.5363485701627213,
|
| 189 |
+
"learning_rate": 9.909033799150947e-06,
|
| 190 |
+
"loss": 0.6323,
|
| 191 |
+
"step": 26
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.48868778280542985,
|
| 195 |
+
"grad_norm": 0.5974955089825978,
|
| 196 |
+
"learning_rate": 9.887776194738433e-06,
|
| 197 |
+
"loss": 0.5826,
|
| 198 |
+
"step": 27
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.5067873303167421,
|
| 202 |
+
"grad_norm": 0.5444695893455697,
|
| 203 |
+
"learning_rate": 9.864316316489873e-06,
|
| 204 |
+
"loss": 0.578,
|
| 205 |
+
"step": 28
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.5248868778280543,
|
| 209 |
+
"grad_norm": 0.4489447311178371,
|
| 210 |
+
"learning_rate": 9.838664734667496e-06,
|
| 211 |
+
"loss": 0.6192,
|
| 212 |
+
"step": 29
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.5429864253393665,
|
| 216 |
+
"grad_norm": 0.5254652960581311,
|
| 217 |
+
"learning_rate": 9.810833007044247e-06,
|
| 218 |
+
"loss": 0.5858,
|
| 219 |
+
"step": 30
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.5610859728506787,
|
| 223 |
+
"grad_norm": 0.43507105065822094,
|
| 224 |
+
"learning_rate": 9.780833673696255e-06,
|
| 225 |
+
"loss": 0.5912,
|
| 226 |
+
"step": 31
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.579185520361991,
|
| 230 |
+
"grad_norm": 0.5884456720678961,
|
| 231 |
+
"learning_rate": 9.74868025135266e-06,
|
| 232 |
+
"loss": 0.601,
|
| 233 |
+
"step": 32
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.5972850678733032,
|
| 237 |
+
"grad_norm": 0.446124158858029,
|
| 238 |
+
"learning_rate": 9.714387227305422e-06,
|
| 239 |
+
"loss": 0.5716,
|
| 240 |
+
"step": 33
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.6153846153846154,
|
| 244 |
+
"grad_norm": 0.4999259943183043,
|
| 245 |
+
"learning_rate": 9.677970052881811e-06,
|
| 246 |
+
"loss": 0.6004,
|
| 247 |
+
"step": 34
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.6334841628959276,
|
| 251 |
+
"grad_norm": 0.5452059230819422,
|
| 252 |
+
"learning_rate": 9.639445136482549e-06,
|
| 253 |
+
"loss": 0.5771,
|
| 254 |
+
"step": 35
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.6515837104072398,
|
| 258 |
+
"grad_norm": 0.45682271068627184,
|
| 259 |
+
"learning_rate": 9.598829836188694e-06,
|
| 260 |
+
"loss": 0.5837,
|
| 261 |
+
"step": 36
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.669683257918552,
|
| 265 |
+
"grad_norm": 0.43493826579652684,
|
| 266 |
+
"learning_rate": 9.55614245194068e-06,
|
| 267 |
+
"loss": 0.5963,
|
| 268 |
+
"step": 37
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.6877828054298643,
|
| 272 |
+
"grad_norm": 0.42543550395893537,
|
| 273 |
+
"learning_rate": 9.511402217292927e-06,
|
| 274 |
+
"loss": 0.5633,
|
| 275 |
+
"step": 38
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.7058823529411765,
|
| 279 |
+
"grad_norm": 0.3891549600693992,
|
| 280 |
+
"learning_rate": 9.464629290747844e-06,
|
| 281 |
+
"loss": 0.5607,
|
| 282 |
+
"step": 39
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.7239819004524887,
|
| 286 |
+
"grad_norm": 0.4002335108145771,
|
| 287 |
+
"learning_rate": 9.415844746673047e-06,
|
| 288 |
+
"loss": 0.5778,
|
| 289 |
+
"step": 40
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.7420814479638009,
|
| 293 |
+
"grad_norm": 0.4287499076319389,
|
| 294 |
+
"learning_rate": 9.365070565805941e-06,
|
| 295 |
+
"loss": 0.5649,
|
| 296 |
+
"step": 41
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.7601809954751131,
|
| 300 |
+
"grad_norm": 0.38942600340952016,
|
| 301 |
+
"learning_rate": 9.312329625349903e-06,
|
| 302 |
+
"loss": 0.5701,
|
| 303 |
+
"step": 42
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.7782805429864253,
|
| 307 |
+
"grad_norm": 0.3306475208382961,
|
| 308 |
+
"learning_rate": 9.257645688666557e-06,
|
| 309 |
+
"loss": 0.5641,
|
| 310 |
+
"step": 43
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.7963800904977375,
|
| 314 |
+
"grad_norm": 0.37084164933852937,
|
| 315 |
+
"learning_rate": 9.201043394568773e-06,
|
| 316 |
+
"loss": 0.5511,
|
| 317 |
+
"step": 44
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.8144796380090498,
|
| 321 |
+
"grad_norm": 0.3494129880763865,
|
| 322 |
+
"learning_rate": 9.142548246219212e-06,
|
| 323 |
+
"loss": 0.5626,
|
| 324 |
+
"step": 45
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.832579185520362,
|
| 328 |
+
"grad_norm": 0.36906169844507203,
|
| 329 |
+
"learning_rate": 9.082186599639429e-06,
|
| 330 |
+
"loss": 0.5659,
|
| 331 |
+
"step": 46
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.8506787330316742,
|
| 335 |
+
"grad_norm": 0.3758901401117832,
|
| 336 |
+
"learning_rate": 9.019985651834703e-06,
|
| 337 |
+
"loss": 0.572,
|
| 338 |
+
"step": 47
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.8687782805429864,
|
| 342 |
+
"grad_norm": 0.32671844293334396,
|
| 343 |
+
"learning_rate": 8.955973428539943e-06,
|
| 344 |
+
"loss": 0.5755,
|
| 345 |
+
"step": 48
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.8868778280542986,
|
| 349 |
+
"grad_norm": 0.4489028958210117,
|
| 350 |
+
"learning_rate": 8.890178771592198e-06,
|
| 351 |
+
"loss": 0.578,
|
| 352 |
+
"step": 49
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.9049773755656109,
|
| 356 |
+
"grad_norm": 0.39698421743018597,
|
| 357 |
+
"learning_rate": 8.822631325935463e-06,
|
| 358 |
+
"loss": 0.5742,
|
| 359 |
+
"step": 50
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.9230769230769231,
|
| 363 |
+
"grad_norm": 0.40912494828005824,
|
| 364 |
+
"learning_rate": 8.753361526263622e-06,
|
| 365 |
+
"loss": 0.5581,
|
| 366 |
+
"step": 51
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.9411764705882353,
|
| 370 |
+
"grad_norm": 0.46734953420703257,
|
| 371 |
+
"learning_rate": 8.682400583307562e-06,
|
| 372 |
+
"loss": 0.5684,
|
| 373 |
+
"step": 52
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.9592760180995475,
|
| 377 |
+
"grad_norm": 0.3467889011356855,
|
| 378 |
+
"learning_rate": 8.609780469772623e-06,
|
| 379 |
+
"loss": 0.556,
|
| 380 |
+
"step": 53
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.9773755656108597,
|
| 384 |
+
"grad_norm": 0.5085559424990912,
|
| 385 |
+
"learning_rate": 8.535533905932739e-06,
|
| 386 |
+
"loss": 0.5514,
|
| 387 |
+
"step": 54
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.995475113122172,
|
| 391 |
+
"grad_norm": 0.4523513658790744,
|
| 392 |
+
"learning_rate": 8.459694344887732e-06,
|
| 393 |
+
"loss": 0.5925,
|
| 394 |
+
"step": 55
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 1.0135746606334841,
|
| 398 |
+
"grad_norm": 0.7340348802970464,
|
| 399 |
+
"learning_rate": 8.382295957490435e-06,
|
| 400 |
+
"loss": 0.9574,
|
| 401 |
+
"step": 56
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 1.0316742081447963,
|
| 405 |
+
"grad_norm": 0.420046446649172,
|
| 406 |
+
"learning_rate": 8.303373616950408e-06,
|
| 407 |
+
"loss": 0.5356,
|
| 408 |
+
"step": 57
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 1.0497737556561086,
|
| 412 |
+
"grad_norm": 0.3421282037573979,
|
| 413 |
+
"learning_rate": 8.222962883121196e-06,
|
| 414 |
+
"loss": 0.5521,
|
| 415 |
+
"step": 58
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 1.0678733031674208,
|
| 419 |
+
"grad_norm": 0.4190729595247935,
|
| 420 |
+
"learning_rate": 8.141099986478212e-06,
|
| 421 |
+
"loss": 0.5362,
|
| 422 |
+
"step": 59
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 1.085972850678733,
|
| 426 |
+
"grad_norm": 0.359499883552521,
|
| 427 |
+
"learning_rate": 8.057821811794457e-06,
|
| 428 |
+
"loss": 0.541,
|
| 429 |
+
"step": 60
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 1.1040723981900453,
|
| 433 |
+
"grad_norm": 0.3756476194471776,
|
| 434 |
+
"learning_rate": 7.973165881521435e-06,
|
| 435 |
+
"loss": 0.5467,
|
| 436 |
+
"step": 61
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 1.1221719457013575,
|
| 440 |
+
"grad_norm": 0.35661012462013447,
|
| 441 |
+
"learning_rate": 7.887170338882742e-06,
|
| 442 |
+
"loss": 0.5462,
|
| 443 |
+
"step": 62
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 1.1402714932126696,
|
| 447 |
+
"grad_norm": 0.3498507388990045,
|
| 448 |
+
"learning_rate": 7.799873930687979e-06,
|
| 449 |
+
"loss": 0.5171,
|
| 450 |
+
"step": 63
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 1.1583710407239818,
|
| 454 |
+
"grad_norm": 0.3170076180552228,
|
| 455 |
+
"learning_rate": 7.711315989874677e-06,
|
| 456 |
+
"loss": 0.507,
|
| 457 |
+
"step": 64
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 1.1764705882352942,
|
| 461 |
+
"grad_norm": 0.3037529085890902,
|
| 462 |
+
"learning_rate": 7.621536417786159e-06,
|
| 463 |
+
"loss": 0.5174,
|
| 464 |
+
"step": 65
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 1.1945701357466063,
|
| 468 |
+
"grad_norm": 0.342570759484934,
|
| 469 |
+
"learning_rate": 7.530575666193283e-06,
|
| 470 |
+
"loss": 0.5317,
|
| 471 |
+
"step": 66
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 1.2126696832579185,
|
| 475 |
+
"grad_norm": 0.30104088055723055,
|
| 476 |
+
"learning_rate": 7.438474719068174e-06,
|
| 477 |
+
"loss": 0.5274,
|
| 478 |
+
"step": 67
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 1.2307692307692308,
|
| 482 |
+
"grad_norm": 0.326669875074583,
|
| 483 |
+
"learning_rate": 7.3452750741181855e-06,
|
| 484 |
+
"loss": 0.5135,
|
| 485 |
+
"step": 68
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 1.248868778280543,
|
| 489 |
+
"grad_norm": 0.3551023399546242,
|
| 490 |
+
"learning_rate": 7.251018724088367e-06,
|
| 491 |
+
"loss": 0.5204,
|
| 492 |
+
"step": 69
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 1.2669683257918551,
|
| 496 |
+
"grad_norm": 0.37895672243853146,
|
| 497 |
+
"learning_rate": 7.155748137840892e-06,
|
| 498 |
+
"loss": 0.5479,
|
| 499 |
+
"step": 70
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 1.2850678733031673,
|
| 503 |
+
"grad_norm": 0.36133670714468175,
|
| 504 |
+
"learning_rate": 7.059506241219964e-06,
|
| 505 |
+
"loss": 0.5434,
|
| 506 |
+
"step": 71
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 1.3031674208144797,
|
| 510 |
+
"grad_norm": 0.33317369601816094,
|
| 511 |
+
"learning_rate": 6.962336397710819e-06,
|
| 512 |
+
"loss": 0.5253,
|
| 513 |
+
"step": 72
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 1.3212669683257918,
|
| 517 |
+
"grad_norm": 0.28803695913966876,
|
| 518 |
+
"learning_rate": 6.864282388901544e-06,
|
| 519 |
+
"loss": 0.5076,
|
| 520 |
+
"step": 73
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 1.3393665158371042,
|
| 524 |
+
"grad_norm": 0.3913327961715048,
|
| 525 |
+
"learning_rate": 6.765388394756504e-06,
|
| 526 |
+
"loss": 0.5365,
|
| 527 |
+
"step": 74
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 1.3574660633484164,
|
| 531 |
+
"grad_norm": 0.2977675871279601,
|
| 532 |
+
"learning_rate": 6.665698973710289e-06,
|
| 533 |
+
"loss": 0.5393,
|
| 534 |
+
"step": 75
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 1.3755656108597285,
|
| 538 |
+
"grad_norm": 0.3702071945183516,
|
| 539 |
+
"learning_rate": 6.565259042591112e-06,
|
| 540 |
+
"loss": 0.517,
|
| 541 |
+
"step": 76
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 1.3936651583710407,
|
| 545 |
+
"grad_norm": 0.35663025248537605,
|
| 546 |
+
"learning_rate": 6.464113856382752e-06,
|
| 547 |
+
"loss": 0.5218,
|
| 548 |
+
"step": 77
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 1.4117647058823528,
|
| 552 |
+
"grad_norm": 0.3846006097732817,
|
| 553 |
+
"learning_rate": 6.3623089878341146e-06,
|
| 554 |
+
"loss": 0.5281,
|
| 555 |
+
"step": 78
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 1.4298642533936652,
|
| 559 |
+
"grad_norm": 0.38145233359354735,
|
| 560 |
+
"learning_rate": 6.259890306925627e-06,
|
| 561 |
+
"loss": 0.5494,
|
| 562 |
+
"step": 79
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 1.4479638009049773,
|
| 566 |
+
"grad_norm": 0.321686326326639,
|
| 567 |
+
"learning_rate": 6.156903960201709e-06,
|
| 568 |
+
"loss": 0.5215,
|
| 569 |
+
"step": 80
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 1.4660633484162897,
|
| 573 |
+
"grad_norm": 0.3553248299309718,
|
| 574 |
+
"learning_rate": 6.053396349978632e-06,
|
| 575 |
+
"loss": 0.5124,
|
| 576 |
+
"step": 81
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 1.4841628959276019,
|
| 580 |
+
"grad_norm": 0.3688316984975195,
|
| 581 |
+
"learning_rate": 5.949414113437142e-06,
|
| 582 |
+
"loss": 0.5297,
|
| 583 |
+
"step": 82
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 1.502262443438914,
|
| 587 |
+
"grad_norm": 0.31268728687027997,
|
| 588 |
+
"learning_rate": 5.8450041016092465e-06,
|
| 589 |
+
"loss": 0.5395,
|
| 590 |
+
"step": 83
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 1.5203619909502262,
|
| 594 |
+
"grad_norm": 0.36390895213403174,
|
| 595 |
+
"learning_rate": 5.740213358268658e-06,
|
| 596 |
+
"loss": 0.5247,
|
| 597 |
+
"step": 84
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 1.5384615384615383,
|
| 601 |
+
"grad_norm": 0.3532101356565584,
|
| 602 |
+
"learning_rate": 5.635089098734394e-06,
|
| 603 |
+
"loss": 0.5121,
|
| 604 |
+
"step": 85
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 1.5565610859728507,
|
| 608 |
+
"grad_norm": 0.30616495012538836,
|
| 609 |
+
"learning_rate": 5.529678688597081e-06,
|
| 610 |
+
"loss": 0.5301,
|
| 611 |
+
"step": 86
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 1.5746606334841629,
|
| 615 |
+
"grad_norm": 0.41168813574652363,
|
| 616 |
+
"learning_rate": 5.4240296223775465e-06,
|
| 617 |
+
"loss": 0.5297,
|
| 618 |
+
"step": 87
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 1.5927601809954752,
|
| 622 |
+
"grad_norm": 0.3480419851531957,
|
| 623 |
+
"learning_rate": 5.318189502127332e-06,
|
| 624 |
+
"loss": 0.5346,
|
| 625 |
+
"step": 88
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 1.6108597285067874,
|
| 629 |
+
"grad_norm": 0.3951211138988914,
|
| 630 |
+
"learning_rate": 5.212206015980742e-06,
|
| 631 |
+
"loss": 0.5442,
|
| 632 |
+
"step": 89
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 1.6289592760180995,
|
| 636 |
+
"grad_norm": 0.32142900262246665,
|
| 637 |
+
"learning_rate": 5.106126916668118e-06,
|
| 638 |
+
"loss": 0.5139,
|
| 639 |
+
"step": 90
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 1.6470588235294117,
|
| 643 |
+
"grad_norm": 0.372934660354091,
|
| 644 |
+
"learning_rate": 5e-06,
|
| 645 |
+
"loss": 0.537,
|
| 646 |
+
"step": 91
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 1.6651583710407238,
|
| 650 |
+
"grad_norm": 0.31940542344772543,
|
| 651 |
+
"learning_rate": 4.8938730833318825e-06,
|
| 652 |
+
"loss": 0.5255,
|
| 653 |
+
"step": 92
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 1.6832579185520362,
|
| 657 |
+
"grad_norm": 0.3484072513306141,
|
| 658 |
+
"learning_rate": 4.78779398401926e-06,
|
| 659 |
+
"loss": 0.5312,
|
| 660 |
+
"step": 93
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 1.7013574660633484,
|
| 664 |
+
"grad_norm": 0.305688116719012,
|
| 665 |
+
"learning_rate": 4.6818104978726685e-06,
|
| 666 |
+
"loss": 0.5404,
|
| 667 |
+
"step": 94
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 1.7194570135746607,
|
| 671 |
+
"grad_norm": 0.2610608476826038,
|
| 672 |
+
"learning_rate": 4.575970377622456e-06,
|
| 673 |
+
"loss": 0.5308,
|
| 674 |
+
"step": 95
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 1.737556561085973,
|
| 678 |
+
"grad_norm": 0.323315649537396,
|
| 679 |
+
"learning_rate": 4.47032131140292e-06,
|
| 680 |
+
"loss": 0.5129,
|
| 681 |
+
"step": 96
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 1.755656108597285,
|
| 685 |
+
"grad_norm": 0.2868605984916918,
|
| 686 |
+
"learning_rate": 4.364910901265607e-06,
|
| 687 |
+
"loss": 0.5324,
|
| 688 |
+
"step": 97
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 1.7737556561085972,
|
| 692 |
+
"grad_norm": 0.2480631165360626,
|
| 693 |
+
"learning_rate": 4.259786641731344e-06,
|
| 694 |
+
"loss": 0.5204,
|
| 695 |
+
"step": 98
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 1.7918552036199094,
|
| 699 |
+
"grad_norm": 0.2845617409603977,
|
| 700 |
+
"learning_rate": 4.154995898390756e-06,
|
| 701 |
+
"loss": 0.5207,
|
| 702 |
+
"step": 99
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 1.8099547511312217,
|
| 706 |
+
"grad_norm": 0.2901557749018309,
|
| 707 |
+
"learning_rate": 4.050585886562858e-06,
|
| 708 |
+
"loss": 0.5401,
|
| 709 |
+
"step": 100
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 1.8280542986425339,
|
| 713 |
+
"grad_norm": 0.2577789829212818,
|
| 714 |
+
"learning_rate": 3.94660365002137e-06,
|
| 715 |
+
"loss": 0.525,
|
| 716 |
+
"step": 101
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 1.8461538461538463,
|
| 720 |
+
"grad_norm": 0.28060392539372403,
|
| 721 |
+
"learning_rate": 3.843096039798293e-06,
|
| 722 |
+
"loss": 0.5267,
|
| 723 |
+
"step": 102
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 1.8642533936651584,
|
| 727 |
+
"grad_norm": 0.27493292579589645,
|
| 728 |
+
"learning_rate": 3.7401096930743753e-06,
|
| 729 |
+
"loss": 0.5225,
|
| 730 |
+
"step": 103
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 1.8823529411764706,
|
| 734 |
+
"grad_norm": 0.2503147317440557,
|
| 735 |
+
"learning_rate": 3.6376910121658867e-06,
|
| 736 |
+
"loss": 0.5183,
|
| 737 |
+
"step": 104
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 1.9004524886877827,
|
| 741 |
+
"grad_norm": 0.29883473530138427,
|
| 742 |
+
"learning_rate": 3.5358861436172487e-06,
|
| 743 |
+
"loss": 0.5338,
|
| 744 |
+
"step": 105
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 1.9185520361990949,
|
| 748 |
+
"grad_norm": 0.3003751720185607,
|
| 749 |
+
"learning_rate": 3.4347409574088896e-06,
|
| 750 |
+
"loss": 0.5248,
|
| 751 |
+
"step": 106
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 1.9366515837104072,
|
| 755 |
+
"grad_norm": 0.296798930565972,
|
| 756 |
+
"learning_rate": 3.3343010262897125e-06,
|
| 757 |
+
"loss": 0.5172,
|
| 758 |
+
"step": 107
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 1.9547511312217196,
|
| 762 |
+
"grad_norm": 0.29960841020898205,
|
| 763 |
+
"learning_rate": 3.234611605243496e-06,
|
| 764 |
+
"loss": 0.5416,
|
| 765 |
+
"step": 108
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 1.9728506787330318,
|
| 769 |
+
"grad_norm": 0.24656341443926255,
|
| 770 |
+
"learning_rate": 3.1357176110984578e-06,
|
| 771 |
+
"loss": 0.5169,
|
| 772 |
+
"step": 109
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 1.990950226244344,
|
| 776 |
+
"grad_norm": 0.24416915079855595,
|
| 777 |
+
"learning_rate": 3.0376636022891813e-06,
|
| 778 |
+
"loss": 0.5031,
|
| 779 |
+
"step": 110
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 2.009049773755656,
|
| 783 |
+
"grad_norm": 0.5106478330596561,
|
| 784 |
+
"learning_rate": 2.9404937587800374e-06,
|
| 785 |
+
"loss": 0.8733,
|
| 786 |
+
"step": 111
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 2.0271493212669682,
|
| 790 |
+
"grad_norm": 0.28895565841398857,
|
| 791 |
+
"learning_rate": 2.8442518621591085e-06,
|
| 792 |
+
"loss": 0.5102,
|
| 793 |
+
"step": 112
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 2.0452488687782804,
|
| 797 |
+
"grad_norm": 0.274250226920425,
|
| 798 |
+
"learning_rate": 2.748981275911633e-06,
|
| 799 |
+
"loss": 0.4991,
|
| 800 |
+
"step": 113
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 2.0633484162895925,
|
| 804 |
+
"grad_norm": 0.2683303919910382,
|
| 805 |
+
"learning_rate": 2.6547249258818162e-06,
|
| 806 |
+
"loss": 0.4811,
|
| 807 |
+
"step": 114
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 2.081447963800905,
|
| 811 |
+
"grad_norm": 0.2670782977498998,
|
| 812 |
+
"learning_rate": 2.5615252809318287e-06,
|
| 813 |
+
"loss": 0.5246,
|
| 814 |
+
"step": 115
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 2.0995475113122173,
|
| 818 |
+
"grad_norm": 0.25059501980164506,
|
| 819 |
+
"learning_rate": 2.469424333806718e-06,
|
| 820 |
+
"loss": 0.5181,
|
| 821 |
+
"step": 116
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 2.1176470588235294,
|
| 825 |
+
"grad_norm": 0.28254518452392197,
|
| 826 |
+
"learning_rate": 2.3784635822138424e-06,
|
| 827 |
+
"loss": 0.5112,
|
| 828 |
+
"step": 117
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 2.1357466063348416,
|
| 832 |
+
"grad_norm": 0.29324248893783117,
|
| 833 |
+
"learning_rate": 2.288684010125325e-06,
|
| 834 |
+
"loss": 0.5114,
|
| 835 |
+
"step": 118
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 2.1538461538461537,
|
| 839 |
+
"grad_norm": 0.2503268575452895,
|
| 840 |
+
"learning_rate": 2.2001260693120236e-06,
|
| 841 |
+
"loss": 0.4967,
|
| 842 |
+
"step": 119
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 2.171945701357466,
|
| 846 |
+
"grad_norm": 0.2699479211486462,
|
| 847 |
+
"learning_rate": 2.1128296611172593e-06,
|
| 848 |
+
"loss": 0.4959,
|
| 849 |
+
"step": 120
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 2.1900452488687785,
|
| 853 |
+
"grad_norm": 0.2580177426665245,
|
| 854 |
+
"learning_rate": 2.0268341184785674e-06,
|
| 855 |
+
"loss": 0.5088,
|
| 856 |
+
"step": 121
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 2.2081447963800906,
|
| 860 |
+
"grad_norm": 0.2508244895204738,
|
| 861 |
+
"learning_rate": 1.9421781882055447e-06,
|
| 862 |
+
"loss": 0.4917,
|
| 863 |
+
"step": 122
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 2.226244343891403,
|
| 867 |
+
"grad_norm": 0.2792571659281552,
|
| 868 |
+
"learning_rate": 1.8589000135217882e-06,
|
| 869 |
+
"loss": 0.5023,
|
| 870 |
+
"step": 123
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 2.244343891402715,
|
| 874 |
+
"grad_norm": 0.23158505061350954,
|
| 875 |
+
"learning_rate": 1.7770371168788042e-06,
|
| 876 |
+
"loss": 0.4863,
|
| 877 |
+
"step": 124
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 2.262443438914027,
|
| 881 |
+
"grad_norm": 0.2683239496999239,
|
| 882 |
+
"learning_rate": 1.6966263830495939e-06,
|
| 883 |
+
"loss": 0.4873,
|
| 884 |
+
"step": 125
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 2.2805429864253393,
|
| 888 |
+
"grad_norm": 0.24178359031260568,
|
| 889 |
+
"learning_rate": 1.6177040425095664e-06,
|
| 890 |
+
"loss": 0.4977,
|
| 891 |
+
"step": 126
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 2.2986425339366514,
|
| 895 |
+
"grad_norm": 0.2542786308206507,
|
| 896 |
+
"learning_rate": 1.5403056551122697e-06,
|
| 897 |
+
"loss": 0.4998,
|
| 898 |
+
"step": 127
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 2.3167420814479636,
|
| 902 |
+
"grad_norm": 0.2345038790062461,
|
| 903 |
+
"learning_rate": 1.4644660940672628e-06,
|
| 904 |
+
"loss": 0.5061,
|
| 905 |
+
"step": 128
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 2.334841628959276,
|
| 909 |
+
"grad_norm": 0.2494018788793317,
|
| 910 |
+
"learning_rate": 1.390219530227378e-06,
|
| 911 |
+
"loss": 0.5169,
|
| 912 |
+
"step": 129
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 2.3529411764705883,
|
| 916 |
+
"grad_norm": 0.23720925343998778,
|
| 917 |
+
"learning_rate": 1.3175994166924394e-06,
|
| 918 |
+
"loss": 0.5023,
|
| 919 |
+
"step": 130
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 2.3710407239819005,
|
| 923 |
+
"grad_norm": 0.20440215406384443,
|
| 924 |
+
"learning_rate": 1.246638473736378e-06,
|
| 925 |
+
"loss": 0.4753,
|
| 926 |
+
"step": 131
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 2.3891402714932126,
|
| 930 |
+
"grad_norm": 0.21449535787548765,
|
| 931 |
+
"learning_rate": 1.1773686740645384e-06,
|
| 932 |
+
"loss": 0.5136,
|
| 933 |
+
"step": 132
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 2.4072398190045248,
|
| 937 |
+
"grad_norm": 0.24588391549876998,
|
| 938 |
+
"learning_rate": 1.1098212284078037e-06,
|
| 939 |
+
"loss": 0.5009,
|
| 940 |
+
"step": 133
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 2.425339366515837,
|
| 944 |
+
"grad_norm": 0.2388777734423218,
|
| 945 |
+
"learning_rate": 1.0440265714600573e-06,
|
| 946 |
+
"loss": 0.5246,
|
| 947 |
+
"step": 134
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 2.4434389140271495,
|
| 951 |
+
"grad_norm": 0.23579449100780014,
|
| 952 |
+
"learning_rate": 9.80014348165298e-07,
|
| 953 |
+
"loss": 0.5026,
|
| 954 |
+
"step": 135
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 2.4615384615384617,
|
| 958 |
+
"grad_norm": 0.22823611772484245,
|
| 959 |
+
"learning_rate": 9.178134003605721e-07,
|
| 960 |
+
"loss": 0.492,
|
| 961 |
+
"step": 136
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 2.479638009049774,
|
| 965 |
+
"grad_norm": 0.21111741055317057,
|
| 966 |
+
"learning_rate": 8.574517537807897e-07,
|
| 967 |
+
"loss": 0.5105,
|
| 968 |
+
"step": 137
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 2.497737556561086,
|
| 972 |
+
"grad_norm": 0.23249058681576157,
|
| 973 |
+
"learning_rate": 7.989566054312286e-07,
|
| 974 |
+
"loss": 0.5143,
|
| 975 |
+
"step": 138
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 2.515837104072398,
|
| 979 |
+
"grad_norm": 0.21764828402899594,
|
| 980 |
+
"learning_rate": 7.423543113334436e-07,
|
| 981 |
+
"loss": 0.4849,
|
| 982 |
+
"step": 139
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 2.5339366515837103,
|
| 986 |
+
"grad_norm": 0.24396341429275906,
|
| 987 |
+
"learning_rate": 6.876703746500984e-07,
|
| 988 |
+
"loss": 0.4903,
|
| 989 |
+
"step": 140
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 2.5520361990950224,
|
| 993 |
+
"grad_norm": 0.22499941187789713,
|
| 994 |
+
"learning_rate": 6.349294341940593e-07,
|
| 995 |
+
"loss": 0.4772,
|
| 996 |
+
"step": 141
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 2.5701357466063346,
|
| 1000 |
+
"grad_norm": 0.23288553371110793,
|
| 1001 |
+
"learning_rate": 5.841552533269534e-07,
|
| 1002 |
+
"loss": 0.5162,
|
| 1003 |
+
"step": 142
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 2.588235294117647,
|
| 1007 |
+
"grad_norm": 0.205460117436211,
|
| 1008 |
+
"learning_rate": 5.353707092521581e-07,
|
| 1009 |
+
"loss": 0.4976,
|
| 1010 |
+
"step": 143
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 2.6063348416289593,
|
| 1014 |
+
"grad_norm": 0.2214846135707125,
|
| 1015 |
+
"learning_rate": 4.885977827070748e-07,
|
| 1016 |
+
"loss": 0.5116,
|
| 1017 |
+
"step": 144
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 2.6244343891402715,
|
| 1021 |
+
"grad_norm": 0.22509550835701087,
|
| 1022 |
+
"learning_rate": 4.43857548059321e-07,
|
| 1023 |
+
"loss": 0.5115,
|
| 1024 |
+
"step": 145
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 2.6425339366515836,
|
| 1028 |
+
"grad_norm": 0.2041589989166428,
|
| 1029 |
+
"learning_rate": 4.0117016381130636e-07,
|
| 1030 |
+
"loss": 0.4798,
|
| 1031 |
+
"step": 146
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 2.660633484162896,
|
| 1035 |
+
"grad_norm": 0.22016570643213332,
|
| 1036 |
+
"learning_rate": 3.6055486351745327e-07,
|
| 1037 |
+
"loss": 0.4998,
|
| 1038 |
+
"step": 147
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 2.6787330316742084,
|
| 1042 |
+
"grad_norm": 0.2478568518194293,
|
| 1043 |
+
"learning_rate": 3.220299471181898e-07,
|
| 1044 |
+
"loss": 0.5039,
|
| 1045 |
+
"step": 148
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 2.6968325791855206,
|
| 1049 |
+
"grad_norm": 0.20717209753663476,
|
| 1050 |
+
"learning_rate": 2.85612772694579e-07,
|
| 1051 |
+
"loss": 0.4928,
|
| 1052 |
+
"step": 149
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 2.7149321266968327,
|
| 1056 |
+
"grad_norm": 0.21806938834032513,
|
| 1057 |
+
"learning_rate": 2.5131974864734063e-07,
|
| 1058 |
+
"loss": 0.5131,
|
| 1059 |
+
"step": 150
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 2.733031674208145,
|
| 1063 |
+
"grad_norm": 0.21612495922499686,
|
| 1064 |
+
"learning_rate": 2.1916632630374579e-07,
|
| 1065 |
+
"loss": 0.5145,
|
| 1066 |
+
"step": 151
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 2.751131221719457,
|
| 1070 |
+
"grad_norm": 0.21635896851957898,
|
| 1071 |
+
"learning_rate": 1.8916699295575324e-07,
|
| 1072 |
+
"loss": 0.4964,
|
| 1073 |
+
"step": 152
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 2.769230769230769,
|
| 1077 |
+
"grad_norm": 0.19798724526464564,
|
| 1078 |
+
"learning_rate": 1.6133526533250566e-07,
|
| 1079 |
+
"loss": 0.4855,
|
| 1080 |
+
"step": 153
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 2.7873303167420813,
|
| 1084 |
+
"grad_norm": 0.20783663967394034,
|
| 1085 |
+
"learning_rate": 1.3568368351012718e-07,
|
| 1086 |
+
"loss": 0.4966,
|
| 1087 |
+
"step": 154
|
| 1088 |
+
},
|
| 1089 |
+
{
|
| 1090 |
+
"epoch": 2.8054298642533935,
|
| 1091 |
+
"grad_norm": 0.19191009162032222,
|
| 1092 |
+
"learning_rate": 1.1222380526156929e-07,
|
| 1093 |
+
"loss": 0.4974,
|
| 1094 |
+
"step": 155
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 2.8235294117647056,
|
| 1098 |
+
"grad_norm": 0.21216955411007024,
|
| 1099 |
+
"learning_rate": 9.096620084905472e-08,
|
| 1100 |
+
"loss": 0.5191,
|
| 1101 |
+
"step": 156
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 2.841628959276018,
|
| 1105 |
+
"grad_norm": 0.21639489053110292,
|
| 1106 |
+
"learning_rate": 7.192044826145772e-08,
|
| 1107 |
+
"loss": 0.5268,
|
| 1108 |
+
"step": 157
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"epoch": 2.8597285067873304,
|
| 1112 |
+
"grad_norm": 0.21244774609292955,
|
| 1113 |
+
"learning_rate": 5.509512889877333e-08,
|
| 1114 |
+
"loss": 0.5088,
|
| 1115 |
+
"step": 158
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 2.8778280542986425,
|
| 1119 |
+
"grad_norm": 0.21215837031481655,
|
| 1120 |
+
"learning_rate": 4.0497823705615836e-08,
|
| 1121 |
+
"loss": 0.4942,
|
| 1122 |
+
"step": 159
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 2.8959276018099547,
|
| 1126 |
+
"grad_norm": 0.2070665647344241,
|
| 1127 |
+
"learning_rate": 2.8135109755487723e-08,
|
| 1128 |
+
"loss": 0.5331,
|
| 1129 |
+
"step": 160
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 2.914027149321267,
|
| 1133 |
+
"grad_norm": 0.19955418175982154,
|
| 1134 |
+
"learning_rate": 1.8012557287367394e-08,
|
| 1135 |
+
"loss": 0.4901,
|
| 1136 |
+
"step": 161
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 2.9321266968325794,
|
| 1140 |
+
"grad_norm": 0.2052841051806031,
|
| 1141 |
+
"learning_rate": 1.0134727195937332e-08,
|
| 1142 |
+
"loss": 0.5096,
|
| 1143 |
+
"step": 162
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 2.9502262443438916,
|
| 1147 |
+
"grad_norm": 0.22007204293809482,
|
| 1148 |
+
"learning_rate": 4.505168976592922e-09,
|
| 1149 |
+
"loss": 0.5248,
|
| 1150 |
+
"step": 163
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 2.9683257918552037,
|
| 1154 |
+
"grad_norm": 0.21247616301770453,
|
| 1155 |
+
"learning_rate": 1.1264191261528557e-09,
|
| 1156 |
+
"loss": 0.5081,
|
| 1157 |
+
"step": 164
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 2.986425339366516,
|
| 1161 |
+
"grad_norm": 0.20780388742772332,
|
| 1162 |
+
"learning_rate": 0.0,
|
| 1163 |
+
"loss": 0.5126,
|
| 1164 |
+
"step": 165
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 2.986425339366516,
|
| 1168 |
+
"step": 165,
|
| 1169 |
+
"total_flos": 518749773889536.0,
|
| 1170 |
+
"train_loss": 0.5635366864276655,
|
| 1171 |
+
"train_runtime": 44509.1752,
|
| 1172 |
+
"train_samples_per_second": 0.357,
|
| 1173 |
+
"train_steps_per_second": 0.004
|
| 1174 |
+
}
|
| 1175 |
+
],
|
| 1176 |
+
"logging_steps": 1.0,
|
| 1177 |
+
"max_steps": 165,
|
| 1178 |
+
"num_input_tokens_seen": 0,
|
| 1179 |
+
"num_train_epochs": 3,
|
| 1180 |
+
"save_steps": 100,
|
| 1181 |
+
"stateful_callbacks": {
|
| 1182 |
+
"TrainerControl": {
|
| 1183 |
+
"args": {
|
| 1184 |
+
"should_epoch_stop": false,
|
| 1185 |
+
"should_evaluate": false,
|
| 1186 |
+
"should_log": false,
|
| 1187 |
+
"should_save": true,
|
| 1188 |
+
"should_training_stop": true
|
| 1189 |
+
},
|
| 1190 |
+
"attributes": {}
|
| 1191 |
+
}
|
| 1192 |
+
},
|
| 1193 |
+
"total_flos": 518749773889536.0,
|
| 1194 |
+
"train_batch_size": 1,
|
| 1195 |
+
"trial_name": null,
|
| 1196 |
+
"trial_params": null
|
| 1197 |
+
}
|
training_loss.png
ADDED
|