End of training
Browse files- README.md +2 -1
- all_results.json +8 -0
- train_results.json +8 -0
- trainer_state.json +2247 -0
- training_loss.png +0 -0
README.md
CHANGED
|
@@ -4,6 +4,7 @@ license: apache-2.0
|
|
| 4 |
base_model: Qwen/Qwen2.5-32B-Instruct
|
| 5 |
tags:
|
| 6 |
- llama-factory
|
|
|
|
| 7 |
- generated_from_trainer
|
| 8 |
model-index:
|
| 9 |
- name: s1K_32b
|
|
@@ -15,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 15 |
|
| 16 |
# s1K_32b
|
| 17 |
|
| 18 |
-
This model is a fine-tuned version of [Qwen/Qwen2.5-32B-Instruct](https://huggingface.co/Qwen/Qwen2.5-32B-Instruct) on
|
| 19 |
|
| 20 |
## Model description
|
| 21 |
|
|
|
|
| 4 |
base_model: Qwen/Qwen2.5-32B-Instruct
|
| 5 |
tags:
|
| 6 |
- llama-factory
|
| 7 |
+
- full
|
| 8 |
- generated_from_trainer
|
| 9 |
model-index:
|
| 10 |
- name: s1K_32b
|
|
|
|
| 16 |
|
| 17 |
# s1K_32b
|
| 18 |
|
| 19 |
+
This model is a fine-tuned version of [Qwen/Qwen2.5-32B-Instruct](https://huggingface.co/Qwen/Qwen2.5-32B-Instruct) on the mlfoundations-dev/s1K_reformat dataset.
|
| 20 |
|
| 21 |
## Model description
|
| 22 |
|
all_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 5.0,
|
| 3 |
+
"total_flos": 32588508020736.0,
|
| 4 |
+
"train_loss": 0.11869781007546754,
|
| 5 |
+
"train_runtime": 5355.6377,
|
| 6 |
+
"train_samples_per_second": 0.934,
|
| 7 |
+
"train_steps_per_second": 0.059
|
| 8 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 5.0,
|
| 3 |
+
"total_flos": 32588508020736.0,
|
| 4 |
+
"train_loss": 0.11869781007546754,
|
| 5 |
+
"train_runtime": 5355.6377,
|
| 6 |
+
"train_samples_per_second": 0.934,
|
| 7 |
+
"train_steps_per_second": 0.059
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,2247 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 5.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 315,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.015873015873015872,
|
| 13 |
+
"grad_norm": 3.3795859813690186,
|
| 14 |
+
"learning_rate": 6.25e-07,
|
| 15 |
+
"loss": 0.4037,
|
| 16 |
+
"step": 1
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.031746031746031744,
|
| 20 |
+
"grad_norm": 4.176924228668213,
|
| 21 |
+
"learning_rate": 1.25e-06,
|
| 22 |
+
"loss": 0.6101,
|
| 23 |
+
"step": 2
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.047619047619047616,
|
| 27 |
+
"grad_norm": 3.2537522315979004,
|
| 28 |
+
"learning_rate": 1.8750000000000003e-06,
|
| 29 |
+
"loss": 0.5335,
|
| 30 |
+
"step": 3
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.06349206349206349,
|
| 34 |
+
"grad_norm": 3.8701913356781006,
|
| 35 |
+
"learning_rate": 2.5e-06,
|
| 36 |
+
"loss": 0.5156,
|
| 37 |
+
"step": 4
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.07936507936507936,
|
| 41 |
+
"grad_norm": 3.2381300926208496,
|
| 42 |
+
"learning_rate": 3.125e-06,
|
| 43 |
+
"loss": 0.5435,
|
| 44 |
+
"step": 5
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.09523809523809523,
|
| 48 |
+
"grad_norm": 2.6569769382476807,
|
| 49 |
+
"learning_rate": 3.7500000000000005e-06,
|
| 50 |
+
"loss": 0.4349,
|
| 51 |
+
"step": 6
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.1111111111111111,
|
| 55 |
+
"grad_norm": 2.641036033630371,
|
| 56 |
+
"learning_rate": 4.3750000000000005e-06,
|
| 57 |
+
"loss": 0.4657,
|
| 58 |
+
"step": 7
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.12698412698412698,
|
| 62 |
+
"grad_norm": 2.500192880630493,
|
| 63 |
+
"learning_rate": 5e-06,
|
| 64 |
+
"loss": 0.387,
|
| 65 |
+
"step": 8
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.14285714285714285,
|
| 69 |
+
"grad_norm": 2.925560474395752,
|
| 70 |
+
"learning_rate": 5.625e-06,
|
| 71 |
+
"loss": 0.5708,
|
| 72 |
+
"step": 9
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.15873015873015872,
|
| 76 |
+
"grad_norm": 1.740651249885559,
|
| 77 |
+
"learning_rate": 6.25e-06,
|
| 78 |
+
"loss": 0.4166,
|
| 79 |
+
"step": 10
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.1746031746031746,
|
| 83 |
+
"grad_norm": 1.8194835186004639,
|
| 84 |
+
"learning_rate": 6.875e-06,
|
| 85 |
+
"loss": 0.4392,
|
| 86 |
+
"step": 11
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.19047619047619047,
|
| 90 |
+
"grad_norm": 1.2873449325561523,
|
| 91 |
+
"learning_rate": 7.500000000000001e-06,
|
| 92 |
+
"loss": 0.3224,
|
| 93 |
+
"step": 12
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.20634920634920634,
|
| 97 |
+
"grad_norm": 1.4529649019241333,
|
| 98 |
+
"learning_rate": 8.125000000000001e-06,
|
| 99 |
+
"loss": 0.3407,
|
| 100 |
+
"step": 13
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.2222222222222222,
|
| 104 |
+
"grad_norm": 1.3093940019607544,
|
| 105 |
+
"learning_rate": 8.750000000000001e-06,
|
| 106 |
+
"loss": 0.3677,
|
| 107 |
+
"step": 14
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.23809523809523808,
|
| 111 |
+
"grad_norm": 1.4765881299972534,
|
| 112 |
+
"learning_rate": 9.375000000000001e-06,
|
| 113 |
+
"loss": 0.3792,
|
| 114 |
+
"step": 15
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.25396825396825395,
|
| 118 |
+
"grad_norm": 5.271663665771484,
|
| 119 |
+
"learning_rate": 1e-05,
|
| 120 |
+
"loss": 0.6346,
|
| 121 |
+
"step": 16
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.2698412698412698,
|
| 125 |
+
"grad_norm": 1.3494993448257446,
|
| 126 |
+
"learning_rate": 9.999724009977419e-06,
|
| 127 |
+
"loss": 0.4433,
|
| 128 |
+
"step": 17
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.2857142857142857,
|
| 132 |
+
"grad_norm": 1.115939736366272,
|
| 133 |
+
"learning_rate": 9.998896070377873e-06,
|
| 134 |
+
"loss": 0.2864,
|
| 135 |
+
"step": 18
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.30158730158730157,
|
| 139 |
+
"grad_norm": 1.5331807136535645,
|
| 140 |
+
"learning_rate": 9.99751627260259e-06,
|
| 141 |
+
"loss": 0.3423,
|
| 142 |
+
"step": 19
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.31746031746031744,
|
| 146 |
+
"grad_norm": 1.5603549480438232,
|
| 147 |
+
"learning_rate": 9.995584768975735e-06,
|
| 148 |
+
"loss": 0.3799,
|
| 149 |
+
"step": 20
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.3333333333333333,
|
| 153 |
+
"grad_norm": 1.4752039909362793,
|
| 154 |
+
"learning_rate": 9.993101772727602e-06,
|
| 155 |
+
"loss": 0.3565,
|
| 156 |
+
"step": 21
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.3492063492063492,
|
| 160 |
+
"grad_norm": 9.61251449584961,
|
| 161 |
+
"learning_rate": 9.990067557971068e-06,
|
| 162 |
+
"loss": 0.7244,
|
| 163 |
+
"step": 22
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.36507936507936506,
|
| 167 |
+
"grad_norm": 1.3895500898361206,
|
| 168 |
+
"learning_rate": 9.986482459671332e-06,
|
| 169 |
+
"loss": 0.3444,
|
| 170 |
+
"step": 23
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.38095238095238093,
|
| 174 |
+
"grad_norm": 1.231410026550293,
|
| 175 |
+
"learning_rate": 9.982346873608936e-06,
|
| 176 |
+
"loss": 0.3221,
|
| 177 |
+
"step": 24
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.3968253968253968,
|
| 181 |
+
"grad_norm": 1.9909180402755737,
|
| 182 |
+
"learning_rate": 9.977661256336081e-06,
|
| 183 |
+
"loss": 0.3724,
|
| 184 |
+
"step": 25
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.4126984126984127,
|
| 188 |
+
"grad_norm": 1.2321062088012695,
|
| 189 |
+
"learning_rate": 9.972426125126208e-06,
|
| 190 |
+
"loss": 0.3391,
|
| 191 |
+
"step": 26
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.42857142857142855,
|
| 195 |
+
"grad_norm": 5.4099321365356445,
|
| 196 |
+
"learning_rate": 9.966642057916915e-06,
|
| 197 |
+
"loss": 0.5806,
|
| 198 |
+
"step": 27
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.4444444444444444,
|
| 202 |
+
"grad_norm": 1.0613089799880981,
|
| 203 |
+
"learning_rate": 9.960309693246135e-06,
|
| 204 |
+
"loss": 0.2237,
|
| 205 |
+
"step": 28
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.4603174603174603,
|
| 209 |
+
"grad_norm": 1.1249053478240967,
|
| 210 |
+
"learning_rate": 9.953429730181653e-06,
|
| 211 |
+
"loss": 0.3618,
|
| 212 |
+
"step": 29
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.47619047619047616,
|
| 216 |
+
"grad_norm": 1.0634913444519043,
|
| 217 |
+
"learning_rate": 9.94600292824394e-06,
|
| 218 |
+
"loss": 0.3177,
|
| 219 |
+
"step": 30
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.49206349206349204,
|
| 223 |
+
"grad_norm": 1.1851016283035278,
|
| 224 |
+
"learning_rate": 9.938030107322284e-06,
|
| 225 |
+
"loss": 0.2799,
|
| 226 |
+
"step": 31
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.5079365079365079,
|
| 230 |
+
"grad_norm": 1.2439275979995728,
|
| 231 |
+
"learning_rate": 9.929512147584297e-06,
|
| 232 |
+
"loss": 0.3415,
|
| 233 |
+
"step": 32
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.5238095238095238,
|
| 237 |
+
"grad_norm": 1.1060391664505005,
|
| 238 |
+
"learning_rate": 9.920449989378741e-06,
|
| 239 |
+
"loss": 0.2888,
|
| 240 |
+
"step": 33
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.5396825396825397,
|
| 244 |
+
"grad_norm": 1.4267876148223877,
|
| 245 |
+
"learning_rate": 9.910844633131712e-06,
|
| 246 |
+
"loss": 0.3724,
|
| 247 |
+
"step": 34
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.5555555555555556,
|
| 251 |
+
"grad_norm": 1.1967415809631348,
|
| 252 |
+
"learning_rate": 9.90069713923621e-06,
|
| 253 |
+
"loss": 0.3362,
|
| 254 |
+
"step": 35
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.5714285714285714,
|
| 258 |
+
"grad_norm": 0.9548850059509277,
|
| 259 |
+
"learning_rate": 9.890008627935057e-06,
|
| 260 |
+
"loss": 0.2776,
|
| 261 |
+
"step": 36
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.5873015873015873,
|
| 265 |
+
"grad_norm": 1.1262502670288086,
|
| 266 |
+
"learning_rate": 9.878780279197246e-06,
|
| 267 |
+
"loss": 0.3215,
|
| 268 |
+
"step": 37
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.6031746031746031,
|
| 272 |
+
"grad_norm": 0.9589006304740906,
|
| 273 |
+
"learning_rate": 9.867013332587667e-06,
|
| 274 |
+
"loss": 0.2871,
|
| 275 |
+
"step": 38
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.6190476190476191,
|
| 279 |
+
"grad_norm": 1.2175616025924683,
|
| 280 |
+
"learning_rate": 9.854709087130261e-06,
|
| 281 |
+
"loss": 0.2672,
|
| 282 |
+
"step": 39
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.6349206349206349,
|
| 286 |
+
"grad_norm": 1.1927433013916016,
|
| 287 |
+
"learning_rate": 9.841868901164621e-06,
|
| 288 |
+
"loss": 0.3167,
|
| 289 |
+
"step": 40
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.6507936507936508,
|
| 293 |
+
"grad_norm": 1.1496555805206299,
|
| 294 |
+
"learning_rate": 9.828494192196037e-06,
|
| 295 |
+
"loss": 0.331,
|
| 296 |
+
"step": 41
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.6666666666666666,
|
| 300 |
+
"grad_norm": 2.9185266494750977,
|
| 301 |
+
"learning_rate": 9.814586436738998e-06,
|
| 302 |
+
"loss": 0.4792,
|
| 303 |
+
"step": 42
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.6825396825396826,
|
| 307 |
+
"grad_norm": 4.013699054718018,
|
| 308 |
+
"learning_rate": 9.8001471701542e-06,
|
| 309 |
+
"loss": 0.3316,
|
| 310 |
+
"step": 43
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.6984126984126984,
|
| 314 |
+
"grad_norm": 1.1543551683425903,
|
| 315 |
+
"learning_rate": 9.785177986479049e-06,
|
| 316 |
+
"loss": 0.2954,
|
| 317 |
+
"step": 44
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.7142857142857143,
|
| 321 |
+
"grad_norm": 0.9748827219009399,
|
| 322 |
+
"learning_rate": 9.76968053825168e-06,
|
| 323 |
+
"loss": 0.2267,
|
| 324 |
+
"step": 45
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.7301587301587301,
|
| 328 |
+
"grad_norm": 1.0915194749832153,
|
| 329 |
+
"learning_rate": 9.753656536328529e-06,
|
| 330 |
+
"loss": 0.3005,
|
| 331 |
+
"step": 46
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.746031746031746,
|
| 335 |
+
"grad_norm": 2.678406238555908,
|
| 336 |
+
"learning_rate": 9.737107749695456e-06,
|
| 337 |
+
"loss": 0.2828,
|
| 338 |
+
"step": 47
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.7619047619047619,
|
| 342 |
+
"grad_norm": 0.9532033801078796,
|
| 343 |
+
"learning_rate": 9.72003600527246e-06,
|
| 344 |
+
"loss": 0.2658,
|
| 345 |
+
"step": 48
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.7777777777777778,
|
| 349 |
+
"grad_norm": 1.1294094324111938,
|
| 350 |
+
"learning_rate": 9.702443187711991e-06,
|
| 351 |
+
"loss": 0.2978,
|
| 352 |
+
"step": 49
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.7936507936507936,
|
| 356 |
+
"grad_norm": 1.067834734916687,
|
| 357 |
+
"learning_rate": 9.6843312391909e-06,
|
| 358 |
+
"loss": 0.2941,
|
| 359 |
+
"step": 50
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.8095238095238095,
|
| 363 |
+
"grad_norm": 1.0761404037475586,
|
| 364 |
+
"learning_rate": 9.665702159196014e-06,
|
| 365 |
+
"loss": 0.3056,
|
| 366 |
+
"step": 51
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.8253968253968254,
|
| 370 |
+
"grad_norm": 4.403201103210449,
|
| 371 |
+
"learning_rate": 9.646558004303419e-06,
|
| 372 |
+
"loss": 0.3728,
|
| 373 |
+
"step": 52
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.8412698412698413,
|
| 377 |
+
"grad_norm": 1.0028859376907349,
|
| 378 |
+
"learning_rate": 9.62690088795141e-06,
|
| 379 |
+
"loss": 0.2765,
|
| 380 |
+
"step": 53
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.8571428571428571,
|
| 384 |
+
"grad_norm": 1.0655666589736938,
|
| 385 |
+
"learning_rate": 9.606732980207186e-06,
|
| 386 |
+
"loss": 0.2912,
|
| 387 |
+
"step": 54
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.873015873015873,
|
| 391 |
+
"grad_norm": 8.566058158874512,
|
| 392 |
+
"learning_rate": 9.586056507527266e-06,
|
| 393 |
+
"loss": 0.5695,
|
| 394 |
+
"step": 55
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.8888888888888888,
|
| 398 |
+
"grad_norm": 1.487505555152893,
|
| 399 |
+
"learning_rate": 9.564873752511719e-06,
|
| 400 |
+
"loss": 0.2906,
|
| 401 |
+
"step": 56
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.9047619047619048,
|
| 405 |
+
"grad_norm": 2.458684206008911,
|
| 406 |
+
"learning_rate": 9.543187053652156e-06,
|
| 407 |
+
"loss": 0.4637,
|
| 408 |
+
"step": 57
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.9206349206349206,
|
| 412 |
+
"grad_norm": 1.1918795108795166,
|
| 413 |
+
"learning_rate": 9.520998805073583e-06,
|
| 414 |
+
"loss": 0.347,
|
| 415 |
+
"step": 58
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.9365079365079365,
|
| 419 |
+
"grad_norm": 1.9880917072296143,
|
| 420 |
+
"learning_rate": 9.498311456270091e-06,
|
| 421 |
+
"loss": 0.3115,
|
| 422 |
+
"step": 59
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.9523809523809523,
|
| 426 |
+
"grad_norm": 1.1867977380752563,
|
| 427 |
+
"learning_rate": 9.475127511834438e-06,
|
| 428 |
+
"loss": 0.3172,
|
| 429 |
+
"step": 60
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.9682539682539683,
|
| 433 |
+
"grad_norm": 1.2388020753860474,
|
| 434 |
+
"learning_rate": 9.451449531181571e-06,
|
| 435 |
+
"loss": 0.252,
|
| 436 |
+
"step": 61
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.9841269841269841,
|
| 440 |
+
"grad_norm": 9.597868919372559,
|
| 441 |
+
"learning_rate": 9.427280128266049e-06,
|
| 442 |
+
"loss": 0.2668,
|
| 443 |
+
"step": 62
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 1.0,
|
| 447 |
+
"grad_norm": 0.9280756711959839,
|
| 448 |
+
"learning_rate": 9.4026219712935e-06,
|
| 449 |
+
"loss": 0.2107,
|
| 450 |
+
"step": 63
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 1.0158730158730158,
|
| 454 |
+
"grad_norm": 0.9878024458885193,
|
| 455 |
+
"learning_rate": 9.377477782426041e-06,
|
| 456 |
+
"loss": 0.2008,
|
| 457 |
+
"step": 64
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 1.0317460317460316,
|
| 461 |
+
"grad_norm": 2.081068992614746,
|
| 462 |
+
"learning_rate": 9.351850337481774e-06,
|
| 463 |
+
"loss": 0.1998,
|
| 464 |
+
"step": 65
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 1.0476190476190477,
|
| 468 |
+
"grad_norm": 0.9242327213287354,
|
| 469 |
+
"learning_rate": 9.325742465628342e-06,
|
| 470 |
+
"loss": 0.1791,
|
| 471 |
+
"step": 66
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 1.0634920634920635,
|
| 475 |
+
"grad_norm": 2.085343599319458,
|
| 476 |
+
"learning_rate": 9.299157049070604e-06,
|
| 477 |
+
"loss": 0.1787,
|
| 478 |
+
"step": 67
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 1.0793650793650793,
|
| 482 |
+
"grad_norm": 0.815376877784729,
|
| 483 |
+
"learning_rate": 9.272097022732444e-06,
|
| 484 |
+
"loss": 0.1678,
|
| 485 |
+
"step": 68
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 1.0952380952380953,
|
| 489 |
+
"grad_norm": 0.8281353116035461,
|
| 490 |
+
"learning_rate": 9.244565373932775e-06,
|
| 491 |
+
"loss": 0.1646,
|
| 492 |
+
"step": 69
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 1.1111111111111112,
|
| 496 |
+
"grad_norm": 0.828342616558075,
|
| 497 |
+
"learning_rate": 9.216565142055745e-06,
|
| 498 |
+
"loss": 0.155,
|
| 499 |
+
"step": 70
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 1.126984126984127,
|
| 503 |
+
"grad_norm": 0.77997225522995,
|
| 504 |
+
"learning_rate": 9.188099418215208e-06,
|
| 505 |
+
"loss": 0.1176,
|
| 506 |
+
"step": 71
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 1.1428571428571428,
|
| 510 |
+
"grad_norm": 0.9490588903427124,
|
| 511 |
+
"learning_rate": 9.159171344913469e-06,
|
| 512 |
+
"loss": 0.1316,
|
| 513 |
+
"step": 72
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 1.1587301587301586,
|
| 517 |
+
"grad_norm": 1.061501145362854,
|
| 518 |
+
"learning_rate": 9.129784115694368e-06,
|
| 519 |
+
"loss": 0.1759,
|
| 520 |
+
"step": 73
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 1.1746031746031746,
|
| 524 |
+
"grad_norm": 1.077555537223816,
|
| 525 |
+
"learning_rate": 9.09994097479073e-06,
|
| 526 |
+
"loss": 0.1607,
|
| 527 |
+
"step": 74
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 1.1904761904761905,
|
| 531 |
+
"grad_norm": 0.8213773965835571,
|
| 532 |
+
"learning_rate": 9.069645216766207e-06,
|
| 533 |
+
"loss": 0.1123,
|
| 534 |
+
"step": 75
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 1.2063492063492063,
|
| 538 |
+
"grad_norm": 0.8540403246879578,
|
| 539 |
+
"learning_rate": 9.038900186151574e-06,
|
| 540 |
+
"loss": 0.1306,
|
| 541 |
+
"step": 76
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 1.2222222222222223,
|
| 545 |
+
"grad_norm": 1.0110580921173096,
|
| 546 |
+
"learning_rate": 9.007709277075512e-06,
|
| 547 |
+
"loss": 0.1644,
|
| 548 |
+
"step": 77
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 1.2380952380952381,
|
| 552 |
+
"grad_norm": 1.3765249252319336,
|
| 553 |
+
"learning_rate": 8.976075932889896e-06,
|
| 554 |
+
"loss": 0.2027,
|
| 555 |
+
"step": 78
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 1.253968253968254,
|
| 559 |
+
"grad_norm": 1.0460267066955566,
|
| 560 |
+
"learning_rate": 8.944003645789678e-06,
|
| 561 |
+
"loss": 0.1635,
|
| 562 |
+
"step": 79
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 1.2698412698412698,
|
| 566 |
+
"grad_norm": 0.8925233483314514,
|
| 567 |
+
"learning_rate": 8.911495956427358e-06,
|
| 568 |
+
"loss": 0.1538,
|
| 569 |
+
"step": 80
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 1.2857142857142856,
|
| 573 |
+
"grad_norm": 1.1790034770965576,
|
| 574 |
+
"learning_rate": 8.8785564535221e-06,
|
| 575 |
+
"loss": 0.164,
|
| 576 |
+
"step": 81
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 1.3015873015873016,
|
| 580 |
+
"grad_norm": 1.1329326629638672,
|
| 581 |
+
"learning_rate": 8.845188773463567e-06,
|
| 582 |
+
"loss": 0.1449,
|
| 583 |
+
"step": 82
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 1.3174603174603174,
|
| 587 |
+
"grad_norm": 4.996532440185547,
|
| 588 |
+
"learning_rate": 8.811396599910467e-06,
|
| 589 |
+
"loss": 0.247,
|
| 590 |
+
"step": 83
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 1.3333333333333333,
|
| 594 |
+
"grad_norm": 0.74964839220047,
|
| 595 |
+
"learning_rate": 8.777183663383897e-06,
|
| 596 |
+
"loss": 0.1259,
|
| 597 |
+
"step": 84
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 1.3492063492063493,
|
| 601 |
+
"grad_norm": 1.8605432510375977,
|
| 602 |
+
"learning_rate": 8.742553740855507e-06,
|
| 603 |
+
"loss": 0.1935,
|
| 604 |
+
"step": 85
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 1.3650793650793651,
|
| 608 |
+
"grad_norm": 0.9615695476531982,
|
| 609 |
+
"learning_rate": 8.707510655330536e-06,
|
| 610 |
+
"loss": 0.1772,
|
| 611 |
+
"step": 86
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 1.380952380952381,
|
| 615 |
+
"grad_norm": 1.0028129816055298,
|
| 616 |
+
"learning_rate": 8.672058275425773e-06,
|
| 617 |
+
"loss": 0.159,
|
| 618 |
+
"step": 87
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 1.3968253968253967,
|
| 622 |
+
"grad_norm": 0.780433177947998,
|
| 623 |
+
"learning_rate": 8.636200514942466e-06,
|
| 624 |
+
"loss": 0.1374,
|
| 625 |
+
"step": 88
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 1.4126984126984126,
|
| 629 |
+
"grad_norm": 2.116007089614868,
|
| 630 |
+
"learning_rate": 8.59994133243427e-06,
|
| 631 |
+
"loss": 0.1951,
|
| 632 |
+
"step": 89
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 1.4285714285714286,
|
| 636 |
+
"grad_norm": 0.8867003917694092,
|
| 637 |
+
"learning_rate": 8.563284730770222e-06,
|
| 638 |
+
"loss": 0.147,
|
| 639 |
+
"step": 90
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 1.4444444444444444,
|
| 643 |
+
"grad_norm": 0.8089895248413086,
|
| 644 |
+
"learning_rate": 8.52623475669285e-06,
|
| 645 |
+
"loss": 0.1109,
|
| 646 |
+
"step": 91
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 1.4603174603174602,
|
| 650 |
+
"grad_norm": 1.093917727470398,
|
| 651 |
+
"learning_rate": 8.488795500371427e-06,
|
| 652 |
+
"loss": 0.1328,
|
| 653 |
+
"step": 92
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 1.4761904761904763,
|
| 657 |
+
"grad_norm": 1.0122655630111694,
|
| 658 |
+
"learning_rate": 8.450971094950433e-06,
|
| 659 |
+
"loss": 0.1998,
|
| 660 |
+
"step": 93
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 1.492063492063492,
|
| 664 |
+
"grad_norm": 3.2237043380737305,
|
| 665 |
+
"learning_rate": 8.412765716093273e-06,
|
| 666 |
+
"loss": 0.1949,
|
| 667 |
+
"step": 94
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 1.507936507936508,
|
| 671 |
+
"grad_norm": 0.8921442031860352,
|
| 672 |
+
"learning_rate": 8.374183581521288e-06,
|
| 673 |
+
"loss": 0.1659,
|
| 674 |
+
"step": 95
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 1.5238095238095237,
|
| 678 |
+
"grad_norm": 0.9087255001068115,
|
| 679 |
+
"learning_rate": 8.335228950548164e-06,
|
| 680 |
+
"loss": 0.1381,
|
| 681 |
+
"step": 96
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 1.5396825396825395,
|
| 685 |
+
"grad_norm": 0.8095484972000122,
|
| 686 |
+
"learning_rate": 8.29590612360969e-06,
|
| 687 |
+
"loss": 0.1418,
|
| 688 |
+
"step": 97
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 1.5555555555555556,
|
| 692 |
+
"grad_norm": 0.9046949744224548,
|
| 693 |
+
"learning_rate": 8.256219441789023e-06,
|
| 694 |
+
"loss": 0.1362,
|
| 695 |
+
"step": 98
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 1.5714285714285714,
|
| 699 |
+
"grad_norm": 1.0229334831237793,
|
| 700 |
+
"learning_rate": 8.216173286337449e-06,
|
| 701 |
+
"loss": 0.1548,
|
| 702 |
+
"step": 99
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 1.5873015873015874,
|
| 706 |
+
"grad_norm": 0.7955384254455566,
|
| 707 |
+
"learning_rate": 8.175772078190706e-06,
|
| 708 |
+
"loss": 0.0964,
|
| 709 |
+
"step": 100
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 1.6031746031746033,
|
| 713 |
+
"grad_norm": 0.8258505463600159,
|
| 714 |
+
"learning_rate": 8.135020277480933e-06,
|
| 715 |
+
"loss": 0.1436,
|
| 716 |
+
"step": 101
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 1.619047619047619,
|
| 720 |
+
"grad_norm": 0.9820398688316345,
|
| 721 |
+
"learning_rate": 8.093922383044293e-06,
|
| 722 |
+
"loss": 0.1875,
|
| 723 |
+
"step": 102
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 1.6349206349206349,
|
| 727 |
+
"grad_norm": 0.8524972796440125,
|
| 728 |
+
"learning_rate": 8.052482931924307e-06,
|
| 729 |
+
"loss": 0.1381,
|
| 730 |
+
"step": 103
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 1.6507936507936507,
|
| 734 |
+
"grad_norm": 1.515363097190857,
|
| 735 |
+
"learning_rate": 8.010706498870997e-06,
|
| 736 |
+
"loss": 0.1783,
|
| 737 |
+
"step": 104
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 1.6666666666666665,
|
| 741 |
+
"grad_norm": 0.9057204127311707,
|
| 742 |
+
"learning_rate": 7.968597695835845e-06,
|
| 743 |
+
"loss": 0.1491,
|
| 744 |
+
"step": 105
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 1.6825396825396826,
|
| 748 |
+
"grad_norm": 0.73180091381073,
|
| 749 |
+
"learning_rate": 7.926161171462647e-06,
|
| 750 |
+
"loss": 0.1216,
|
| 751 |
+
"step": 106
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 1.6984126984126984,
|
| 755 |
+
"grad_norm": 1.321354627609253,
|
| 756 |
+
"learning_rate": 7.883401610574338e-06,
|
| 757 |
+
"loss": 0.1496,
|
| 758 |
+
"step": 107
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 1.7142857142857144,
|
| 762 |
+
"grad_norm": 0.9403553009033203,
|
| 763 |
+
"learning_rate": 7.84032373365578e-06,
|
| 764 |
+
"loss": 0.1509,
|
| 765 |
+
"step": 108
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 1.7301587301587302,
|
| 769 |
+
"grad_norm": 0.9378424286842346,
|
| 770 |
+
"learning_rate": 7.796932296332666e-06,
|
| 771 |
+
"loss": 0.16,
|
| 772 |
+
"step": 109
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 1.746031746031746,
|
| 776 |
+
"grad_norm": 0.8329445123672485,
|
| 777 |
+
"learning_rate": 7.753232088846505e-06,
|
| 778 |
+
"loss": 0.0972,
|
| 779 |
+
"step": 110
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 1.7619047619047619,
|
| 783 |
+
"grad_norm": 1.2565405368804932,
|
| 784 |
+
"learning_rate": 7.709227935525796e-06,
|
| 785 |
+
"loss": 0.1837,
|
| 786 |
+
"step": 111
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 1.7777777777777777,
|
| 790 |
+
"grad_norm": 0.9211390614509583,
|
| 791 |
+
"learning_rate": 7.664924694253444e-06,
|
| 792 |
+
"loss": 0.1472,
|
| 793 |
+
"step": 112
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 1.7936507936507935,
|
| 797 |
+
"grad_norm": 0.8232191801071167,
|
| 798 |
+
"learning_rate": 7.620327255930475e-06,
|
| 799 |
+
"loss": 0.123,
|
| 800 |
+
"step": 113
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 1.8095238095238095,
|
| 804 |
+
"grad_norm": 0.8965691328048706,
|
| 805 |
+
"learning_rate": 7.575440543936092e-06,
|
| 806 |
+
"loss": 0.1329,
|
| 807 |
+
"step": 114
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 1.8253968253968254,
|
| 811 |
+
"grad_norm": 0.9043343663215637,
|
| 812 |
+
"learning_rate": 7.530269513584158e-06,
|
| 813 |
+
"loss": 0.1268,
|
| 814 |
+
"step": 115
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 1.8412698412698414,
|
| 818 |
+
"grad_norm": 0.822960376739502,
|
| 819 |
+
"learning_rate": 7.484819151576148e-06,
|
| 820 |
+
"loss": 0.128,
|
| 821 |
+
"step": 116
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 1.8571428571428572,
|
| 825 |
+
"grad_norm": 1.0179648399353027,
|
| 826 |
+
"learning_rate": 7.439094475450638e-06,
|
| 827 |
+
"loss": 0.1732,
|
| 828 |
+
"step": 117
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 1.873015873015873,
|
| 832 |
+
"grad_norm": 0.888350784778595,
|
| 833 |
+
"learning_rate": 7.393100533029383e-06,
|
| 834 |
+
"loss": 0.1608,
|
| 835 |
+
"step": 118
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 1.8888888888888888,
|
| 839 |
+
"grad_norm": 0.8138338923454285,
|
| 840 |
+
"learning_rate": 7.346842401860069e-06,
|
| 841 |
+
"loss": 0.1534,
|
| 842 |
+
"step": 119
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 1.9047619047619047,
|
| 846 |
+
"grad_norm": 1.0096592903137207,
|
| 847 |
+
"learning_rate": 7.300325188655762e-06,
|
| 848 |
+
"loss": 0.1705,
|
| 849 |
+
"step": 120
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 1.9206349206349205,
|
| 853 |
+
"grad_norm": 0.8139647841453552,
|
| 854 |
+
"learning_rate": 7.253554028731149e-06,
|
| 855 |
+
"loss": 0.1453,
|
| 856 |
+
"step": 121
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 1.9365079365079365,
|
| 860 |
+
"grad_norm": 1.51543128490448,
|
| 861 |
+
"learning_rate": 7.206534085435626e-06,
|
| 862 |
+
"loss": 0.2008,
|
| 863 |
+
"step": 122
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 1.9523809523809523,
|
| 867 |
+
"grad_norm": 0.944955587387085,
|
| 868 |
+
"learning_rate": 7.159270549583278e-06,
|
| 869 |
+
"loss": 0.1793,
|
| 870 |
+
"step": 123
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 1.9682539682539684,
|
| 874 |
+
"grad_norm": 0.943600594997406,
|
| 875 |
+
"learning_rate": 7.111768638879834e-06,
|
| 876 |
+
"loss": 0.1783,
|
| 877 |
+
"step": 124
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 1.9841269841269842,
|
| 881 |
+
"grad_norm": 0.8988723158836365,
|
| 882 |
+
"learning_rate": 7.064033597346658e-06,
|
| 883 |
+
"loss": 0.1472,
|
| 884 |
+
"step": 125
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 2.0,
|
| 888 |
+
"grad_norm": 0.8375394940376282,
|
| 889 |
+
"learning_rate": 7.016070694741824e-06,
|
| 890 |
+
"loss": 0.0907,
|
| 891 |
+
"step": 126
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 2.015873015873016,
|
| 895 |
+
"grad_norm": 0.5929964780807495,
|
| 896 |
+
"learning_rate": 6.967885225978366e-06,
|
| 897 |
+
"loss": 0.0541,
|
| 898 |
+
"step": 127
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 2.0317460317460316,
|
| 902 |
+
"grad_norm": 0.7823465466499329,
|
| 903 |
+
"learning_rate": 6.919482510539723e-06,
|
| 904 |
+
"loss": 0.0553,
|
| 905 |
+
"step": 128
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 2.0476190476190474,
|
| 909 |
+
"grad_norm": 0.6394754648208618,
|
| 910 |
+
"learning_rate": 6.870867891892511e-06,
|
| 911 |
+
"loss": 0.0539,
|
| 912 |
+
"step": 129
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 2.0634920634920633,
|
| 916 |
+
"grad_norm": 0.6169945001602173,
|
| 917 |
+
"learning_rate": 6.822046736896607e-06,
|
| 918 |
+
"loss": 0.0446,
|
| 919 |
+
"step": 130
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 2.0793650793650795,
|
| 923 |
+
"grad_norm": 0.6922060251235962,
|
| 924 |
+
"learning_rate": 6.773024435212678e-06,
|
| 925 |
+
"loss": 0.0474,
|
| 926 |
+
"step": 131
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 2.0952380952380953,
|
| 930 |
+
"grad_norm": 0.6317225098609924,
|
| 931 |
+
"learning_rate": 6.723806398707186e-06,
|
| 932 |
+
"loss": 0.0472,
|
| 933 |
+
"step": 132
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 2.111111111111111,
|
| 937 |
+
"grad_norm": 0.6741769909858704,
|
| 938 |
+
"learning_rate": 6.674398060854931e-06,
|
| 939 |
+
"loss": 0.042,
|
| 940 |
+
"step": 133
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 2.126984126984127,
|
| 944 |
+
"grad_norm": 0.9402458071708679,
|
| 945 |
+
"learning_rate": 6.624804876139227e-06,
|
| 946 |
+
"loss": 0.0647,
|
| 947 |
+
"step": 134
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 2.142857142857143,
|
| 951 |
+
"grad_norm": 0.8518017530441284,
|
| 952 |
+
"learning_rate": 6.57503231944974e-06,
|
| 953 |
+
"loss": 0.0559,
|
| 954 |
+
"step": 135
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 2.1587301587301586,
|
| 958 |
+
"grad_norm": 0.9338895678520203,
|
| 959 |
+
"learning_rate": 6.525085885478088e-06,
|
| 960 |
+
"loss": 0.0482,
|
| 961 |
+
"step": 136
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 2.1746031746031744,
|
| 965 |
+
"grad_norm": 0.8100844025611877,
|
| 966 |
+
"learning_rate": 6.4749710881112485e-06,
|
| 967 |
+
"loss": 0.0599,
|
| 968 |
+
"step": 137
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 2.1904761904761907,
|
| 972 |
+
"grad_norm": 1.0090680122375488,
|
| 973 |
+
"learning_rate": 6.424693459822843e-06,
|
| 974 |
+
"loss": 0.0642,
|
| 975 |
+
"step": 138
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 2.2063492063492065,
|
| 979 |
+
"grad_norm": 0.6279119253158569,
|
| 980 |
+
"learning_rate": 6.374258551062377e-06,
|
| 981 |
+
"loss": 0.0514,
|
| 982 |
+
"step": 139
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 2.2222222222222223,
|
| 986 |
+
"grad_norm": 0.6508628129959106,
|
| 987 |
+
"learning_rate": 6.3236719296424985e-06,
|
| 988 |
+
"loss": 0.0464,
|
| 989 |
+
"step": 140
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 2.238095238095238,
|
| 993 |
+
"grad_norm": 0.586854100227356,
|
| 994 |
+
"learning_rate": 6.272939180124316e-06,
|
| 995 |
+
"loss": 0.0407,
|
| 996 |
+
"step": 141
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 2.253968253968254,
|
| 1000 |
+
"grad_norm": 0.6627777814865112,
|
| 1001 |
+
"learning_rate": 6.222065903200909e-06,
|
| 1002 |
+
"loss": 0.0487,
|
| 1003 |
+
"step": 142
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 2.2698412698412698,
|
| 1007 |
+
"grad_norm": 0.7252594232559204,
|
| 1008 |
+
"learning_rate": 6.171057715079012e-06,
|
| 1009 |
+
"loss": 0.0568,
|
| 1010 |
+
"step": 143
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 2.2857142857142856,
|
| 1014 |
+
"grad_norm": 0.5209100246429443,
|
| 1015 |
+
"learning_rate": 6.119920246859025e-06,
|
| 1016 |
+
"loss": 0.0363,
|
| 1017 |
+
"step": 144
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 2.3015873015873014,
|
| 1021 |
+
"grad_norm": 0.5737756490707397,
|
| 1022 |
+
"learning_rate": 6.068659143913349e-06,
|
| 1023 |
+
"loss": 0.0303,
|
| 1024 |
+
"step": 145
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 2.317460317460317,
|
| 1028 |
+
"grad_norm": 1.3126976490020752,
|
| 1029 |
+
"learning_rate": 6.0172800652631706e-06,
|
| 1030 |
+
"loss": 0.074,
|
| 1031 |
+
"step": 146
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 2.3333333333333335,
|
| 1035 |
+
"grad_norm": 0.6515079140663147,
|
| 1036 |
+
"learning_rate": 5.965788682953717e-06,
|
| 1037 |
+
"loss": 0.047,
|
| 1038 |
+
"step": 147
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 2.3492063492063493,
|
| 1042 |
+
"grad_norm": 0.6837308406829834,
|
| 1043 |
+
"learning_rate": 5.914190681428098e-06,
|
| 1044 |
+
"loss": 0.0354,
|
| 1045 |
+
"step": 148
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 2.365079365079365,
|
| 1049 |
+
"grad_norm": 1.8536609411239624,
|
| 1050 |
+
"learning_rate": 5.862491756899753e-06,
|
| 1051 |
+
"loss": 0.0605,
|
| 1052 |
+
"step": 149
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 2.380952380952381,
|
| 1056 |
+
"grad_norm": 0.7980743646621704,
|
| 1057 |
+
"learning_rate": 5.8106976167236236e-06,
|
| 1058 |
+
"loss": 0.0443,
|
| 1059 |
+
"step": 150
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 2.3968253968253967,
|
| 1063 |
+
"grad_norm": 0.7513721585273743,
|
| 1064 |
+
"learning_rate": 5.758813978766077e-06,
|
| 1065 |
+
"loss": 0.0554,
|
| 1066 |
+
"step": 151
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 2.4126984126984126,
|
| 1070 |
+
"grad_norm": 0.9976074695587158,
|
| 1071 |
+
"learning_rate": 5.706846570773677e-06,
|
| 1072 |
+
"loss": 0.0629,
|
| 1073 |
+
"step": 152
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 2.4285714285714284,
|
| 1077 |
+
"grad_norm": 0.6811009049415588,
|
| 1078 |
+
"learning_rate": 5.654801129740863e-06,
|
| 1079 |
+
"loss": 0.0395,
|
| 1080 |
+
"step": 153
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 2.4444444444444446,
|
| 1084 |
+
"grad_norm": 0.8673996329307556,
|
| 1085 |
+
"learning_rate": 5.6026834012766155e-06,
|
| 1086 |
+
"loss": 0.0517,
|
| 1087 |
+
"step": 154
|
| 1088 |
+
},
|
| 1089 |
+
{
|
| 1090 |
+
"epoch": 2.4603174603174605,
|
| 1091 |
+
"grad_norm": 0.6963090300559998,
|
| 1092 |
+
"learning_rate": 5.550499138970158e-06,
|
| 1093 |
+
"loss": 0.0416,
|
| 1094 |
+
"step": 155
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 2.4761904761904763,
|
| 1098 |
+
"grad_norm": 0.6987242102622986,
|
| 1099 |
+
"learning_rate": 5.4982541037557825e-06,
|
| 1100 |
+
"loss": 0.0416,
|
| 1101 |
+
"step": 156
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 2.492063492063492,
|
| 1105 |
+
"grad_norm": 1.312720775604248,
|
| 1106 |
+
"learning_rate": 5.44595406327687e-06,
|
| 1107 |
+
"loss": 0.0499,
|
| 1108 |
+
"step": 157
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"epoch": 2.507936507936508,
|
| 1112 |
+
"grad_norm": 0.7223400473594666,
|
| 1113 |
+
"learning_rate": 5.393604791249158e-06,
|
| 1114 |
+
"loss": 0.0617,
|
| 1115 |
+
"step": 158
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 2.5238095238095237,
|
| 1119 |
+
"grad_norm": 0.5593965649604797,
|
| 1120 |
+
"learning_rate": 5.341212066823356e-06,
|
| 1121 |
+
"loss": 0.0341,
|
| 1122 |
+
"step": 159
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 2.5396825396825395,
|
| 1126 |
+
"grad_norm": 0.80050128698349,
|
| 1127 |
+
"learning_rate": 5.288781673947143e-06,
|
| 1128 |
+
"loss": 0.0818,
|
| 1129 |
+
"step": 160
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 2.5555555555555554,
|
| 1133 |
+
"grad_norm": 0.9423879981040955,
|
| 1134 |
+
"learning_rate": 5.2363194007266435e-06,
|
| 1135 |
+
"loss": 0.0533,
|
| 1136 |
+
"step": 161
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 2.571428571428571,
|
| 1140 |
+
"grad_norm": 0.5583639740943909,
|
| 1141 |
+
"learning_rate": 5.183831038787449e-06,
|
| 1142 |
+
"loss": 0.0308,
|
| 1143 |
+
"step": 162
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 2.5873015873015874,
|
| 1147 |
+
"grad_norm": 0.9041386842727661,
|
| 1148 |
+
"learning_rate": 5.131322382635236e-06,
|
| 1149 |
+
"loss": 0.0393,
|
| 1150 |
+
"step": 163
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 2.6031746031746033,
|
| 1154 |
+
"grad_norm": 0.7514568567276001,
|
| 1155 |
+
"learning_rate": 5.078799229016083e-06,
|
| 1156 |
+
"loss": 0.0379,
|
| 1157 |
+
"step": 164
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 2.619047619047619,
|
| 1161 |
+
"grad_norm": 0.6620778441429138,
|
| 1162 |
+
"learning_rate": 5.0262673762765316e-06,
|
| 1163 |
+
"loss": 0.042,
|
| 1164 |
+
"step": 165
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 2.634920634920635,
|
| 1168 |
+
"grad_norm": 0.7500037550926208,
|
| 1169 |
+
"learning_rate": 4.973732623723471e-06,
|
| 1170 |
+
"loss": 0.0494,
|
| 1171 |
+
"step": 166
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 2.6507936507936507,
|
| 1175 |
+
"grad_norm": 0.6531999707221985,
|
| 1176 |
+
"learning_rate": 4.921200770983919e-06,
|
| 1177 |
+
"loss": 0.0364,
|
| 1178 |
+
"step": 167
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 2.6666666666666665,
|
| 1182 |
+
"grad_norm": 1.0730338096618652,
|
| 1183 |
+
"learning_rate": 4.8686776173647655e-06,
|
| 1184 |
+
"loss": 0.0554,
|
| 1185 |
+
"step": 168
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 2.682539682539683,
|
| 1189 |
+
"grad_norm": 0.6723483204841614,
|
| 1190 |
+
"learning_rate": 4.816168961212553e-06,
|
| 1191 |
+
"loss": 0.0532,
|
| 1192 |
+
"step": 169
|
| 1193 |
+
},
|
| 1194 |
+
{
|
| 1195 |
+
"epoch": 2.6984126984126986,
|
| 1196 |
+
"grad_norm": 0.7021626234054565,
|
| 1197 |
+
"learning_rate": 4.763680599273357e-06,
|
| 1198 |
+
"loss": 0.0611,
|
| 1199 |
+
"step": 170
|
| 1200 |
+
},
|
| 1201 |
+
{
|
| 1202 |
+
"epoch": 2.7142857142857144,
|
| 1203 |
+
"grad_norm": 0.7522814273834229,
|
| 1204 |
+
"learning_rate": 4.711218326052859e-06,
|
| 1205 |
+
"loss": 0.0426,
|
| 1206 |
+
"step": 171
|
| 1207 |
+
},
|
| 1208 |
+
{
|
| 1209 |
+
"epoch": 2.7301587301587302,
|
| 1210 |
+
"grad_norm": 0.6586665511131287,
|
| 1211 |
+
"learning_rate": 4.6587879331766465e-06,
|
| 1212 |
+
"loss": 0.0368,
|
| 1213 |
+
"step": 172
|
| 1214 |
+
},
|
| 1215 |
+
{
|
| 1216 |
+
"epoch": 2.746031746031746,
|
| 1217 |
+
"grad_norm": 1.2219913005828857,
|
| 1218 |
+
"learning_rate": 4.606395208750844e-06,
|
| 1219 |
+
"loss": 0.0852,
|
| 1220 |
+
"step": 173
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 2.761904761904762,
|
| 1224 |
+
"grad_norm": 0.6554266810417175,
|
| 1225 |
+
"learning_rate": 4.554045936723132e-06,
|
| 1226 |
+
"loss": 0.0516,
|
| 1227 |
+
"step": 174
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"epoch": 2.7777777777777777,
|
| 1231 |
+
"grad_norm": 0.5427403450012207,
|
| 1232 |
+
"learning_rate": 4.501745896244219e-06,
|
| 1233 |
+
"loss": 0.0388,
|
| 1234 |
+
"step": 175
|
| 1235 |
+
},
|
| 1236 |
+
{
|
| 1237 |
+
"epoch": 2.7936507936507935,
|
| 1238 |
+
"grad_norm": 0.5815466046333313,
|
| 1239 |
+
"learning_rate": 4.4495008610298435e-06,
|
| 1240 |
+
"loss": 0.0452,
|
| 1241 |
+
"step": 176
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 2.8095238095238093,
|
| 1245 |
+
"grad_norm": 0.957595944404602,
|
| 1246 |
+
"learning_rate": 4.397316598723385e-06,
|
| 1247 |
+
"loss": 0.0584,
|
| 1248 |
+
"step": 177
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 2.825396825396825,
|
| 1252 |
+
"grad_norm": 0.6840035915374756,
|
| 1253 |
+
"learning_rate": 4.345198870259139e-06,
|
| 1254 |
+
"loss": 0.0503,
|
| 1255 |
+
"step": 178
|
| 1256 |
+
},
|
| 1257 |
+
{
|
| 1258 |
+
"epoch": 2.8412698412698414,
|
| 1259 |
+
"grad_norm": 0.7481380105018616,
|
| 1260 |
+
"learning_rate": 4.2931534292263265e-06,
|
| 1261 |
+
"loss": 0.0561,
|
| 1262 |
+
"step": 179
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 2.857142857142857,
|
| 1266 |
+
"grad_norm": 0.5073666572570801,
|
| 1267 |
+
"learning_rate": 4.241186021233925e-06,
|
| 1268 |
+
"loss": 0.0339,
|
| 1269 |
+
"step": 180
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 2.873015873015873,
|
| 1273 |
+
"grad_norm": 0.6581488251686096,
|
| 1274 |
+
"learning_rate": 4.189302383276378e-06,
|
| 1275 |
+
"loss": 0.0629,
|
| 1276 |
+
"step": 181
|
| 1277 |
+
},
|
| 1278 |
+
{
|
| 1279 |
+
"epoch": 2.888888888888889,
|
| 1280 |
+
"grad_norm": 0.623148500919342,
|
| 1281 |
+
"learning_rate": 4.137508243100249e-06,
|
| 1282 |
+
"loss": 0.0521,
|
| 1283 |
+
"step": 182
|
| 1284 |
+
},
|
| 1285 |
+
{
|
| 1286 |
+
"epoch": 2.9047619047619047,
|
| 1287 |
+
"grad_norm": 0.621598482131958,
|
| 1288 |
+
"learning_rate": 4.085809318571905e-06,
|
| 1289 |
+
"loss": 0.0458,
|
| 1290 |
+
"step": 183
|
| 1291 |
+
},
|
| 1292 |
+
{
|
| 1293 |
+
"epoch": 2.9206349206349205,
|
| 1294 |
+
"grad_norm": 0.6772786974906921,
|
| 1295 |
+
"learning_rate": 4.034211317046285e-06,
|
| 1296 |
+
"loss": 0.044,
|
| 1297 |
+
"step": 184
|
| 1298 |
+
},
|
| 1299 |
+
{
|
| 1300 |
+
"epoch": 2.9365079365079367,
|
| 1301 |
+
"grad_norm": 1.262326955795288,
|
| 1302 |
+
"learning_rate": 3.982719934736832e-06,
|
| 1303 |
+
"loss": 0.046,
|
| 1304 |
+
"step": 185
|
| 1305 |
+
},
|
| 1306 |
+
{
|
| 1307 |
+
"epoch": 2.9523809523809526,
|
| 1308 |
+
"grad_norm": 0.6345784068107605,
|
| 1309 |
+
"learning_rate": 3.931340856086652e-06,
|
| 1310 |
+
"loss": 0.042,
|
| 1311 |
+
"step": 186
|
| 1312 |
+
},
|
| 1313 |
+
{
|
| 1314 |
+
"epoch": 2.9682539682539684,
|
| 1315 |
+
"grad_norm": 0.5799025297164917,
|
| 1316 |
+
"learning_rate": 3.880079753140978e-06,
|
| 1317 |
+
"loss": 0.0467,
|
| 1318 |
+
"step": 187
|
| 1319 |
+
},
|
| 1320 |
+
{
|
| 1321 |
+
"epoch": 2.984126984126984,
|
| 1322 |
+
"grad_norm": 0.6458066701889038,
|
| 1323 |
+
"learning_rate": 3.82894228492099e-06,
|
| 1324 |
+
"loss": 0.047,
|
| 1325 |
+
"step": 188
|
| 1326 |
+
},
|
| 1327 |
+
{
|
| 1328 |
+
"epoch": 3.0,
|
| 1329 |
+
"grad_norm": 0.5344202518463135,
|
| 1330 |
+
"learning_rate": 3.777934096799094e-06,
|
| 1331 |
+
"loss": 0.0418,
|
| 1332 |
+
"step": 189
|
| 1333 |
+
},
|
| 1334 |
+
{
|
| 1335 |
+
"epoch": 3.015873015873016,
|
| 1336 |
+
"grad_norm": 0.3208453953266144,
|
| 1337 |
+
"learning_rate": 3.7270608198756852e-06,
|
| 1338 |
+
"loss": 0.0124,
|
| 1339 |
+
"step": 190
|
| 1340 |
+
},
|
| 1341 |
+
{
|
| 1342 |
+
"epoch": 3.0317460317460316,
|
| 1343 |
+
"grad_norm": 0.48605024814605713,
|
| 1344 |
+
"learning_rate": 3.676328070357503e-06,
|
| 1345 |
+
"loss": 0.0332,
|
| 1346 |
+
"step": 191
|
| 1347 |
+
},
|
| 1348 |
+
{
|
| 1349 |
+
"epoch": 3.0476190476190474,
|
| 1350 |
+
"grad_norm": 0.5282842516899109,
|
| 1351 |
+
"learning_rate": 3.6257414489376217e-06,
|
| 1352 |
+
"loss": 0.0155,
|
| 1353 |
+
"step": 192
|
| 1354 |
+
},
|
| 1355 |
+
{
|
| 1356 |
+
"epoch": 3.0634920634920633,
|
| 1357 |
+
"grad_norm": 0.3682861030101776,
|
| 1358 |
+
"learning_rate": 3.5753065401771577e-06,
|
| 1359 |
+
"loss": 0.017,
|
| 1360 |
+
"step": 193
|
| 1361 |
+
},
|
| 1362 |
+
{
|
| 1363 |
+
"epoch": 3.0793650793650795,
|
| 1364 |
+
"grad_norm": 0.4284667372703552,
|
| 1365 |
+
"learning_rate": 3.5250289118887515e-06,
|
| 1366 |
+
"loss": 0.0107,
|
| 1367 |
+
"step": 194
|
| 1368 |
+
},
|
| 1369 |
+
{
|
| 1370 |
+
"epoch": 3.0952380952380953,
|
| 1371 |
+
"grad_norm": 0.36055704951286316,
|
| 1372 |
+
"learning_rate": 3.4749141145219118e-06,
|
| 1373 |
+
"loss": 0.0116,
|
| 1374 |
+
"step": 195
|
| 1375 |
+
},
|
| 1376 |
+
{
|
| 1377 |
+
"epoch": 3.111111111111111,
|
| 1378 |
+
"grad_norm": 0.42703211307525635,
|
| 1379 |
+
"learning_rate": 3.424967680550261e-06,
|
| 1380 |
+
"loss": 0.0089,
|
| 1381 |
+
"step": 196
|
| 1382 |
+
},
|
| 1383 |
+
{
|
| 1384 |
+
"epoch": 3.126984126984127,
|
| 1385 |
+
"grad_norm": 0.4422130882740021,
|
| 1386 |
+
"learning_rate": 3.3751951238607745e-06,
|
| 1387 |
+
"loss": 0.015,
|
| 1388 |
+
"step": 197
|
| 1389 |
+
},
|
| 1390 |
+
{
|
| 1391 |
+
"epoch": 3.142857142857143,
|
| 1392 |
+
"grad_norm": 1.8617546558380127,
|
| 1393 |
+
"learning_rate": 3.3256019391450696e-06,
|
| 1394 |
+
"loss": 0.044,
|
| 1395 |
+
"step": 198
|
| 1396 |
+
},
|
| 1397 |
+
{
|
| 1398 |
+
"epoch": 3.1587301587301586,
|
| 1399 |
+
"grad_norm": 0.4572526216506958,
|
| 1400 |
+
"learning_rate": 3.2761936012928147e-06,
|
| 1401 |
+
"loss": 0.0092,
|
| 1402 |
+
"step": 199
|
| 1403 |
+
},
|
| 1404 |
+
{
|
| 1405 |
+
"epoch": 3.1746031746031744,
|
| 1406 |
+
"grad_norm": 0.606041431427002,
|
| 1407 |
+
"learning_rate": 3.226975564787322e-06,
|
| 1408 |
+
"loss": 0.013,
|
| 1409 |
+
"step": 200
|
| 1410 |
+
},
|
| 1411 |
+
{
|
| 1412 |
+
"epoch": 3.1904761904761907,
|
| 1413 |
+
"grad_norm": 0.6772161722183228,
|
| 1414 |
+
"learning_rate": 3.177953263103394e-06,
|
| 1415 |
+
"loss": 0.0233,
|
| 1416 |
+
"step": 201
|
| 1417 |
+
},
|
| 1418 |
+
{
|
| 1419 |
+
"epoch": 3.2063492063492065,
|
| 1420 |
+
"grad_norm": 0.4099721908569336,
|
| 1421 |
+
"learning_rate": 3.1291321081074887e-06,
|
| 1422 |
+
"loss": 0.0074,
|
| 1423 |
+
"step": 202
|
| 1424 |
+
},
|
| 1425 |
+
{
|
| 1426 |
+
"epoch": 3.2222222222222223,
|
| 1427 |
+
"grad_norm": 0.34896424412727356,
|
| 1428 |
+
"learning_rate": 3.0805174894602775e-06,
|
| 1429 |
+
"loss": 0.0081,
|
| 1430 |
+
"step": 203
|
| 1431 |
+
},
|
| 1432 |
+
{
|
| 1433 |
+
"epoch": 3.238095238095238,
|
| 1434 |
+
"grad_norm": 0.5158839821815491,
|
| 1435 |
+
"learning_rate": 3.032114774021636e-06,
|
| 1436 |
+
"loss": 0.0174,
|
| 1437 |
+
"step": 204
|
| 1438 |
+
},
|
| 1439 |
+
{
|
| 1440 |
+
"epoch": 3.253968253968254,
|
| 1441 |
+
"grad_norm": 0.6538414359092712,
|
| 1442 |
+
"learning_rate": 2.9839293052581767e-06,
|
| 1443 |
+
"loss": 0.0188,
|
| 1444 |
+
"step": 205
|
| 1445 |
+
},
|
| 1446 |
+
{
|
| 1447 |
+
"epoch": 3.2698412698412698,
|
| 1448 |
+
"grad_norm": 0.4532378911972046,
|
| 1449 |
+
"learning_rate": 2.9359664026533443e-06,
|
| 1450 |
+
"loss": 0.0123,
|
| 1451 |
+
"step": 206
|
| 1452 |
+
},
|
| 1453 |
+
{
|
| 1454 |
+
"epoch": 3.2857142857142856,
|
| 1455 |
+
"grad_norm": 0.4588467478752136,
|
| 1456 |
+
"learning_rate": 2.8882313611201684e-06,
|
| 1457 |
+
"loss": 0.0154,
|
| 1458 |
+
"step": 207
|
| 1459 |
+
},
|
| 1460 |
+
{
|
| 1461 |
+
"epoch": 3.3015873015873014,
|
| 1462 |
+
"grad_norm": 0.6380066871643066,
|
| 1463 |
+
"learning_rate": 2.8407294504167238e-06,
|
| 1464 |
+
"loss": 0.0168,
|
| 1465 |
+
"step": 208
|
| 1466 |
+
},
|
| 1467 |
+
{
|
| 1468 |
+
"epoch": 3.317460317460317,
|
| 1469 |
+
"grad_norm": 0.3832227289676666,
|
| 1470 |
+
"learning_rate": 2.793465914564375e-06,
|
| 1471 |
+
"loss": 0.0118,
|
| 1472 |
+
"step": 209
|
| 1473 |
+
},
|
| 1474 |
+
{
|
| 1475 |
+
"epoch": 3.3333333333333335,
|
| 1476 |
+
"grad_norm": 0.4849812090396881,
|
| 1477 |
+
"learning_rate": 2.7464459712688517e-06,
|
| 1478 |
+
"loss": 0.0156,
|
| 1479 |
+
"step": 210
|
| 1480 |
+
},
|
| 1481 |
+
{
|
| 1482 |
+
"epoch": 3.3492063492063493,
|
| 1483 |
+
"grad_norm": 0.5382815003395081,
|
| 1484 |
+
"learning_rate": 2.6996748113442397e-06,
|
| 1485 |
+
"loss": 0.0162,
|
| 1486 |
+
"step": 211
|
| 1487 |
+
},
|
| 1488 |
+
{
|
| 1489 |
+
"epoch": 3.365079365079365,
|
| 1490 |
+
"grad_norm": 0.4217240810394287,
|
| 1491 |
+
"learning_rate": 2.653157598139932e-06,
|
| 1492 |
+
"loss": 0.0144,
|
| 1493 |
+
"step": 212
|
| 1494 |
+
},
|
| 1495 |
+
{
|
| 1496 |
+
"epoch": 3.380952380952381,
|
| 1497 |
+
"grad_norm": 0.3669329881668091,
|
| 1498 |
+
"learning_rate": 2.6068994669706184e-06,
|
| 1499 |
+
"loss": 0.0068,
|
| 1500 |
+
"step": 213
|
| 1501 |
+
},
|
| 1502 |
+
{
|
| 1503 |
+
"epoch": 3.3968253968253967,
|
| 1504 |
+
"grad_norm": 0.3910522162914276,
|
| 1505 |
+
"learning_rate": 2.560905524549364e-06,
|
| 1506 |
+
"loss": 0.0134,
|
| 1507 |
+
"step": 214
|
| 1508 |
+
},
|
| 1509 |
+
{
|
| 1510 |
+
"epoch": 3.4126984126984126,
|
| 1511 |
+
"grad_norm": 0.3835786283016205,
|
| 1512 |
+
"learning_rate": 2.515180848423853e-06,
|
| 1513 |
+
"loss": 0.009,
|
| 1514 |
+
"step": 215
|
| 1515 |
+
},
|
| 1516 |
+
{
|
| 1517 |
+
"epoch": 3.4285714285714284,
|
| 1518 |
+
"grad_norm": 0.39538145065307617,
|
| 1519 |
+
"learning_rate": 2.469730486415842e-06,
|
| 1520 |
+
"loss": 0.0154,
|
| 1521 |
+
"step": 216
|
| 1522 |
+
},
|
| 1523 |
+
{
|
| 1524 |
+
"epoch": 3.4444444444444446,
|
| 1525 |
+
"grad_norm": 0.48559364676475525,
|
| 1526 |
+
"learning_rate": 2.4245594560639086e-06,
|
| 1527 |
+
"loss": 0.0159,
|
| 1528 |
+
"step": 217
|
| 1529 |
+
},
|
| 1530 |
+
{
|
| 1531 |
+
"epoch": 3.4603174603174605,
|
| 1532 |
+
"grad_norm": 0.6111780405044556,
|
| 1533 |
+
"learning_rate": 2.379672744069527e-06,
|
| 1534 |
+
"loss": 0.0216,
|
| 1535 |
+
"step": 218
|
| 1536 |
+
},
|
| 1537 |
+
{
|
| 1538 |
+
"epoch": 3.4761904761904763,
|
| 1539 |
+
"grad_norm": 0.48673927783966064,
|
| 1540 |
+
"learning_rate": 2.335075305746558e-06,
|
| 1541 |
+
"loss": 0.0124,
|
| 1542 |
+
"step": 219
|
| 1543 |
+
},
|
| 1544 |
+
{
|
| 1545 |
+
"epoch": 3.492063492063492,
|
| 1546 |
+
"grad_norm": 0.4308861792087555,
|
| 1547 |
+
"learning_rate": 2.2907720644742064e-06,
|
| 1548 |
+
"loss": 0.0112,
|
| 1549 |
+
"step": 220
|
| 1550 |
+
},
|
| 1551 |
+
{
|
| 1552 |
+
"epoch": 3.507936507936508,
|
| 1553 |
+
"grad_norm": 0.39692750573158264,
|
| 1554 |
+
"learning_rate": 2.2467679111534963e-06,
|
| 1555 |
+
"loss": 0.0162,
|
| 1556 |
+
"step": 221
|
| 1557 |
+
},
|
| 1558 |
+
{
|
| 1559 |
+
"epoch": 3.5238095238095237,
|
| 1560 |
+
"grad_norm": 0.4090782701969147,
|
| 1561 |
+
"learning_rate": 2.2030677036673345e-06,
|
| 1562 |
+
"loss": 0.0214,
|
| 1563 |
+
"step": 222
|
| 1564 |
+
},
|
| 1565 |
+
{
|
| 1566 |
+
"epoch": 3.5396825396825395,
|
| 1567 |
+
"grad_norm": 0.4135613441467285,
|
| 1568 |
+
"learning_rate": 2.159676266344222e-06,
|
| 1569 |
+
"loss": 0.0157,
|
| 1570 |
+
"step": 223
|
| 1571 |
+
},
|
| 1572 |
+
{
|
| 1573 |
+
"epoch": 3.5555555555555554,
|
| 1574 |
+
"grad_norm": 0.8922034502029419,
|
| 1575 |
+
"learning_rate": 2.1165983894256647e-06,
|
| 1576 |
+
"loss": 0.0355,
|
| 1577 |
+
"step": 224
|
| 1578 |
+
},
|
| 1579 |
+
{
|
| 1580 |
+
"epoch": 3.571428571428571,
|
| 1581 |
+
"grad_norm": 0.25958430767059326,
|
| 1582 |
+
"learning_rate": 2.0738388285373532e-06,
|
| 1583 |
+
"loss": 0.0057,
|
| 1584 |
+
"step": 225
|
| 1585 |
+
},
|
| 1586 |
+
{
|
| 1587 |
+
"epoch": 3.5873015873015874,
|
| 1588 |
+
"grad_norm": 0.3267192244529724,
|
| 1589 |
+
"learning_rate": 2.0314023041641567e-06,
|
| 1590 |
+
"loss": 0.0089,
|
| 1591 |
+
"step": 226
|
| 1592 |
+
},
|
| 1593 |
+
{
|
| 1594 |
+
"epoch": 3.6031746031746033,
|
| 1595 |
+
"grad_norm": 0.4624428451061249,
|
| 1596 |
+
"learning_rate": 1.9892935011290037e-06,
|
| 1597 |
+
"loss": 0.0159,
|
| 1598 |
+
"step": 227
|
| 1599 |
+
},
|
| 1600 |
+
{
|
| 1601 |
+
"epoch": 3.619047619047619,
|
| 1602 |
+
"grad_norm": 0.2781091630458832,
|
| 1603 |
+
"learning_rate": 1.947517068075694e-06,
|
| 1604 |
+
"loss": 0.0083,
|
| 1605 |
+
"step": 228
|
| 1606 |
+
},
|
| 1607 |
+
{
|
| 1608 |
+
"epoch": 3.634920634920635,
|
| 1609 |
+
"grad_norm": 0.31553471088409424,
|
| 1610 |
+
"learning_rate": 1.9060776169557083e-06,
|
| 1611 |
+
"loss": 0.0137,
|
| 1612 |
+
"step": 229
|
| 1613 |
+
},
|
| 1614 |
+
{
|
| 1615 |
+
"epoch": 3.6507936507936507,
|
| 1616 |
+
"grad_norm": 0.5683518052101135,
|
| 1617 |
+
"learning_rate": 1.864979722519068e-06,
|
| 1618 |
+
"loss": 0.0298,
|
| 1619 |
+
"step": 230
|
| 1620 |
+
},
|
| 1621 |
+
{
|
| 1622 |
+
"epoch": 3.6666666666666665,
|
| 1623 |
+
"grad_norm": 0.3412047028541565,
|
| 1624 |
+
"learning_rate": 1.8242279218092968e-06,
|
| 1625 |
+
"loss": 0.0137,
|
| 1626 |
+
"step": 231
|
| 1627 |
+
},
|
| 1628 |
+
{
|
| 1629 |
+
"epoch": 3.682539682539683,
|
| 1630 |
+
"grad_norm": 0.5104885101318359,
|
| 1631 |
+
"learning_rate": 1.7838267136625536e-06,
|
| 1632 |
+
"loss": 0.0114,
|
| 1633 |
+
"step": 232
|
| 1634 |
+
},
|
| 1635 |
+
{
|
| 1636 |
+
"epoch": 3.6984126984126986,
|
| 1637 |
+
"grad_norm": 0.4502926170825958,
|
| 1638 |
+
"learning_rate": 1.743780558210979e-06,
|
| 1639 |
+
"loss": 0.013,
|
| 1640 |
+
"step": 233
|
| 1641 |
+
},
|
| 1642 |
+
{
|
| 1643 |
+
"epoch": 3.7142857142857144,
|
| 1644 |
+
"grad_norm": 0.48987045884132385,
|
| 1645 |
+
"learning_rate": 1.704093876390312e-06,
|
| 1646 |
+
"loss": 0.0217,
|
| 1647 |
+
"step": 234
|
| 1648 |
+
},
|
| 1649 |
+
{
|
| 1650 |
+
"epoch": 3.7301587301587302,
|
| 1651 |
+
"grad_norm": 0.3420720398426056,
|
| 1652 |
+
"learning_rate": 1.664771049451837e-06,
|
| 1653 |
+
"loss": 0.0089,
|
| 1654 |
+
"step": 235
|
| 1655 |
+
},
|
| 1656 |
+
{
|
| 1657 |
+
"epoch": 3.746031746031746,
|
| 1658 |
+
"grad_norm": 0.3107622563838959,
|
| 1659 |
+
"learning_rate": 1.6258164184787123e-06,
|
| 1660 |
+
"loss": 0.0085,
|
| 1661 |
+
"step": 236
|
| 1662 |
+
},
|
| 1663 |
+
{
|
| 1664 |
+
"epoch": 3.761904761904762,
|
| 1665 |
+
"grad_norm": 0.5032781958580017,
|
| 1666 |
+
"learning_rate": 1.5872342839067305e-06,
|
| 1667 |
+
"loss": 0.0075,
|
| 1668 |
+
"step": 237
|
| 1669 |
+
},
|
| 1670 |
+
{
|
| 1671 |
+
"epoch": 3.7777777777777777,
|
| 1672 |
+
"grad_norm": 0.3874753415584564,
|
| 1673 |
+
"learning_rate": 1.5490289050495678e-06,
|
| 1674 |
+
"loss": 0.0108,
|
| 1675 |
+
"step": 238
|
| 1676 |
+
},
|
| 1677 |
+
{
|
| 1678 |
+
"epoch": 3.7936507936507935,
|
| 1679 |
+
"grad_norm": 0.30904141068458557,
|
| 1680 |
+
"learning_rate": 1.511204499628574e-06,
|
| 1681 |
+
"loss": 0.0086,
|
| 1682 |
+
"step": 239
|
| 1683 |
+
},
|
| 1684 |
+
{
|
| 1685 |
+
"epoch": 3.8095238095238093,
|
| 1686 |
+
"grad_norm": 0.39750581979751587,
|
| 1687 |
+
"learning_rate": 1.4737652433071515e-06,
|
| 1688 |
+
"loss": 0.0135,
|
| 1689 |
+
"step": 240
|
| 1690 |
+
},
|
| 1691 |
+
{
|
| 1692 |
+
"epoch": 3.825396825396825,
|
| 1693 |
+
"grad_norm": 0.5290718674659729,
|
| 1694 |
+
"learning_rate": 1.4367152692297799e-06,
|
| 1695 |
+
"loss": 0.0187,
|
| 1696 |
+
"step": 241
|
| 1697 |
+
},
|
| 1698 |
+
{
|
| 1699 |
+
"epoch": 3.8412698412698414,
|
| 1700 |
+
"grad_norm": 0.4600606858730316,
|
| 1701 |
+
"learning_rate": 1.4000586675657312e-06,
|
| 1702 |
+
"loss": 0.0111,
|
| 1703 |
+
"step": 242
|
| 1704 |
+
},
|
| 1705 |
+
{
|
| 1706 |
+
"epoch": 3.857142857142857,
|
| 1707 |
+
"grad_norm": 0.461280882358551,
|
| 1708 |
+
"learning_rate": 1.3637994850575342e-06,
|
| 1709 |
+
"loss": 0.0147,
|
| 1710 |
+
"step": 243
|
| 1711 |
+
},
|
| 1712 |
+
{
|
| 1713 |
+
"epoch": 3.873015873015873,
|
| 1714 |
+
"grad_norm": 0.5423941016197205,
|
| 1715 |
+
"learning_rate": 1.3279417245742288e-06,
|
| 1716 |
+
"loss": 0.0236,
|
| 1717 |
+
"step": 244
|
| 1718 |
+
},
|
| 1719 |
+
{
|
| 1720 |
+
"epoch": 3.888888888888889,
|
| 1721 |
+
"grad_norm": 0.39030736684799194,
|
| 1722 |
+
"learning_rate": 1.2924893446694648e-06,
|
| 1723 |
+
"loss": 0.006,
|
| 1724 |
+
"step": 245
|
| 1725 |
+
},
|
| 1726 |
+
{
|
| 1727 |
+
"epoch": 3.9047619047619047,
|
| 1728 |
+
"grad_norm": 0.5754550099372864,
|
| 1729 |
+
"learning_rate": 1.257446259144494e-06,
|
| 1730 |
+
"loss": 0.0286,
|
| 1731 |
+
"step": 246
|
| 1732 |
+
},
|
| 1733 |
+
{
|
| 1734 |
+
"epoch": 3.9206349206349205,
|
| 1735 |
+
"grad_norm": 0.35015997290611267,
|
| 1736 |
+
"learning_rate": 1.222816336616104e-06,
|
| 1737 |
+
"loss": 0.0143,
|
| 1738 |
+
"step": 247
|
| 1739 |
+
},
|
| 1740 |
+
{
|
| 1741 |
+
"epoch": 3.9365079365079367,
|
| 1742 |
+
"grad_norm": 0.4731229245662689,
|
| 1743 |
+
"learning_rate": 1.1886034000895341e-06,
|
| 1744 |
+
"loss": 0.0177,
|
| 1745 |
+
"step": 248
|
| 1746 |
+
},
|
| 1747 |
+
{
|
| 1748 |
+
"epoch": 3.9523809523809526,
|
| 1749 |
+
"grad_norm": 0.3958592116832733,
|
| 1750 |
+
"learning_rate": 1.1548112265364336e-06,
|
| 1751 |
+
"loss": 0.0132,
|
| 1752 |
+
"step": 249
|
| 1753 |
+
},
|
| 1754 |
+
{
|
| 1755 |
+
"epoch": 3.9682539682539684,
|
| 1756 |
+
"grad_norm": 0.4273618161678314,
|
| 1757 |
+
"learning_rate": 1.1214435464779006e-06,
|
| 1758 |
+
"loss": 0.0206,
|
| 1759 |
+
"step": 250
|
| 1760 |
+
},
|
| 1761 |
+
{
|
| 1762 |
+
"epoch": 3.984126984126984,
|
| 1763 |
+
"grad_norm": 0.429865300655365,
|
| 1764 |
+
"learning_rate": 1.088504043572643e-06,
|
| 1765 |
+
"loss": 0.0131,
|
| 1766 |
+
"step": 251
|
| 1767 |
+
},
|
| 1768 |
+
{
|
| 1769 |
+
"epoch": 4.0,
|
| 1770 |
+
"grad_norm": 0.28878334164619446,
|
| 1771 |
+
"learning_rate": 1.055996354210323e-06,
|
| 1772 |
+
"loss": 0.0079,
|
| 1773 |
+
"step": 252
|
| 1774 |
+
},
|
| 1775 |
+
{
|
| 1776 |
+
"epoch": 4.015873015873016,
|
| 1777 |
+
"grad_norm": 0.12734107673168182,
|
| 1778 |
+
"learning_rate": 1.0239240671101065e-06,
|
| 1779 |
+
"loss": 0.0027,
|
| 1780 |
+
"step": 253
|
| 1781 |
+
},
|
| 1782 |
+
{
|
| 1783 |
+
"epoch": 4.031746031746032,
|
| 1784 |
+
"grad_norm": 0.24097059667110443,
|
| 1785 |
+
"learning_rate": 9.922907229244905e-07,
|
| 1786 |
+
"loss": 0.0063,
|
| 1787 |
+
"step": 254
|
| 1788 |
+
},
|
| 1789 |
+
{
|
| 1790 |
+
"epoch": 4.0476190476190474,
|
| 1791 |
+
"grad_norm": 0.18659502267837524,
|
| 1792 |
+
"learning_rate": 9.610998138484262e-07,
|
| 1793 |
+
"loss": 0.0032,
|
| 1794 |
+
"step": 255
|
| 1795 |
+
},
|
| 1796 |
+
{
|
| 1797 |
+
"epoch": 4.063492063492063,
|
| 1798 |
+
"grad_norm": 0.16850638389587402,
|
| 1799 |
+
"learning_rate": 9.303547832337934e-07,
|
| 1800 |
+
"loss": 0.0036,
|
| 1801 |
+
"step": 256
|
| 1802 |
+
},
|
| 1803 |
+
{
|
| 1804 |
+
"epoch": 4.079365079365079,
|
| 1805 |
+
"grad_norm": 0.13990715146064758,
|
| 1806 |
+
"learning_rate": 9.000590252092701e-07,
|
| 1807 |
+
"loss": 0.0034,
|
| 1808 |
+
"step": 257
|
| 1809 |
+
},
|
| 1810 |
+
{
|
| 1811 |
+
"epoch": 4.095238095238095,
|
| 1812 |
+
"grad_norm": 0.16486185789108276,
|
| 1813 |
+
"learning_rate": 8.702158843056319e-07,
|
| 1814 |
+
"loss": 0.0029,
|
| 1815 |
+
"step": 258
|
| 1816 |
+
},
|
| 1817 |
+
{
|
| 1818 |
+
"epoch": 4.111111111111111,
|
| 1819 |
+
"grad_norm": 0.13398580253124237,
|
| 1820 |
+
"learning_rate": 8.408286550865319e-07,
|
| 1821 |
+
"loss": 0.0041,
|
| 1822 |
+
"step": 259
|
| 1823 |
+
},
|
| 1824 |
+
{
|
| 1825 |
+
"epoch": 4.1269841269841265,
|
| 1826 |
+
"grad_norm": 0.14831550419330597,
|
| 1827 |
+
"learning_rate": 8.119005817847924e-07,
|
| 1828 |
+
"loss": 0.0039,
|
| 1829 |
+
"step": 260
|
| 1830 |
+
},
|
| 1831 |
+
{
|
| 1832 |
+
"epoch": 4.142857142857143,
|
| 1833 |
+
"grad_norm": 0.1601114124059677,
|
| 1834 |
+
"learning_rate": 7.834348579442552e-07,
|
| 1835 |
+
"loss": 0.0034,
|
| 1836 |
+
"step": 261
|
| 1837 |
+
},
|
| 1838 |
+
{
|
| 1839 |
+
"epoch": 4.158730158730159,
|
| 1840 |
+
"grad_norm": 0.2705823481082916,
|
| 1841 |
+
"learning_rate": 7.554346260672263e-07,
|
| 1842 |
+
"loss": 0.0054,
|
| 1843 |
+
"step": 262
|
| 1844 |
+
},
|
| 1845 |
+
{
|
| 1846 |
+
"epoch": 4.174603174603175,
|
| 1847 |
+
"grad_norm": 0.17291611433029175,
|
| 1848 |
+
"learning_rate": 7.279029772675572e-07,
|
| 1849 |
+
"loss": 0.0025,
|
| 1850 |
+
"step": 263
|
| 1851 |
+
},
|
| 1852 |
+
{
|
| 1853 |
+
"epoch": 4.190476190476191,
|
| 1854 |
+
"grad_norm": 0.11752714961767197,
|
| 1855 |
+
"learning_rate": 7.008429509293979e-07,
|
| 1856 |
+
"loss": 0.002,
|
| 1857 |
+
"step": 264
|
| 1858 |
+
},
|
| 1859 |
+
{
|
| 1860 |
+
"epoch": 4.2063492063492065,
|
| 1861 |
+
"grad_norm": 0.2471189945936203,
|
| 1862 |
+
"learning_rate": 6.742575343716584e-07,
|
| 1863 |
+
"loss": 0.0053,
|
| 1864 |
+
"step": 265
|
| 1865 |
+
},
|
| 1866 |
+
{
|
| 1867 |
+
"epoch": 4.222222222222222,
|
| 1868 |
+
"grad_norm": 0.2652353048324585,
|
| 1869 |
+
"learning_rate": 6.481496625182271e-07,
|
| 1870 |
+
"loss": 0.0035,
|
| 1871 |
+
"step": 266
|
| 1872 |
+
},
|
| 1873 |
+
{
|
| 1874 |
+
"epoch": 4.238095238095238,
|
| 1875 |
+
"grad_norm": 0.5520457029342651,
|
| 1876 |
+
"learning_rate": 6.225222175739598e-07,
|
| 1877 |
+
"loss": 0.0094,
|
| 1878 |
+
"step": 267
|
| 1879 |
+
},
|
| 1880 |
+
{
|
| 1881 |
+
"epoch": 4.253968253968254,
|
| 1882 |
+
"grad_norm": 0.1428154557943344,
|
| 1883 |
+
"learning_rate": 5.973780287065007e-07,
|
| 1884 |
+
"loss": 0.002,
|
| 1885 |
+
"step": 268
|
| 1886 |
+
},
|
| 1887 |
+
{
|
| 1888 |
+
"epoch": 4.26984126984127,
|
| 1889 |
+
"grad_norm": 0.38164809346199036,
|
| 1890 |
+
"learning_rate": 5.727198717339511e-07,
|
| 1891 |
+
"loss": 0.0185,
|
| 1892 |
+
"step": 269
|
| 1893 |
+
},
|
| 1894 |
+
{
|
| 1895 |
+
"epoch": 4.285714285714286,
|
| 1896 |
+
"grad_norm": 0.2093685418367386,
|
| 1897 |
+
"learning_rate": 5.485504688184307e-07,
|
| 1898 |
+
"loss": 0.0016,
|
| 1899 |
+
"step": 270
|
| 1900 |
+
},
|
| 1901 |
+
{
|
| 1902 |
+
"epoch": 4.301587301587301,
|
| 1903 |
+
"grad_norm": 0.15652765333652496,
|
| 1904 |
+
"learning_rate": 5.24872488165562e-07,
|
| 1905 |
+
"loss": 0.0029,
|
| 1906 |
+
"step": 271
|
| 1907 |
+
},
|
| 1908 |
+
{
|
| 1909 |
+
"epoch": 4.317460317460317,
|
| 1910 |
+
"grad_norm": 0.17758126556873322,
|
| 1911 |
+
"learning_rate": 5.016885437299113e-07,
|
| 1912 |
+
"loss": 0.003,
|
| 1913 |
+
"step": 272
|
| 1914 |
+
},
|
| 1915 |
+
{
|
| 1916 |
+
"epoch": 4.333333333333333,
|
| 1917 |
+
"grad_norm": 0.18992979824543,
|
| 1918 |
+
"learning_rate": 4.790011949264173e-07,
|
| 1919 |
+
"loss": 0.0064,
|
| 1920 |
+
"step": 273
|
| 1921 |
+
},
|
| 1922 |
+
{
|
| 1923 |
+
"epoch": 4.349206349206349,
|
| 1924 |
+
"grad_norm": 0.10369639843702316,
|
| 1925 |
+
"learning_rate": 4.5681294634784437e-07,
|
| 1926 |
+
"loss": 0.0014,
|
| 1927 |
+
"step": 274
|
| 1928 |
+
},
|
| 1929 |
+
{
|
| 1930 |
+
"epoch": 4.365079365079365,
|
| 1931 |
+
"grad_norm": 0.09521545469760895,
|
| 1932 |
+
"learning_rate": 4.3512624748828225e-07,
|
| 1933 |
+
"loss": 0.0012,
|
| 1934 |
+
"step": 275
|
| 1935 |
+
},
|
| 1936 |
+
{
|
| 1937 |
+
"epoch": 4.380952380952381,
|
| 1938 |
+
"grad_norm": 0.310982346534729,
|
| 1939 |
+
"learning_rate": 4.139434924727359e-07,
|
| 1940 |
+
"loss": 0.0059,
|
| 1941 |
+
"step": 276
|
| 1942 |
+
},
|
| 1943 |
+
{
|
| 1944 |
+
"epoch": 4.396825396825397,
|
| 1945 |
+
"grad_norm": 0.3128473162651062,
|
| 1946 |
+
"learning_rate": 3.9326701979281623e-07,
|
| 1947 |
+
"loss": 0.0059,
|
| 1948 |
+
"step": 277
|
| 1949 |
+
},
|
| 1950 |
+
{
|
| 1951 |
+
"epoch": 4.412698412698413,
|
| 1952 |
+
"grad_norm": 0.33223262429237366,
|
| 1953 |
+
"learning_rate": 3.7309911204858997e-07,
|
| 1954 |
+
"loss": 0.0031,
|
| 1955 |
+
"step": 278
|
| 1956 |
+
},
|
| 1957 |
+
{
|
| 1958 |
+
"epoch": 4.428571428571429,
|
| 1959 |
+
"grad_norm": 0.13230553269386292,
|
| 1960 |
+
"learning_rate": 3.534419956965823e-07,
|
| 1961 |
+
"loss": 0.0024,
|
| 1962 |
+
"step": 279
|
| 1963 |
+
},
|
| 1964 |
+
{
|
| 1965 |
+
"epoch": 4.444444444444445,
|
| 1966 |
+
"grad_norm": 0.20840996503829956,
|
| 1967 |
+
"learning_rate": 3.3429784080398765e-07,
|
| 1968 |
+
"loss": 0.005,
|
| 1969 |
+
"step": 280
|
| 1970 |
+
},
|
| 1971 |
+
{
|
| 1972 |
+
"epoch": 4.4603174603174605,
|
| 1973 |
+
"grad_norm": 0.21250075101852417,
|
| 1974 |
+
"learning_rate": 3.1566876080910193e-07,
|
| 1975 |
+
"loss": 0.0042,
|
| 1976 |
+
"step": 281
|
| 1977 |
+
},
|
| 1978 |
+
{
|
| 1979 |
+
"epoch": 4.476190476190476,
|
| 1980 |
+
"grad_norm": 0.14549191296100616,
|
| 1981 |
+
"learning_rate": 2.9755681228800904e-07,
|
| 1982 |
+
"loss": 0.0023,
|
| 1983 |
+
"step": 282
|
| 1984 |
+
},
|
| 1985 |
+
{
|
| 1986 |
+
"epoch": 4.492063492063492,
|
| 1987 |
+
"grad_norm": 0.27009841799736023,
|
| 1988 |
+
"learning_rate": 2.799639947275412e-07,
|
| 1989 |
+
"loss": 0.0037,
|
| 1990 |
+
"step": 283
|
| 1991 |
+
},
|
| 1992 |
+
{
|
| 1993 |
+
"epoch": 4.507936507936508,
|
| 1994 |
+
"grad_norm": 0.18486513197422028,
|
| 1995 |
+
"learning_rate": 2.6289225030454556e-07,
|
| 1996 |
+
"loss": 0.0025,
|
| 1997 |
+
"step": 284
|
| 1998 |
+
},
|
| 1999 |
+
{
|
| 2000 |
+
"epoch": 4.523809523809524,
|
| 2001 |
+
"grad_norm": 0.2685333490371704,
|
| 2002 |
+
"learning_rate": 2.4634346367147233e-07,
|
| 2003 |
+
"loss": 0.0038,
|
| 2004 |
+
"step": 285
|
| 2005 |
+
},
|
| 2006 |
+
{
|
| 2007 |
+
"epoch": 4.5396825396825395,
|
| 2008 |
+
"grad_norm": 0.22350570559501648,
|
| 2009 |
+
"learning_rate": 2.303194617483212e-07,
|
| 2010 |
+
"loss": 0.0052,
|
| 2011 |
+
"step": 286
|
| 2012 |
+
},
|
| 2013 |
+
{
|
| 2014 |
+
"epoch": 4.555555555555555,
|
| 2015 |
+
"grad_norm": 0.1270400732755661,
|
| 2016 |
+
"learning_rate": 2.1482201352095277e-07,
|
| 2017 |
+
"loss": 0.0037,
|
| 2018 |
+
"step": 287
|
| 2019 |
+
},
|
| 2020 |
+
{
|
| 2021 |
+
"epoch": 4.571428571428571,
|
| 2022 |
+
"grad_norm": 0.14703014492988586,
|
| 2023 |
+
"learning_rate": 1.998528298458019e-07,
|
| 2024 |
+
"loss": 0.0026,
|
| 2025 |
+
"step": 288
|
| 2026 |
+
},
|
| 2027 |
+
{
|
| 2028 |
+
"epoch": 4.587301587301587,
|
| 2029 |
+
"grad_norm": 0.05500807240605354,
|
| 2030 |
+
"learning_rate": 1.8541356326100436e-07,
|
| 2031 |
+
"loss": 0.0006,
|
| 2032 |
+
"step": 289
|
| 2033 |
+
},
|
| 2034 |
+
{
|
| 2035 |
+
"epoch": 4.603174603174603,
|
| 2036 |
+
"grad_norm": 0.23130561411380768,
|
| 2037 |
+
"learning_rate": 1.7150580780396385e-07,
|
| 2038 |
+
"loss": 0.0041,
|
| 2039 |
+
"step": 290
|
| 2040 |
+
},
|
| 2041 |
+
{
|
| 2042 |
+
"epoch": 4.619047619047619,
|
| 2043 |
+
"grad_norm": 0.14613750576972961,
|
| 2044 |
+
"learning_rate": 1.5813109883537792e-07,
|
| 2045 |
+
"loss": 0.0022,
|
| 2046 |
+
"step": 291
|
| 2047 |
+
},
|
| 2048 |
+
{
|
| 2049 |
+
"epoch": 4.634920634920634,
|
| 2050 |
+
"grad_norm": 0.19623324275016785,
|
| 2051 |
+
"learning_rate": 1.4529091286973994e-07,
|
| 2052 |
+
"loss": 0.0029,
|
| 2053 |
+
"step": 292
|
| 2054 |
+
},
|
| 2055 |
+
{
|
| 2056 |
+
"epoch": 4.650793650793651,
|
| 2057 |
+
"grad_norm": 0.2580835819244385,
|
| 2058 |
+
"learning_rate": 1.3298666741233424e-07,
|
| 2059 |
+
"loss": 0.0047,
|
| 2060 |
+
"step": 293
|
| 2061 |
+
},
|
| 2062 |
+
{
|
| 2063 |
+
"epoch": 4.666666666666667,
|
| 2064 |
+
"grad_norm": 0.19105015695095062,
|
| 2065 |
+
"learning_rate": 1.2121972080275378e-07,
|
| 2066 |
+
"loss": 0.0041,
|
| 2067 |
+
"step": 294
|
| 2068 |
+
},
|
| 2069 |
+
{
|
| 2070 |
+
"epoch": 4.682539682539683,
|
| 2071 |
+
"grad_norm": 0.18912175297737122,
|
| 2072 |
+
"learning_rate": 1.0999137206494315e-07,
|
| 2073 |
+
"loss": 0.0045,
|
| 2074 |
+
"step": 295
|
| 2075 |
+
},
|
| 2076 |
+
{
|
| 2077 |
+
"epoch": 4.698412698412699,
|
| 2078 |
+
"grad_norm": 0.2492668479681015,
|
| 2079 |
+
"learning_rate": 9.93028607637908e-08,
|
| 2080 |
+
"loss": 0.0067,
|
| 2081 |
+
"step": 296
|
| 2082 |
+
},
|
| 2083 |
+
{
|
| 2084 |
+
"epoch": 4.714285714285714,
|
| 2085 |
+
"grad_norm": 0.1554499715566635,
|
| 2086 |
+
"learning_rate": 8.915536686828764e-08,
|
| 2087 |
+
"loss": 0.0085,
|
| 2088 |
+
"step": 297
|
| 2089 |
+
},
|
| 2090 |
+
{
|
| 2091 |
+
"epoch": 4.73015873015873,
|
| 2092 |
+
"grad_norm": 0.16742238402366638,
|
| 2093 |
+
"learning_rate": 7.955001062125989e-08,
|
| 2094 |
+
"loss": 0.0036,
|
| 2095 |
+
"step": 298
|
| 2096 |
+
},
|
| 2097 |
+
{
|
| 2098 |
+
"epoch": 4.746031746031746,
|
| 2099 |
+
"grad_norm": 0.1783057302236557,
|
| 2100 |
+
"learning_rate": 7.048785241570321e-08,
|
| 2101 |
+
"loss": 0.0028,
|
| 2102 |
+
"step": 299
|
| 2103 |
+
},
|
| 2104 |
+
{
|
| 2105 |
+
"epoch": 4.761904761904762,
|
| 2106 |
+
"grad_norm": 0.1566859483718872,
|
| 2107 |
+
"learning_rate": 6.19698926777168e-08,
|
| 2108 |
+
"loss": 0.0033,
|
| 2109 |
+
"step": 300
|
| 2110 |
+
},
|
| 2111 |
+
{
|
| 2112 |
+
"epoch": 4.777777777777778,
|
| 2113 |
+
"grad_norm": 0.1521420031785965,
|
| 2114 |
+
"learning_rate": 5.399707175606117e-08,
|
| 2115 |
+
"loss": 0.0016,
|
| 2116 |
+
"step": 301
|
| 2117 |
+
},
|
| 2118 |
+
{
|
| 2119 |
+
"epoch": 4.7936507936507935,
|
| 2120 |
+
"grad_norm": 0.17157761752605438,
|
| 2121 |
+
"learning_rate": 4.657026981834623e-08,
|
| 2122 |
+
"loss": 0.002,
|
| 2123 |
+
"step": 302
|
| 2124 |
+
},
|
| 2125 |
+
{
|
| 2126 |
+
"epoch": 4.809523809523809,
|
| 2127 |
+
"grad_norm": 0.1927778720855713,
|
| 2128 |
+
"learning_rate": 3.9690306753866204e-08,
|
| 2129 |
+
"loss": 0.0029,
|
| 2130 |
+
"step": 303
|
| 2131 |
+
},
|
| 2132 |
+
{
|
| 2133 |
+
"epoch": 4.825396825396825,
|
| 2134 |
+
"grad_norm": 0.15908092260360718,
|
| 2135 |
+
"learning_rate": 3.3357942083085404e-08,
|
| 2136 |
+
"loss": 0.0016,
|
| 2137 |
+
"step": 304
|
| 2138 |
+
},
|
| 2139 |
+
{
|
| 2140 |
+
"epoch": 4.841269841269841,
|
| 2141 |
+
"grad_norm": 0.175292506814003,
|
| 2142 |
+
"learning_rate": 2.7573874873791372e-08,
|
| 2143 |
+
"loss": 0.0043,
|
| 2144 |
+
"step": 305
|
| 2145 |
+
},
|
| 2146 |
+
{
|
| 2147 |
+
"epoch": 4.857142857142857,
|
| 2148 |
+
"grad_norm": 0.121407650411129,
|
| 2149 |
+
"learning_rate": 2.233874366391997e-08,
|
| 2150 |
+
"loss": 0.001,
|
| 2151 |
+
"step": 306
|
| 2152 |
+
},
|
| 2153 |
+
{
|
| 2154 |
+
"epoch": 4.8730158730158735,
|
| 2155 |
+
"grad_norm": 0.2875802218914032,
|
| 2156 |
+
"learning_rate": 1.7653126391063425e-08,
|
| 2157 |
+
"loss": 0.005,
|
| 2158 |
+
"step": 307
|
| 2159 |
+
},
|
| 2160 |
+
{
|
| 2161 |
+
"epoch": 4.888888888888889,
|
| 2162 |
+
"grad_norm": 0.18822738528251648,
|
| 2163 |
+
"learning_rate": 1.3517540328669143e-08,
|
| 2164 |
+
"loss": 0.0025,
|
| 2165 |
+
"step": 308
|
| 2166 |
+
},
|
| 2167 |
+
{
|
| 2168 |
+
"epoch": 4.904761904761905,
|
| 2169 |
+
"grad_norm": 0.15360276401042938,
|
| 2170 |
+
"learning_rate": 9.93244202893262e-09,
|
| 2171 |
+
"loss": 0.0023,
|
| 2172 |
+
"step": 309
|
| 2173 |
+
},
|
| 2174 |
+
{
|
| 2175 |
+
"epoch": 4.920634920634921,
|
| 2176 |
+
"grad_norm": 0.29353615641593933,
|
| 2177 |
+
"learning_rate": 6.898227272398306e-09,
|
| 2178 |
+
"loss": 0.0065,
|
| 2179 |
+
"step": 310
|
| 2180 |
+
},
|
| 2181 |
+
{
|
| 2182 |
+
"epoch": 4.936507936507937,
|
| 2183 |
+
"grad_norm": 0.5181974172592163,
|
| 2184 |
+
"learning_rate": 4.415231024265665e-09,
|
| 2185 |
+
"loss": 0.0108,
|
| 2186 |
+
"step": 311
|
| 2187 |
+
},
|
| 2188 |
+
{
|
| 2189 |
+
"epoch": 4.9523809523809526,
|
| 2190 |
+
"grad_norm": 0.15942662954330444,
|
| 2191 |
+
"learning_rate": 2.4837273974115393e-09,
|
| 2192 |
+
"loss": 0.0026,
|
| 2193 |
+
"step": 312
|
| 2194 |
+
},
|
| 2195 |
+
{
|
| 2196 |
+
"epoch": 4.968253968253968,
|
| 2197 |
+
"grad_norm": 0.1265093982219696,
|
| 2198 |
+
"learning_rate": 1.1039296221276863e-09,
|
| 2199 |
+
"loss": 0.0116,
|
| 2200 |
+
"step": 313
|
| 2201 |
+
},
|
| 2202 |
+
{
|
| 2203 |
+
"epoch": 4.984126984126984,
|
| 2204 |
+
"grad_norm": 0.29910922050476074,
|
| 2205 |
+
"learning_rate": 2.7599002258127395e-10,
|
| 2206 |
+
"loss": 0.0022,
|
| 2207 |
+
"step": 314
|
| 2208 |
+
},
|
| 2209 |
+
{
|
| 2210 |
+
"epoch": 5.0,
|
| 2211 |
+
"grad_norm": 0.09251413494348526,
|
| 2212 |
+
"learning_rate": 0.0,
|
| 2213 |
+
"loss": 0.0015,
|
| 2214 |
+
"step": 315
|
| 2215 |
+
},
|
| 2216 |
+
{
|
| 2217 |
+
"epoch": 5.0,
|
| 2218 |
+
"step": 315,
|
| 2219 |
+
"total_flos": 32588508020736.0,
|
| 2220 |
+
"train_loss": 0.11869781007546754,
|
| 2221 |
+
"train_runtime": 5355.6377,
|
| 2222 |
+
"train_samples_per_second": 0.934,
|
| 2223 |
+
"train_steps_per_second": 0.059
|
| 2224 |
+
}
|
| 2225 |
+
],
|
| 2226 |
+
"logging_steps": 1.0,
|
| 2227 |
+
"max_steps": 315,
|
| 2228 |
+
"num_input_tokens_seen": 0,
|
| 2229 |
+
"num_train_epochs": 5,
|
| 2230 |
+
"save_steps": 500,
|
| 2231 |
+
"stateful_callbacks": {
|
| 2232 |
+
"TrainerControl": {
|
| 2233 |
+
"args": {
|
| 2234 |
+
"should_epoch_stop": false,
|
| 2235 |
+
"should_evaluate": false,
|
| 2236 |
+
"should_log": false,
|
| 2237 |
+
"should_save": true,
|
| 2238 |
+
"should_training_stop": true
|
| 2239 |
+
},
|
| 2240 |
+
"attributes": {}
|
| 2241 |
+
}
|
| 2242 |
+
},
|
| 2243 |
+
"total_flos": 32588508020736.0,
|
| 2244 |
+
"train_batch_size": 1,
|
| 2245 |
+
"trial_name": null,
|
| 2246 |
+
"trial_params": null
|
| 2247 |
+
}
|
training_loss.png
ADDED
|