End of training
Browse files- README.md +2 -1
- all_results.json +8 -0
- train_results.json +8 -0
- trainer_state.json +1617 -0
- training_loss.png +0 -0
README.md
CHANGED
|
@@ -4,6 +4,7 @@ license: apache-2.0
|
|
| 4 |
base_model: Qwen/Qwen2.5-7B-Instruct
|
| 5 |
tags:
|
| 6 |
- llama-factory
|
|
|
|
| 7 |
- generated_from_trainer
|
| 8 |
model-index:
|
| 9 |
- name: no_pipeline_math_100k
|
|
@@ -15,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 15 |
|
| 16 |
# no_pipeline_math_100k
|
| 17 |
|
| 18 |
-
This model is a fine-tuned version of [Qwen/Qwen2.5-7B-Instruct](https://huggingface.co/Qwen/Qwen2.5-7B-Instruct) on
|
| 19 |
|
| 20 |
## Model description
|
| 21 |
|
|
|
|
| 4 |
base_model: Qwen/Qwen2.5-7B-Instruct
|
| 5 |
tags:
|
| 6 |
- llama-factory
|
| 7 |
+
- full
|
| 8 |
- generated_from_trainer
|
| 9 |
model-index:
|
| 10 |
- name: no_pipeline_math_100k
|
|
|
|
| 16 |
|
| 17 |
# no_pipeline_math_100k
|
| 18 |
|
| 19 |
+
This model is a fine-tuned version of [Qwen/Qwen2.5-7B-Instruct](https://huggingface.co/Qwen/Qwen2.5-7B-Instruct) on the mlfoundations-dev/no_pipeline_math_100k dataset.
|
| 20 |
|
| 21 |
## Model description
|
| 22 |
|
all_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 4.987654320987654,
|
| 3 |
+
"total_flos": 5.796779967165497e+18,
|
| 4 |
+
"train_loss": 0.06433244016435412,
|
| 5 |
+
"train_runtime": 4471.4422,
|
| 6 |
+
"train_samples_per_second": 26.061,
|
| 7 |
+
"train_steps_per_second": 0.05
|
| 8 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 4.987654320987654,
|
| 3 |
+
"total_flos": 5.796779967165497e+18,
|
| 4 |
+
"train_loss": 0.06433244016435412,
|
| 5 |
+
"train_runtime": 4471.4422,
|
| 6 |
+
"train_samples_per_second": 26.061,
|
| 7 |
+
"train_steps_per_second": 0.05
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1617 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 4.987654320987654,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 225,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.02194787379972565,
|
| 13 |
+
"grad_norm": 6.221677266588304,
|
| 14 |
+
"learning_rate": 3.4782608695652175e-06,
|
| 15 |
+
"loss": 0.8718,
|
| 16 |
+
"step": 1
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.0438957475994513,
|
| 20 |
+
"grad_norm": 6.253264896473354,
|
| 21 |
+
"learning_rate": 6.956521739130435e-06,
|
| 22 |
+
"loss": 0.8766,
|
| 23 |
+
"step": 2
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.06584362139917696,
|
| 27 |
+
"grad_norm": 5.741674847301876,
|
| 28 |
+
"learning_rate": 1.0434782608695653e-05,
|
| 29 |
+
"loss": 0.8549,
|
| 30 |
+
"step": 3
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.0877914951989026,
|
| 34 |
+
"grad_norm": 2.3877482149850215,
|
| 35 |
+
"learning_rate": 1.391304347826087e-05,
|
| 36 |
+
"loss": 0.7607,
|
| 37 |
+
"step": 4
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.10973936899862825,
|
| 41 |
+
"grad_norm": 4.563966782136464,
|
| 42 |
+
"learning_rate": 1.739130434782609e-05,
|
| 43 |
+
"loss": 0.7707,
|
| 44 |
+
"step": 5
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.13168724279835392,
|
| 48 |
+
"grad_norm": 5.736100045190483,
|
| 49 |
+
"learning_rate": 2.0869565217391306e-05,
|
| 50 |
+
"loss": 0.7229,
|
| 51 |
+
"step": 6
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.15363511659807957,
|
| 55 |
+
"grad_norm": 5.958244309479874,
|
| 56 |
+
"learning_rate": 2.4347826086956526e-05,
|
| 57 |
+
"loss": 0.7388,
|
| 58 |
+
"step": 7
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.1755829903978052,
|
| 62 |
+
"grad_norm": 2.9204724652960246,
|
| 63 |
+
"learning_rate": 2.782608695652174e-05,
|
| 64 |
+
"loss": 0.6832,
|
| 65 |
+
"step": 8
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.19753086419753085,
|
| 69 |
+
"grad_norm": 2.3986017638483466,
|
| 70 |
+
"learning_rate": 3.130434782608696e-05,
|
| 71 |
+
"loss": 0.649,
|
| 72 |
+
"step": 9
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.2194787379972565,
|
| 76 |
+
"grad_norm": 1.6178731632508974,
|
| 77 |
+
"learning_rate": 3.478260869565218e-05,
|
| 78 |
+
"loss": 0.6185,
|
| 79 |
+
"step": 10
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.24142661179698216,
|
| 83 |
+
"grad_norm": 1.6322225805622788,
|
| 84 |
+
"learning_rate": 3.8260869565217395e-05,
|
| 85 |
+
"loss": 0.6045,
|
| 86 |
+
"step": 11
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.26337448559670784,
|
| 90 |
+
"grad_norm": 1.538396522627224,
|
| 91 |
+
"learning_rate": 4.173913043478261e-05,
|
| 92 |
+
"loss": 0.5934,
|
| 93 |
+
"step": 12
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.28532235939643347,
|
| 97 |
+
"grad_norm": 1.765556579505549,
|
| 98 |
+
"learning_rate": 4.521739130434783e-05,
|
| 99 |
+
"loss": 0.6012,
|
| 100 |
+
"step": 13
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.30727023319615915,
|
| 104 |
+
"grad_norm": 1.9039962720526176,
|
| 105 |
+
"learning_rate": 4.869565217391305e-05,
|
| 106 |
+
"loss": 0.5794,
|
| 107 |
+
"step": 14
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.3292181069958848,
|
| 111 |
+
"grad_norm": 0.9685232076545514,
|
| 112 |
+
"learning_rate": 5.217391304347826e-05,
|
| 113 |
+
"loss": 0.5783,
|
| 114 |
+
"step": 15
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.3511659807956104,
|
| 118 |
+
"grad_norm": 2.3819790183508056,
|
| 119 |
+
"learning_rate": 5.565217391304348e-05,
|
| 120 |
+
"loss": 0.586,
|
| 121 |
+
"step": 16
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.3731138545953361,
|
| 125 |
+
"grad_norm": 1.4218859481860902,
|
| 126 |
+
"learning_rate": 5.91304347826087e-05,
|
| 127 |
+
"loss": 0.5584,
|
| 128 |
+
"step": 17
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.3950617283950617,
|
| 132 |
+
"grad_norm": 1.705949751999593,
|
| 133 |
+
"learning_rate": 6.260869565217392e-05,
|
| 134 |
+
"loss": 0.5785,
|
| 135 |
+
"step": 18
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.4170096021947874,
|
| 139 |
+
"grad_norm": 1.4612385852414467,
|
| 140 |
+
"learning_rate": 6.608695652173914e-05,
|
| 141 |
+
"loss": 0.562,
|
| 142 |
+
"step": 19
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.438957475994513,
|
| 146 |
+
"grad_norm": 0.9429650218903195,
|
| 147 |
+
"learning_rate": 6.956521739130436e-05,
|
| 148 |
+
"loss": 0.5586,
|
| 149 |
+
"step": 20
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.4609053497942387,
|
| 153 |
+
"grad_norm": 1.8504062874353713,
|
| 154 |
+
"learning_rate": 7.304347826086957e-05,
|
| 155 |
+
"loss": 0.5648,
|
| 156 |
+
"step": 21
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.4828532235939643,
|
| 160 |
+
"grad_norm": 1.0359118447076778,
|
| 161 |
+
"learning_rate": 7.652173913043479e-05,
|
| 162 |
+
"loss": 0.5448,
|
| 163 |
+
"step": 22
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.50480109739369,
|
| 167 |
+
"grad_norm": 1.6864037499836835,
|
| 168 |
+
"learning_rate": 8e-05,
|
| 169 |
+
"loss": 0.5592,
|
| 170 |
+
"step": 23
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.5267489711934157,
|
| 174 |
+
"grad_norm": 1.2434910951524751,
|
| 175 |
+
"learning_rate": 7.999516253040599e-05,
|
| 176 |
+
"loss": 0.5427,
|
| 177 |
+
"step": 24
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.5486968449931413,
|
| 181 |
+
"grad_norm": 1.3000636051596768,
|
| 182 |
+
"learning_rate": 7.998065129167953e-05,
|
| 183 |
+
"loss": 0.5373,
|
| 184 |
+
"step": 25
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.5706447187928669,
|
| 188 |
+
"grad_norm": 2.684317103873129,
|
| 189 |
+
"learning_rate": 7.995646979370445e-05,
|
| 190 |
+
"loss": 0.5399,
|
| 191 |
+
"step": 26
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.5925925925925926,
|
| 195 |
+
"grad_norm": 1.4021363857201035,
|
| 196 |
+
"learning_rate": 7.992262388534378e-05,
|
| 197 |
+
"loss": 0.5434,
|
| 198 |
+
"step": 27
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.6145404663923183,
|
| 202 |
+
"grad_norm": 1.9180218278821586,
|
| 203 |
+
"learning_rate": 7.987912175302518e-05,
|
| 204 |
+
"loss": 0.5344,
|
| 205 |
+
"step": 28
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.6364883401920439,
|
| 209 |
+
"grad_norm": 1.152289451077819,
|
| 210 |
+
"learning_rate": 7.982597391876076e-05,
|
| 211 |
+
"loss": 0.531,
|
| 212 |
+
"step": 29
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.6584362139917695,
|
| 216 |
+
"grad_norm": 1.8585951370881602,
|
| 217 |
+
"learning_rate": 7.976319323760211e-05,
|
| 218 |
+
"loss": 0.5416,
|
| 219 |
+
"step": 30
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.6803840877914952,
|
| 223 |
+
"grad_norm": 1.3000646333049437,
|
| 224 |
+
"learning_rate": 7.969079489453107e-05,
|
| 225 |
+
"loss": 0.5263,
|
| 226 |
+
"step": 31
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.7023319615912208,
|
| 230 |
+
"grad_norm": 1.6442425191809935,
|
| 231 |
+
"learning_rate": 7.960879640078679e-05,
|
| 232 |
+
"loss": 0.5282,
|
| 233 |
+
"step": 32
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.7242798353909465,
|
| 237 |
+
"grad_norm": 1.3282337661300934,
|
| 238 |
+
"learning_rate": 7.951721758963028e-05,
|
| 239 |
+
"loss": 0.5125,
|
| 240 |
+
"step": 33
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.7462277091906722,
|
| 244 |
+
"grad_norm": 1.1249842886051156,
|
| 245 |
+
"learning_rate": 7.941608061154726e-05,
|
| 246 |
+
"loss": 0.5066,
|
| 247 |
+
"step": 34
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.7681755829903978,
|
| 251 |
+
"grad_norm": 0.9153543962795861,
|
| 252 |
+
"learning_rate": 7.930540992889056e-05,
|
| 253 |
+
"loss": 0.5041,
|
| 254 |
+
"step": 35
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.7901234567901234,
|
| 258 |
+
"grad_norm": 0.8714948501569295,
|
| 259 |
+
"learning_rate": 7.918523230996327e-05,
|
| 260 |
+
"loss": 0.5078,
|
| 261 |
+
"step": 36
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.8120713305898491,
|
| 265 |
+
"grad_norm": 1.17614468677297,
|
| 266 |
+
"learning_rate": 7.905557682254429e-05,
|
| 267 |
+
"loss": 0.5132,
|
| 268 |
+
"step": 37
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.8340192043895748,
|
| 272 |
+
"grad_norm": 0.9070928796685083,
|
| 273 |
+
"learning_rate": 7.891647482685751e-05,
|
| 274 |
+
"loss": 0.4954,
|
| 275 |
+
"step": 38
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.8559670781893004,
|
| 279 |
+
"grad_norm": 0.6348754426964612,
|
| 280 |
+
"learning_rate": 7.876795996798665e-05,
|
| 281 |
+
"loss": 0.5002,
|
| 282 |
+
"step": 39
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.877914951989026,
|
| 286 |
+
"grad_norm": 0.7240683284862476,
|
| 287 |
+
"learning_rate": 7.861006816773743e-05,
|
| 288 |
+
"loss": 0.4974,
|
| 289 |
+
"step": 40
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.8998628257887518,
|
| 293 |
+
"grad_norm": 0.6967632259977047,
|
| 294 |
+
"learning_rate": 7.844283761594899e-05,
|
| 295 |
+
"loss": 0.4916,
|
| 296 |
+
"step": 41
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.9218106995884774,
|
| 300 |
+
"grad_norm": 0.7523038917826653,
|
| 301 |
+
"learning_rate": 7.82663087612568e-05,
|
| 302 |
+
"loss": 0.4943,
|
| 303 |
+
"step": 42
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.943758573388203,
|
| 307 |
+
"grad_norm": 0.9009479018151323,
|
| 308 |
+
"learning_rate": 7.80805243013092e-05,
|
| 309 |
+
"loss": 0.4948,
|
| 310 |
+
"step": 43
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.9657064471879286,
|
| 314 |
+
"grad_norm": 1.1416188350869176,
|
| 315 |
+
"learning_rate": 7.788552917244002e-05,
|
| 316 |
+
"loss": 0.4963,
|
| 317 |
+
"step": 44
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.9876543209876543,
|
| 321 |
+
"grad_norm": 0.7724926247499572,
|
| 322 |
+
"learning_rate": 7.768137053879957e-05,
|
| 323 |
+
"loss": 0.4827,
|
| 324 |
+
"step": 45
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 1.00960219478738,
|
| 328 |
+
"grad_norm": 0.7378875298449712,
|
| 329 |
+
"learning_rate": 7.7468097780947e-05,
|
| 330 |
+
"loss": 0.4808,
|
| 331 |
+
"step": 46
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 1.0315500685871055,
|
| 335 |
+
"grad_norm": 0.8154669296425998,
|
| 336 |
+
"learning_rate": 7.724576248390639e-05,
|
| 337 |
+
"loss": 0.4702,
|
| 338 |
+
"step": 47
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 1.0534979423868314,
|
| 342 |
+
"grad_norm": 0.7479771108510165,
|
| 343 |
+
"learning_rate": 7.701441842468968e-05,
|
| 344 |
+
"loss": 0.457,
|
| 345 |
+
"step": 48
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 1.075445816186557,
|
| 349 |
+
"grad_norm": 0.8109214505078363,
|
| 350 |
+
"learning_rate": 7.677412155928946e-05,
|
| 351 |
+
"loss": 0.4581,
|
| 352 |
+
"step": 49
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 1.0973936899862826,
|
| 356 |
+
"grad_norm": 0.620878499140312,
|
| 357 |
+
"learning_rate": 7.652493000914476e-05,
|
| 358 |
+
"loss": 0.4643,
|
| 359 |
+
"step": 50
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 1.1193415637860082,
|
| 363 |
+
"grad_norm": 0.41252079635226235,
|
| 364 |
+
"learning_rate": 7.62669040470829e-05,
|
| 365 |
+
"loss": 0.4643,
|
| 366 |
+
"step": 51
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 1.1412894375857339,
|
| 370 |
+
"grad_norm": 0.512997391865451,
|
| 371 |
+
"learning_rate": 7.60001060827412e-05,
|
| 372 |
+
"loss": 0.4566,
|
| 373 |
+
"step": 52
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 1.1632373113854595,
|
| 377 |
+
"grad_norm": 0.5520989749480719,
|
| 378 |
+
"learning_rate": 7.572460064747167e-05,
|
| 379 |
+
"loss": 0.4553,
|
| 380 |
+
"step": 53
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 1.1851851851851851,
|
| 384 |
+
"grad_norm": 0.48682230835165313,
|
| 385 |
+
"learning_rate": 7.544045437873259e-05,
|
| 386 |
+
"loss": 0.4503,
|
| 387 |
+
"step": 54
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 1.2071330589849107,
|
| 391 |
+
"grad_norm": 0.40890214708757283,
|
| 392 |
+
"learning_rate": 7.514773600397076e-05,
|
| 393 |
+
"loss": 0.449,
|
| 394 |
+
"step": 55
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 1.2290809327846364,
|
| 398 |
+
"grad_norm": 0.4496051089290013,
|
| 399 |
+
"learning_rate": 7.484651632399802e-05,
|
| 400 |
+
"loss": 0.4486,
|
| 401 |
+
"step": 56
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 1.2510288065843622,
|
| 405 |
+
"grad_norm": 0.3385328491192872,
|
| 406 |
+
"learning_rate": 7.453686819586655e-05,
|
| 407 |
+
"loss": 0.4473,
|
| 408 |
+
"step": 57
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 1.2729766803840878,
|
| 412 |
+
"grad_norm": 0.5042801234415435,
|
| 413 |
+
"learning_rate": 7.421886651524656e-05,
|
| 414 |
+
"loss": 0.4419,
|
| 415 |
+
"step": 58
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 1.2949245541838135,
|
| 419 |
+
"grad_norm": 0.4013529469572051,
|
| 420 |
+
"learning_rate": 7.38925881983111e-05,
|
| 421 |
+
"loss": 0.4463,
|
| 422 |
+
"step": 59
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 1.316872427983539,
|
| 426 |
+
"grad_norm": 0.3526194553768413,
|
| 427 |
+
"learning_rate": 7.355811216313206e-05,
|
| 428 |
+
"loss": 0.4446,
|
| 429 |
+
"step": 60
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 1.3388203017832647,
|
| 433 |
+
"grad_norm": 0.3798310275778352,
|
| 434 |
+
"learning_rate": 7.321551931059191e-05,
|
| 435 |
+
"loss": 0.4548,
|
| 436 |
+
"step": 61
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 1.3607681755829903,
|
| 440 |
+
"grad_norm": 0.4050313385315871,
|
| 441 |
+
"learning_rate": 7.286489250481604e-05,
|
| 442 |
+
"loss": 0.4473,
|
| 443 |
+
"step": 62
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 1.382716049382716,
|
| 447 |
+
"grad_norm": 0.5307566330866562,
|
| 448 |
+
"learning_rate": 7.250631655313001e-05,
|
| 449 |
+
"loss": 0.4428,
|
| 450 |
+
"step": 63
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 1.4046639231824418,
|
| 454 |
+
"grad_norm": 0.5886517904245452,
|
| 455 |
+
"learning_rate": 7.213987818554704e-05,
|
| 456 |
+
"loss": 0.4439,
|
| 457 |
+
"step": 64
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 1.4266117969821672,
|
| 461 |
+
"grad_norm": 0.7870712747634117,
|
| 462 |
+
"learning_rate": 7.176566603379015e-05,
|
| 463 |
+
"loss": 0.4553,
|
| 464 |
+
"step": 65
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 1.448559670781893,
|
| 468 |
+
"grad_norm": 1.0045619121977905,
|
| 469 |
+
"learning_rate": 7.138377060985465e-05,
|
| 470 |
+
"loss": 0.4467,
|
| 471 |
+
"step": 66
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 1.4705075445816187,
|
| 475 |
+
"grad_norm": 1.0300458801690289,
|
| 476 |
+
"learning_rate": 7.09942842841156e-05,
|
| 477 |
+
"loss": 0.4481,
|
| 478 |
+
"step": 67
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 1.4924554183813443,
|
| 482 |
+
"grad_norm": 0.6788257591381656,
|
| 483 |
+
"learning_rate": 7.059730126298591e-05,
|
| 484 |
+
"loss": 0.4446,
|
| 485 |
+
"step": 68
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 1.51440329218107,
|
| 489 |
+
"grad_norm": 0.5581173116767874,
|
| 490 |
+
"learning_rate": 7.019291756613029e-05,
|
| 491 |
+
"loss": 0.4397,
|
| 492 |
+
"step": 69
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 1.5363511659807956,
|
| 496 |
+
"grad_norm": 0.740704532124767,
|
| 497 |
+
"learning_rate": 6.978123100324061e-05,
|
| 498 |
+
"loss": 0.4402,
|
| 499 |
+
"step": 70
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 1.5582990397805214,
|
| 503 |
+
"grad_norm": 0.7781231979915648,
|
| 504 |
+
"learning_rate": 6.936234115037842e-05,
|
| 505 |
+
"loss": 0.4429,
|
| 506 |
+
"step": 71
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 1.5802469135802468,
|
| 510 |
+
"grad_norm": 0.6101020638409748,
|
| 511 |
+
"learning_rate": 6.893634932589e-05,
|
| 512 |
+
"loss": 0.4433,
|
| 513 |
+
"step": 72
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 1.6021947873799727,
|
| 517 |
+
"grad_norm": 0.6338457181714169,
|
| 518 |
+
"learning_rate": 6.85033585659003e-05,
|
| 519 |
+
"loss": 0.4389,
|
| 520 |
+
"step": 73
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 1.624142661179698,
|
| 524 |
+
"grad_norm": 0.6492977834572081,
|
| 525 |
+
"learning_rate": 6.806347359939107e-05,
|
| 526 |
+
"loss": 0.4351,
|
| 527 |
+
"step": 74
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 1.646090534979424,
|
| 531 |
+
"grad_norm": 0.392861233479167,
|
| 532 |
+
"learning_rate": 6.761680082286988e-05,
|
| 533 |
+
"loss": 0.4399,
|
| 534 |
+
"step": 75
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 1.6680384087791496,
|
| 538 |
+
"grad_norm": 0.4720757546111325,
|
| 539 |
+
"learning_rate": 6.716344827463545e-05,
|
| 540 |
+
"loss": 0.4349,
|
| 541 |
+
"step": 76
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 1.6899862825788752,
|
| 545 |
+
"grad_norm": 0.518992673759433,
|
| 546 |
+
"learning_rate": 6.670352560864615e-05,
|
| 547 |
+
"loss": 0.4384,
|
| 548 |
+
"step": 77
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 1.7119341563786008,
|
| 552 |
+
"grad_norm": 0.3064648926060582,
|
| 553 |
+
"learning_rate": 6.62371440679976e-05,
|
| 554 |
+
"loss": 0.4298,
|
| 555 |
+
"step": 78
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 1.7338820301783264,
|
| 559 |
+
"grad_norm": 0.2952380631214821,
|
| 560 |
+
"learning_rate": 6.576441645801592e-05,
|
| 561 |
+
"loss": 0.4384,
|
| 562 |
+
"step": 79
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 1.7558299039780523,
|
| 566 |
+
"grad_norm": 0.3619917164563617,
|
| 567 |
+
"learning_rate": 6.528545711897307e-05,
|
| 568 |
+
"loss": 0.4367,
|
| 569 |
+
"step": 80
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 1.7777777777777777,
|
| 573 |
+
"grad_norm": 0.3352348275889567,
|
| 574 |
+
"learning_rate": 6.480038189843101e-05,
|
| 575 |
+
"loss": 0.4411,
|
| 576 |
+
"step": 81
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 1.7997256515775035,
|
| 580 |
+
"grad_norm": 0.3510930618032657,
|
| 581 |
+
"learning_rate": 6.430930812322127e-05,
|
| 582 |
+
"loss": 0.4386,
|
| 583 |
+
"step": 82
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 1.821673525377229,
|
| 587 |
+
"grad_norm": 0.4010609683995807,
|
| 588 |
+
"learning_rate": 6.381235457106664e-05,
|
| 589 |
+
"loss": 0.4298,
|
| 590 |
+
"step": 83
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 1.8436213991769548,
|
| 594 |
+
"grad_norm": 0.36451457683299865,
|
| 595 |
+
"learning_rate": 6.330964144185204e-05,
|
| 596 |
+
"loss": 0.4371,
|
| 597 |
+
"step": 84
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 1.8655692729766804,
|
| 601 |
+
"grad_norm": 0.27838534336167414,
|
| 602 |
+
"learning_rate": 6.280129032855132e-05,
|
| 603 |
+
"loss": 0.4305,
|
| 604 |
+
"step": 85
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 1.887517146776406,
|
| 608 |
+
"grad_norm": 0.2966462979039974,
|
| 609 |
+
"learning_rate": 6.228742418781714e-05,
|
| 610 |
+
"loss": 0.4388,
|
| 611 |
+
"step": 86
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 1.9094650205761317,
|
| 615 |
+
"grad_norm": 0.26944878326607974,
|
| 616 |
+
"learning_rate": 6.17681673102411e-05,
|
| 617 |
+
"loss": 0.4345,
|
| 618 |
+
"step": 87
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 1.9314128943758573,
|
| 622 |
+
"grad_norm": 0.2712568832140498,
|
| 623 |
+
"learning_rate": 6.1243645290291e-05,
|
| 624 |
+
"loss": 0.4308,
|
| 625 |
+
"step": 88
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 1.9533607681755831,
|
| 629 |
+
"grad_norm": 0.2866166967589882,
|
| 630 |
+
"learning_rate": 6.0713984995933016e-05,
|
| 631 |
+
"loss": 0.4325,
|
| 632 |
+
"step": 89
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 1.9753086419753085,
|
| 636 |
+
"grad_norm": 0.23491982047748894,
|
| 637 |
+
"learning_rate": 6.01793145379456e-05,
|
| 638 |
+
"loss": 0.4321,
|
| 639 |
+
"step": 90
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 1.9972565157750344,
|
| 643 |
+
"grad_norm": 0.205586198763508,
|
| 644 |
+
"learning_rate": 5.9639763238932893e-05,
|
| 645 |
+
"loss": 0.4345,
|
| 646 |
+
"step": 91
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 2.01920438957476,
|
| 650 |
+
"grad_norm": 0.27481178105933207,
|
| 651 |
+
"learning_rate": 5.909546160204508e-05,
|
| 652 |
+
"loss": 0.4013,
|
| 653 |
+
"step": 92
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 2.0411522633744856,
|
| 657 |
+
"grad_norm": 0.30806418646748673,
|
| 658 |
+
"learning_rate": 5.8546541279413094e-05,
|
| 659 |
+
"loss": 0.3976,
|
| 660 |
+
"step": 93
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 2.063100137174211,
|
| 664 |
+
"grad_norm": 0.3291669273065511,
|
| 665 |
+
"learning_rate": 5.799313504030545e-05,
|
| 666 |
+
"loss": 0.3947,
|
| 667 |
+
"step": 94
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 2.085048010973937,
|
| 671 |
+
"grad_norm": 0.37474732162243923,
|
| 672 |
+
"learning_rate": 5.743537673901485e-05,
|
| 673 |
+
"loss": 0.3971,
|
| 674 |
+
"step": 95
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 2.1069958847736627,
|
| 678 |
+
"grad_norm": 0.46251544205958794,
|
| 679 |
+
"learning_rate": 5.68734012824825e-05,
|
| 680 |
+
"loss": 0.396,
|
| 681 |
+
"step": 96
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 2.128943758573388,
|
| 685 |
+
"grad_norm": 0.54656137847099,
|
| 686 |
+
"learning_rate": 5.6307344597667555e-05,
|
| 687 |
+
"loss": 0.4027,
|
| 688 |
+
"step": 97
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 2.150891632373114,
|
| 692 |
+
"grad_norm": 0.6049235984601056,
|
| 693 |
+
"learning_rate": 5.5737343598670104e-05,
|
| 694 |
+
"loss": 0.4005,
|
| 695 |
+
"step": 98
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 2.1728395061728394,
|
| 699 |
+
"grad_norm": 0.5234566946423048,
|
| 700 |
+
"learning_rate": 5.5163536153615185e-05,
|
| 701 |
+
"loss": 0.3955,
|
| 702 |
+
"step": 99
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 2.1947873799725652,
|
| 706 |
+
"grad_norm": 0.4013877298974358,
|
| 707 |
+
"learning_rate": 5.4586061051306204e-05,
|
| 708 |
+
"loss": 0.4028,
|
| 709 |
+
"step": 100
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 2.2167352537722906,
|
| 713 |
+
"grad_norm": 0.3845465210881421,
|
| 714 |
+
"learning_rate": 5.4005057967655634e-05,
|
| 715 |
+
"loss": 0.3976,
|
| 716 |
+
"step": 101
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 2.2386831275720165,
|
| 720 |
+
"grad_norm": 0.408459585720323,
|
| 721 |
+
"learning_rate": 5.3420667431901e-05,
|
| 722 |
+
"loss": 0.3975,
|
| 723 |
+
"step": 102
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 2.260631001371742,
|
| 727 |
+
"grad_norm": 0.37864175367916847,
|
| 728 |
+
"learning_rate": 5.283303079261471e-05,
|
| 729 |
+
"loss": 0.3936,
|
| 730 |
+
"step": 103
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 2.2825788751714677,
|
| 734 |
+
"grad_norm": 0.3827028198880068,
|
| 735 |
+
"learning_rate": 5.22422901835155e-05,
|
| 736 |
+
"loss": 0.3936,
|
| 737 |
+
"step": 104
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 2.3045267489711936,
|
| 741 |
+
"grad_norm": 0.3484431146343949,
|
| 742 |
+
"learning_rate": 5.164858848909009e-05,
|
| 743 |
+
"loss": 0.388,
|
| 744 |
+
"step": 105
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 2.326474622770919,
|
| 748 |
+
"grad_norm": 0.3095716919680074,
|
| 749 |
+
"learning_rate": 5.1052069310033216e-05,
|
| 750 |
+
"loss": 0.3988,
|
| 751 |
+
"step": 106
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 2.348422496570645,
|
| 755 |
+
"grad_norm": 0.39008928135047216,
|
| 756 |
+
"learning_rate": 5.0452876928514434e-05,
|
| 757 |
+
"loss": 0.3966,
|
| 758 |
+
"step": 107
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 2.3703703703703702,
|
| 762 |
+
"grad_norm": 0.44827732087715033,
|
| 763 |
+
"learning_rate": 4.9851156273280064e-05,
|
| 764 |
+
"loss": 0.3953,
|
| 765 |
+
"step": 108
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 2.392318244170096,
|
| 769 |
+
"grad_norm": 0.2796243136669204,
|
| 770 |
+
"learning_rate": 4.92470528845988e-05,
|
| 771 |
+
"loss": 0.3964,
|
| 772 |
+
"step": 109
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 2.4142661179698215,
|
| 776 |
+
"grad_norm": 0.3186398495286058,
|
| 777 |
+
"learning_rate": 4.8640712879059354e-05,
|
| 778 |
+
"loss": 0.3923,
|
| 779 |
+
"step": 110
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 2.4362139917695473,
|
| 783 |
+
"grad_norm": 0.35322375158512703,
|
| 784 |
+
"learning_rate": 4.8032282914228743e-05,
|
| 785 |
+
"loss": 0.3886,
|
| 786 |
+
"step": 111
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 2.4581618655692727,
|
| 790 |
+
"grad_norm": 0.18461936703017578,
|
| 791 |
+
"learning_rate": 4.742191015317974e-05,
|
| 792 |
+
"loss": 0.3917,
|
| 793 |
+
"step": 112
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 2.4801097393689986,
|
| 797 |
+
"grad_norm": 0.33683641542461873,
|
| 798 |
+
"learning_rate": 4.680974222889595e-05,
|
| 799 |
+
"loss": 0.394,
|
| 800 |
+
"step": 113
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 2.5020576131687244,
|
| 804 |
+
"grad_norm": 0.3169424542788605,
|
| 805 |
+
"learning_rate": 4.6195927208563394e-05,
|
| 806 |
+
"loss": 0.3871,
|
| 807 |
+
"step": 114
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 2.52400548696845,
|
| 811 |
+
"grad_norm": 0.15430100844009037,
|
| 812 |
+
"learning_rate": 4.558061355775693e-05,
|
| 813 |
+
"loss": 0.3857,
|
| 814 |
+
"step": 115
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 2.5459533607681757,
|
| 818 |
+
"grad_norm": 0.2820303722197663,
|
| 819 |
+
"learning_rate": 4.496395010453038e-05,
|
| 820 |
+
"loss": 0.3886,
|
| 821 |
+
"step": 116
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 2.567901234567901,
|
| 825 |
+
"grad_norm": 0.24770492735610264,
|
| 826 |
+
"learning_rate": 4.4346086003418985e-05,
|
| 827 |
+
"loss": 0.3882,
|
| 828 |
+
"step": 117
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 2.589849108367627,
|
| 832 |
+
"grad_norm": 0.15993713137779939,
|
| 833 |
+
"learning_rate": 4.372717069936287e-05,
|
| 834 |
+
"loss": 0.3891,
|
| 835 |
+
"step": 118
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 2.611796982167353,
|
| 839 |
+
"grad_norm": 0.24474244306163354,
|
| 840 |
+
"learning_rate": 4.310735389156026e-05,
|
| 841 |
+
"loss": 0.392,
|
| 842 |
+
"step": 119
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 2.633744855967078,
|
| 846 |
+
"grad_norm": 0.1905642453010369,
|
| 847 |
+
"learning_rate": 4.248678549725923e-05,
|
| 848 |
+
"loss": 0.3917,
|
| 849 |
+
"step": 120
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 2.6556927297668036,
|
| 853 |
+
"grad_norm": 0.15617551076050876,
|
| 854 |
+
"learning_rate": 4.18656156154967e-05,
|
| 855 |
+
"loss": 0.3945,
|
| 856 |
+
"step": 121
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 2.6776406035665294,
|
| 860 |
+
"grad_norm": 0.21098249998156268,
|
| 861 |
+
"learning_rate": 4.124399449079348e-05,
|
| 862 |
+
"loss": 0.3866,
|
| 863 |
+
"step": 122
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 2.6995884773662553,
|
| 867 |
+
"grad_norm": 0.17151651983910782,
|
| 868 |
+
"learning_rate": 4.0622072476814045e-05,
|
| 869 |
+
"loss": 0.3872,
|
| 870 |
+
"step": 123
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 2.7215363511659807,
|
| 874 |
+
"grad_norm": 0.14636727205415834,
|
| 875 |
+
"learning_rate": 4e-05,
|
| 876 |
+
"loss": 0.3919,
|
| 877 |
+
"step": 124
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 2.7434842249657065,
|
| 881 |
+
"grad_norm": 0.21883566171534793,
|
| 882 |
+
"learning_rate": 3.937792752318597e-05,
|
| 883 |
+
"loss": 0.3854,
|
| 884 |
+
"step": 125
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 2.765432098765432,
|
| 888 |
+
"grad_norm": 0.15504692956699287,
|
| 889 |
+
"learning_rate": 3.8756005509206535e-05,
|
| 890 |
+
"loss": 0.3883,
|
| 891 |
+
"step": 126
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 2.787379972565158,
|
| 895 |
+
"grad_norm": 0.21026818171396724,
|
| 896 |
+
"learning_rate": 3.81343843845033e-05,
|
| 897 |
+
"loss": 0.3883,
|
| 898 |
+
"step": 127
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 2.8093278463648836,
|
| 902 |
+
"grad_norm": 0.1846577189184653,
|
| 903 |
+
"learning_rate": 3.751321450274078e-05,
|
| 904 |
+
"loss": 0.3921,
|
| 905 |
+
"step": 128
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 2.831275720164609,
|
| 909 |
+
"grad_norm": 0.16874522256869426,
|
| 910 |
+
"learning_rate": 3.689264610843975e-05,
|
| 911 |
+
"loss": 0.3911,
|
| 912 |
+
"step": 129
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 2.8532235939643344,
|
| 916 |
+
"grad_norm": 0.16523585002563423,
|
| 917 |
+
"learning_rate": 3.627282930063714e-05,
|
| 918 |
+
"loss": 0.3886,
|
| 919 |
+
"step": 130
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 2.8751714677640603,
|
| 923 |
+
"grad_norm": 0.14959343517939727,
|
| 924 |
+
"learning_rate": 3.565391399658102e-05,
|
| 925 |
+
"loss": 0.392,
|
| 926 |
+
"step": 131
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 2.897119341563786,
|
| 930 |
+
"grad_norm": 0.13667256750591955,
|
| 931 |
+
"learning_rate": 3.503604989546963e-05,
|
| 932 |
+
"loss": 0.3846,
|
| 933 |
+
"step": 132
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 2.9190672153635115,
|
| 937 |
+
"grad_norm": 0.14202188950732816,
|
| 938 |
+
"learning_rate": 3.4419386442243084e-05,
|
| 939 |
+
"loss": 0.3902,
|
| 940 |
+
"step": 133
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 2.9410150891632374,
|
| 944 |
+
"grad_norm": 0.14821759155695716,
|
| 945 |
+
"learning_rate": 3.380407279143661e-05,
|
| 946 |
+
"loss": 0.3867,
|
| 947 |
+
"step": 134
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 2.962962962962963,
|
| 951 |
+
"grad_norm": 0.1667332312440918,
|
| 952 |
+
"learning_rate": 3.3190257771104055e-05,
|
| 953 |
+
"loss": 0.3927,
|
| 954 |
+
"step": 135
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 2.9849108367626886,
|
| 958 |
+
"grad_norm": 0.14452591810498722,
|
| 959 |
+
"learning_rate": 3.257808984682027e-05,
|
| 960 |
+
"loss": 0.3855,
|
| 961 |
+
"step": 136
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 3.006858710562414,
|
| 965 |
+
"grad_norm": 0.18251829909357556,
|
| 966 |
+
"learning_rate": 3.196771708577127e-05,
|
| 967 |
+
"loss": 0.3837,
|
| 968 |
+
"step": 137
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 3.02880658436214,
|
| 972 |
+
"grad_norm": 0.17364339659347747,
|
| 973 |
+
"learning_rate": 3.135928712094067e-05,
|
| 974 |
+
"loss": 0.359,
|
| 975 |
+
"step": 138
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 3.0507544581618657,
|
| 979 |
+
"grad_norm": 0.17747923559690468,
|
| 980 |
+
"learning_rate": 3.075294711540123e-05,
|
| 981 |
+
"loss": 0.3557,
|
| 982 |
+
"step": 139
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 3.072702331961591,
|
| 986 |
+
"grad_norm": 0.1787283180499878,
|
| 987 |
+
"learning_rate": 3.0148843726719953e-05,
|
| 988 |
+
"loss": 0.3516,
|
| 989 |
+
"step": 140
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 3.094650205761317,
|
| 993 |
+
"grad_norm": 0.19311413443289224,
|
| 994 |
+
"learning_rate": 2.9547123071485586e-05,
|
| 995 |
+
"loss": 0.3532,
|
| 996 |
+
"step": 141
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 3.1165980795610424,
|
| 1000 |
+
"grad_norm": 0.15814186428435004,
|
| 1001 |
+
"learning_rate": 2.8947930689966798e-05,
|
| 1002 |
+
"loss": 0.3598,
|
| 1003 |
+
"step": 142
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 3.1385459533607682,
|
| 1007 |
+
"grad_norm": 0.17222178104775263,
|
| 1008 |
+
"learning_rate": 2.8351411510909926e-05,
|
| 1009 |
+
"loss": 0.3567,
|
| 1010 |
+
"step": 143
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 3.1604938271604937,
|
| 1014 |
+
"grad_norm": 0.15304672551321433,
|
| 1015 |
+
"learning_rate": 2.7757709816484512e-05,
|
| 1016 |
+
"loss": 0.3634,
|
| 1017 |
+
"step": 144
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 3.1824417009602195,
|
| 1021 |
+
"grad_norm": 0.15457992995371023,
|
| 1022 |
+
"learning_rate": 2.71669692073853e-05,
|
| 1023 |
+
"loss": 0.3542,
|
| 1024 |
+
"step": 145
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 3.2043895747599453,
|
| 1028 |
+
"grad_norm": 0.1479490816574568,
|
| 1029 |
+
"learning_rate": 2.6579332568099006e-05,
|
| 1030 |
+
"loss": 0.3576,
|
| 1031 |
+
"step": 146
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 3.2263374485596708,
|
| 1035 |
+
"grad_norm": 0.15469338097636914,
|
| 1036 |
+
"learning_rate": 2.5994942032344376e-05,
|
| 1037 |
+
"loss": 0.357,
|
| 1038 |
+
"step": 147
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 3.2482853223593966,
|
| 1042 |
+
"grad_norm": 0.13891659760939676,
|
| 1043 |
+
"learning_rate": 2.54139389486938e-05,
|
| 1044 |
+
"loss": 0.3559,
|
| 1045 |
+
"step": 148
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 3.270233196159122,
|
| 1049 |
+
"grad_norm": 0.143615476750949,
|
| 1050 |
+
"learning_rate": 2.4836463846384832e-05,
|
| 1051 |
+
"loss": 0.3529,
|
| 1052 |
+
"step": 149
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 3.292181069958848,
|
| 1056 |
+
"grad_norm": 0.1420809451110692,
|
| 1057 |
+
"learning_rate": 2.4262656401329913e-05,
|
| 1058 |
+
"loss": 0.359,
|
| 1059 |
+
"step": 150
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 3.3141289437585733,
|
| 1063 |
+
"grad_norm": 0.14089610783501166,
|
| 1064 |
+
"learning_rate": 2.3692655402332455e-05,
|
| 1065 |
+
"loss": 0.3565,
|
| 1066 |
+
"step": 151
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 3.336076817558299,
|
| 1070 |
+
"grad_norm": 0.1402483457342098,
|
| 1071 |
+
"learning_rate": 2.3126598717517514e-05,
|
| 1072 |
+
"loss": 0.3508,
|
| 1073 |
+
"step": 152
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 3.3580246913580245,
|
| 1077 |
+
"grad_norm": 0.12317925977087589,
|
| 1078 |
+
"learning_rate": 2.256462326098516e-05,
|
| 1079 |
+
"loss": 0.3558,
|
| 1080 |
+
"step": 153
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 3.3799725651577504,
|
| 1084 |
+
"grad_norm": 0.1452523632260013,
|
| 1085 |
+
"learning_rate": 2.200686495969457e-05,
|
| 1086 |
+
"loss": 0.3546,
|
| 1087 |
+
"step": 154
|
| 1088 |
+
},
|
| 1089 |
+
{
|
| 1090 |
+
"epoch": 3.401920438957476,
|
| 1091 |
+
"grad_norm": 0.11788796867275445,
|
| 1092 |
+
"learning_rate": 2.1453458720586902e-05,
|
| 1093 |
+
"loss": 0.3562,
|
| 1094 |
+
"step": 155
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 3.4238683127572016,
|
| 1098 |
+
"grad_norm": 0.14156518711907154,
|
| 1099 |
+
"learning_rate": 2.0904538397954913e-05,
|
| 1100 |
+
"loss": 0.3595,
|
| 1101 |
+
"step": 156
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 3.4458161865569275,
|
| 1105 |
+
"grad_norm": 0.1152521891751902,
|
| 1106 |
+
"learning_rate": 2.0360236761067117e-05,
|
| 1107 |
+
"loss": 0.3572,
|
| 1108 |
+
"step": 157
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"epoch": 3.467764060356653,
|
| 1112 |
+
"grad_norm": 0.1292120630664868,
|
| 1113 |
+
"learning_rate": 1.9820685462054413e-05,
|
| 1114 |
+
"loss": 0.3608,
|
| 1115 |
+
"step": 158
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 3.4897119341563787,
|
| 1119 |
+
"grad_norm": 0.12307673484565831,
|
| 1120 |
+
"learning_rate": 1.9286015004066984e-05,
|
| 1121 |
+
"loss": 0.3571,
|
| 1122 |
+
"step": 159
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 3.511659807956104,
|
| 1126 |
+
"grad_norm": 0.11387530947074216,
|
| 1127 |
+
"learning_rate": 1.8756354709708998e-05,
|
| 1128 |
+
"loss": 0.3593,
|
| 1129 |
+
"step": 160
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 3.53360768175583,
|
| 1133 |
+
"grad_norm": 0.1276131097306211,
|
| 1134 |
+
"learning_rate": 1.8231832689758903e-05,
|
| 1135 |
+
"loss": 0.3528,
|
| 1136 |
+
"step": 161
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 3.5555555555555554,
|
| 1140 |
+
"grad_norm": 0.11562929678821103,
|
| 1141 |
+
"learning_rate": 1.771257581218287e-05,
|
| 1142 |
+
"loss": 0.3562,
|
| 1143 |
+
"step": 162
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 3.577503429355281,
|
| 1147 |
+
"grad_norm": 0.11195216588496365,
|
| 1148 |
+
"learning_rate": 1.7198709671448696e-05,
|
| 1149 |
+
"loss": 0.3568,
|
| 1150 |
+
"step": 163
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 3.599451303155007,
|
| 1154 |
+
"grad_norm": 0.10790427870554664,
|
| 1155 |
+
"learning_rate": 1.6690358558147967e-05,
|
| 1156 |
+
"loss": 0.3522,
|
| 1157 |
+
"step": 164
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 3.6213991769547325,
|
| 1161 |
+
"grad_norm": 0.11612616520204616,
|
| 1162 |
+
"learning_rate": 1.6187645428933372e-05,
|
| 1163 |
+
"loss": 0.3584,
|
| 1164 |
+
"step": 165
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 3.6433470507544583,
|
| 1168 |
+
"grad_norm": 0.11400211648602655,
|
| 1169 |
+
"learning_rate": 1.5690691876778746e-05,
|
| 1170 |
+
"loss": 0.3622,
|
| 1171 |
+
"step": 166
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 3.6652949245541837,
|
| 1175 |
+
"grad_norm": 0.1065206298660862,
|
| 1176 |
+
"learning_rate": 1.5199618101569003e-05,
|
| 1177 |
+
"loss": 0.3544,
|
| 1178 |
+
"step": 167
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 3.6872427983539096,
|
| 1182 |
+
"grad_norm": 0.10461159135503716,
|
| 1183 |
+
"learning_rate": 1.4714542881026947e-05,
|
| 1184 |
+
"loss": 0.3519,
|
| 1185 |
+
"step": 168
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 3.709190672153635,
|
| 1189 |
+
"grad_norm": 0.12285024075507715,
|
| 1190 |
+
"learning_rate": 1.4235583541984092e-05,
|
| 1191 |
+
"loss": 0.3621,
|
| 1192 |
+
"step": 169
|
| 1193 |
+
},
|
| 1194 |
+
{
|
| 1195 |
+
"epoch": 3.731138545953361,
|
| 1196 |
+
"grad_norm": 0.10841173453177154,
|
| 1197 |
+
"learning_rate": 1.3762855932002404e-05,
|
| 1198 |
+
"loss": 0.3554,
|
| 1199 |
+
"step": 170
|
| 1200 |
+
},
|
| 1201 |
+
{
|
| 1202 |
+
"epoch": 3.753086419753086,
|
| 1203 |
+
"grad_norm": 0.10185093830948505,
|
| 1204 |
+
"learning_rate": 1.3296474391353854e-05,
|
| 1205 |
+
"loss": 0.3602,
|
| 1206 |
+
"step": 171
|
| 1207 |
+
},
|
| 1208 |
+
{
|
| 1209 |
+
"epoch": 3.775034293552812,
|
| 1210 |
+
"grad_norm": 0.10890679245723875,
|
| 1211 |
+
"learning_rate": 1.2836551725364572e-05,
|
| 1212 |
+
"loss": 0.3489,
|
| 1213 |
+
"step": 172
|
| 1214 |
+
},
|
| 1215 |
+
{
|
| 1216 |
+
"epoch": 3.796982167352538,
|
| 1217 |
+
"grad_norm": 0.10191643511945396,
|
| 1218 |
+
"learning_rate": 1.2383199177130135e-05,
|
| 1219 |
+
"loss": 0.3536,
|
| 1220 |
+
"step": 173
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 3.8189300411522633,
|
| 1224 |
+
"grad_norm": 0.09752716587838449,
|
| 1225 |
+
"learning_rate": 1.1936526400608938e-05,
|
| 1226 |
+
"loss": 0.3517,
|
| 1227 |
+
"step": 174
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"epoch": 3.840877914951989,
|
| 1231 |
+
"grad_norm": 0.09567252814321132,
|
| 1232 |
+
"learning_rate": 1.1496641434099725e-05,
|
| 1233 |
+
"loss": 0.3499,
|
| 1234 |
+
"step": 175
|
| 1235 |
+
},
|
| 1236 |
+
{
|
| 1237 |
+
"epoch": 3.8628257887517146,
|
| 1238 |
+
"grad_norm": 0.10384979007917372,
|
| 1239 |
+
"learning_rate": 1.1063650674110011e-05,
|
| 1240 |
+
"loss": 0.3562,
|
| 1241 |
+
"step": 176
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 3.8847736625514404,
|
| 1245 |
+
"grad_norm": 0.09268596571493516,
|
| 1246 |
+
"learning_rate": 1.0637658849621593e-05,
|
| 1247 |
+
"loss": 0.3521,
|
| 1248 |
+
"step": 177
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 3.9067215363511663,
|
| 1252 |
+
"grad_norm": 0.08906274897668208,
|
| 1253 |
+
"learning_rate": 1.0218768996759399e-05,
|
| 1254 |
+
"loss": 0.3528,
|
| 1255 |
+
"step": 178
|
| 1256 |
+
},
|
| 1257 |
+
{
|
| 1258 |
+
"epoch": 3.9286694101508917,
|
| 1259 |
+
"grad_norm": 0.09311271485149279,
|
| 1260 |
+
"learning_rate": 9.807082433869727e-06,
|
| 1261 |
+
"loss": 0.359,
|
| 1262 |
+
"step": 179
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 3.950617283950617,
|
| 1266 |
+
"grad_norm": 0.09966102704447338,
|
| 1267 |
+
"learning_rate": 9.402698737014098e-06,
|
| 1268 |
+
"loss": 0.3643,
|
| 1269 |
+
"step": 180
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 3.972565157750343,
|
| 1273 |
+
"grad_norm": 0.08428095680890077,
|
| 1274 |
+
"learning_rate": 9.005715715884409e-06,
|
| 1275 |
+
"loss": 0.3564,
|
| 1276 |
+
"step": 181
|
| 1277 |
+
},
|
| 1278 |
+
{
|
| 1279 |
+
"epoch": 3.9945130315500688,
|
| 1280 |
+
"grad_norm": 0.098824142446022,
|
| 1281 |
+
"learning_rate": 8.616229390145361e-06,
|
| 1282 |
+
"loss": 0.3561,
|
| 1283 |
+
"step": 182
|
| 1284 |
+
},
|
| 1285 |
+
{
|
| 1286 |
+
"epoch": 4.065843621399177,
|
| 1287 |
+
"grad_norm": 0.13982352922009822,
|
| 1288 |
+
"learning_rate": 8.23433396620986e-06,
|
| 1289 |
+
"loss": 0.341,
|
| 1290 |
+
"step": 183
|
| 1291 |
+
},
|
| 1292 |
+
{
|
| 1293 |
+
"epoch": 4.0877914951989025,
|
| 1294 |
+
"grad_norm": 0.11357790917439833,
|
| 1295 |
+
"learning_rate": 7.86012181445297e-06,
|
| 1296 |
+
"loss": 0.3427,
|
| 1297 |
+
"step": 184
|
| 1298 |
+
},
|
| 1299 |
+
{
|
| 1300 |
+
"epoch": 4.109739368998628,
|
| 1301 |
+
"grad_norm": 0.09508337442980856,
|
| 1302 |
+
"learning_rate": 7.4936834468699945e-06,
|
| 1303 |
+
"loss": 0.337,
|
| 1304 |
+
"step": 185
|
| 1305 |
+
},
|
| 1306 |
+
{
|
| 1307 |
+
"epoch": 4.131687242798354,
|
| 1308 |
+
"grad_norm": 0.10345088268864072,
|
| 1309 |
+
"learning_rate": 7.135107495183975e-06,
|
| 1310 |
+
"loss": 0.3345,
|
| 1311 |
+
"step": 186
|
| 1312 |
+
},
|
| 1313 |
+
{
|
| 1314 |
+
"epoch": 4.153635116598079,
|
| 1315 |
+
"grad_norm": 0.1048793029567903,
|
| 1316 |
+
"learning_rate": 6.784480689408099e-06,
|
| 1317 |
+
"loss": 0.3297,
|
| 1318 |
+
"step": 187
|
| 1319 |
+
},
|
| 1320 |
+
{
|
| 1321 |
+
"epoch": 4.175582990397805,
|
| 1322 |
+
"grad_norm": 0.10481831928351991,
|
| 1323 |
+
"learning_rate": 6.441887836867962e-06,
|
| 1324 |
+
"loss": 0.3414,
|
| 1325 |
+
"step": 188
|
| 1326 |
+
},
|
| 1327 |
+
{
|
| 1328 |
+
"epoch": 4.197530864197531,
|
| 1329 |
+
"grad_norm": 0.11340261438354692,
|
| 1330 |
+
"learning_rate": 6.107411801688905e-06,
|
| 1331 |
+
"loss": 0.3387,
|
| 1332 |
+
"step": 189
|
| 1333 |
+
},
|
| 1334 |
+
{
|
| 1335 |
+
"epoch": 4.219478737997257,
|
| 1336 |
+
"grad_norm": 0.10558490986119158,
|
| 1337 |
+
"learning_rate": 5.781133484753451e-06,
|
| 1338 |
+
"loss": 0.3386,
|
| 1339 |
+
"step": 190
|
| 1340 |
+
},
|
| 1341 |
+
{
|
| 1342 |
+
"epoch": 4.2414266117969825,
|
| 1343 |
+
"grad_norm": 0.09563394121242606,
|
| 1344 |
+
"learning_rate": 5.463131804133461e-06,
|
| 1345 |
+
"loss": 0.3392,
|
| 1346 |
+
"step": 191
|
| 1347 |
+
},
|
| 1348 |
+
{
|
| 1349 |
+
"epoch": 4.2633744855967075,
|
| 1350 |
+
"grad_norm": 0.097884687531601,
|
| 1351 |
+
"learning_rate": 5.1534836760019824e-06,
|
| 1352 |
+
"loss": 0.3406,
|
| 1353 |
+
"step": 192
|
| 1354 |
+
},
|
| 1355 |
+
{
|
| 1356 |
+
"epoch": 4.285322359396433,
|
| 1357 |
+
"grad_norm": 0.09586101042431322,
|
| 1358 |
+
"learning_rate": 4.852263996029259e-06,
|
| 1359 |
+
"loss": 0.3347,
|
| 1360 |
+
"step": 193
|
| 1361 |
+
},
|
| 1362 |
+
{
|
| 1363 |
+
"epoch": 4.307270233196159,
|
| 1364 |
+
"grad_norm": 0.09449695957524798,
|
| 1365 |
+
"learning_rate": 4.559545621267414e-06,
|
| 1366 |
+
"loss": 0.3316,
|
| 1367 |
+
"step": 194
|
| 1368 |
+
},
|
| 1369 |
+
{
|
| 1370 |
+
"epoch": 4.329218106995885,
|
| 1371 |
+
"grad_norm": 0.09218455432645911,
|
| 1372 |
+
"learning_rate": 4.275399352528342e-06,
|
| 1373 |
+
"loss": 0.3382,
|
| 1374 |
+
"step": 195
|
| 1375 |
+
},
|
| 1376 |
+
{
|
| 1377 |
+
"epoch": 4.35116598079561,
|
| 1378 |
+
"grad_norm": 0.09122205514695,
|
| 1379 |
+
"learning_rate": 3.999893917258799e-06,
|
| 1380 |
+
"loss": 0.3421,
|
| 1381 |
+
"step": 196
|
| 1382 |
+
},
|
| 1383 |
+
{
|
| 1384 |
+
"epoch": 4.373113854595336,
|
| 1385 |
+
"grad_norm": 0.08884394525292823,
|
| 1386 |
+
"learning_rate": 3.733095952917101e-06,
|
| 1387 |
+
"loss": 0.3359,
|
| 1388 |
+
"step": 197
|
| 1389 |
+
},
|
| 1390 |
+
{
|
| 1391 |
+
"epoch": 4.395061728395062,
|
| 1392 |
+
"grad_norm": 0.08788073543014369,
|
| 1393 |
+
"learning_rate": 3.4750699908552464e-06,
|
| 1394 |
+
"loss": 0.3348,
|
| 1395 |
+
"step": 198
|
| 1396 |
+
},
|
| 1397 |
+
{
|
| 1398 |
+
"epoch": 4.4170096021947876,
|
| 1399 |
+
"grad_norm": 0.0851504561450446,
|
| 1400 |
+
"learning_rate": 3.225878440710544e-06,
|
| 1401 |
+
"loss": 0.338,
|
| 1402 |
+
"step": 199
|
| 1403 |
+
},
|
| 1404 |
+
{
|
| 1405 |
+
"epoch": 4.438957475994513,
|
| 1406 |
+
"grad_norm": 0.08806355910065146,
|
| 1407 |
+
"learning_rate": 2.9855815753103436e-06,
|
| 1408 |
+
"loss": 0.3408,
|
| 1409 |
+
"step": 200
|
| 1410 |
+
},
|
| 1411 |
+
{
|
| 1412 |
+
"epoch": 4.460905349794238,
|
| 1413 |
+
"grad_norm": 0.08443831027497156,
|
| 1414 |
+
"learning_rate": 2.754237516093623e-06,
|
| 1415 |
+
"loss": 0.3336,
|
| 1416 |
+
"step": 201
|
| 1417 |
+
},
|
| 1418 |
+
{
|
| 1419 |
+
"epoch": 4.482853223593964,
|
| 1420 |
+
"grad_norm": 0.0846796048507417,
|
| 1421 |
+
"learning_rate": 2.5319022190529997e-06,
|
| 1422 |
+
"loss": 0.3347,
|
| 1423 |
+
"step": 202
|
| 1424 |
+
},
|
| 1425 |
+
{
|
| 1426 |
+
"epoch": 4.50480109739369,
|
| 1427 |
+
"grad_norm": 0.08433609787768302,
|
| 1428 |
+
"learning_rate": 2.3186294612004365e-06,
|
| 1429 |
+
"loss": 0.3359,
|
| 1430 |
+
"step": 203
|
| 1431 |
+
},
|
| 1432 |
+
{
|
| 1433 |
+
"epoch": 4.526748971193416,
|
| 1434 |
+
"grad_norm": 0.08573881642215767,
|
| 1435 |
+
"learning_rate": 2.1144708275599955e-06,
|
| 1436 |
+
"loss": 0.3411,
|
| 1437 |
+
"step": 204
|
| 1438 |
+
},
|
| 1439 |
+
{
|
| 1440 |
+
"epoch": 4.548696844993142,
|
| 1441 |
+
"grad_norm": 0.08265953354626032,
|
| 1442 |
+
"learning_rate": 1.9194756986908025e-06,
|
| 1443 |
+
"loss": 0.3351,
|
| 1444 |
+
"step": 205
|
| 1445 |
+
},
|
| 1446 |
+
{
|
| 1447 |
+
"epoch": 4.570644718792867,
|
| 1448 |
+
"grad_norm": 0.07968266677719599,
|
| 1449 |
+
"learning_rate": 1.7336912387432115e-06,
|
| 1450 |
+
"loss": 0.3329,
|
| 1451 |
+
"step": 206
|
| 1452 |
+
},
|
| 1453 |
+
{
|
| 1454 |
+
"epoch": 4.592592592592593,
|
| 1455 |
+
"grad_norm": 0.07672083627731244,
|
| 1456 |
+
"learning_rate": 1.5571623840510185e-06,
|
| 1457 |
+
"loss": 0.3344,
|
| 1458 |
+
"step": 207
|
| 1459 |
+
},
|
| 1460 |
+
{
|
| 1461 |
+
"epoch": 4.614540466392318,
|
| 1462 |
+
"grad_norm": 0.07767236853772148,
|
| 1463 |
+
"learning_rate": 1.3899318322625744e-06,
|
| 1464 |
+
"loss": 0.3353,
|
| 1465 |
+
"step": 208
|
| 1466 |
+
},
|
| 1467 |
+
{
|
| 1468 |
+
"epoch": 4.636488340192044,
|
| 1469 |
+
"grad_norm": 0.079617742448113,
|
| 1470 |
+
"learning_rate": 1.2320400320133551e-06,
|
| 1471 |
+
"loss": 0.3351,
|
| 1472 |
+
"step": 209
|
| 1473 |
+
},
|
| 1474 |
+
{
|
| 1475 |
+
"epoch": 4.658436213991769,
|
| 1476 |
+
"grad_norm": 0.07817205047145995,
|
| 1477 |
+
"learning_rate": 1.0835251731425013e-06,
|
| 1478 |
+
"loss": 0.3389,
|
| 1479 |
+
"step": 210
|
| 1480 |
+
},
|
| 1481 |
+
{
|
| 1482 |
+
"epoch": 4.680384087791495,
|
| 1483 |
+
"grad_norm": 0.079996413423676,
|
| 1484 |
+
"learning_rate": 9.444231774557199e-07,
|
| 1485 |
+
"loss": 0.3407,
|
| 1486 |
+
"step": 211
|
| 1487 |
+
},
|
| 1488 |
+
{
|
| 1489 |
+
"epoch": 4.702331961591221,
|
| 1490 |
+
"grad_norm": 0.07582115660185369,
|
| 1491 |
+
"learning_rate": 8.147676900367308e-07,
|
| 1492 |
+
"loss": 0.3412,
|
| 1493 |
+
"step": 212
|
| 1494 |
+
},
|
| 1495 |
+
{
|
| 1496 |
+
"epoch": 4.724279835390947,
|
| 1497 |
+
"grad_norm": 0.07655208386535386,
|
| 1498 |
+
"learning_rate": 6.945900711094534e-07,
|
| 1499 |
+
"loss": 0.3308,
|
| 1500 |
+
"step": 213
|
| 1501 |
+
},
|
| 1502 |
+
{
|
| 1503 |
+
"epoch": 4.746227709190672,
|
| 1504 |
+
"grad_norm": 0.07575601990026438,
|
| 1505 |
+
"learning_rate": 5.839193884527472e-07,
|
| 1506 |
+
"loss": 0.3326,
|
| 1507 |
+
"step": 214
|
| 1508 |
+
},
|
| 1509 |
+
{
|
| 1510 |
+
"epoch": 4.768175582990398,
|
| 1511 |
+
"grad_norm": 0.07445403870736927,
|
| 1512 |
+
"learning_rate": 4.827824103697332e-07,
|
| 1513 |
+
"loss": 0.3312,
|
| 1514 |
+
"step": 215
|
| 1515 |
+
},
|
| 1516 |
+
{
|
| 1517 |
+
"epoch": 4.790123456790123,
|
| 1518 |
+
"grad_norm": 0.07565335743085141,
|
| 1519 |
+
"learning_rate": 3.912035992132257e-07,
|
| 1520 |
+
"loss": 0.3396,
|
| 1521 |
+
"step": 216
|
| 1522 |
+
},
|
| 1523 |
+
{
|
| 1524 |
+
"epoch": 4.812071330589849,
|
| 1525 |
+
"grad_norm": 0.07504738171668904,
|
| 1526 |
+
"learning_rate": 3.0920510546894156e-07,
|
| 1527 |
+
"loss": 0.3279,
|
| 1528 |
+
"step": 217
|
| 1529 |
+
},
|
| 1530 |
+
{
|
| 1531 |
+
"epoch": 4.834019204389575,
|
| 1532 |
+
"grad_norm": 0.07495711123910379,
|
| 1533 |
+
"learning_rate": 2.3680676239789647e-07,
|
| 1534 |
+
"loss": 0.3373,
|
| 1535 |
+
"step": 218
|
| 1536 |
+
},
|
| 1537 |
+
{
|
| 1538 |
+
"epoch": 4.8559670781893,
|
| 1539 |
+
"grad_norm": 0.07581516860262182,
|
| 1540 |
+
"learning_rate": 1.740260812392558e-07,
|
| 1541 |
+
"loss": 0.3377,
|
| 1542 |
+
"step": 219
|
| 1543 |
+
},
|
| 1544 |
+
{
|
| 1545 |
+
"epoch": 4.877914951989026,
|
| 1546 |
+
"grad_norm": 0.07440451827362232,
|
| 1547 |
+
"learning_rate": 1.208782469748293e-07,
|
| 1548 |
+
"loss": 0.3375,
|
| 1549 |
+
"step": 220
|
| 1550 |
+
},
|
| 1551 |
+
{
|
| 1552 |
+
"epoch": 4.899862825788752,
|
| 1553 |
+
"grad_norm": 0.07405892919582897,
|
| 1554 |
+
"learning_rate": 7.737611465622686e-08,
|
| 1555 |
+
"loss": 0.3314,
|
| 1556 |
+
"step": 221
|
| 1557 |
+
},
|
| 1558 |
+
{
|
| 1559 |
+
"epoch": 4.921810699588478,
|
| 1560 |
+
"grad_norm": 0.07208525046101905,
|
| 1561 |
+
"learning_rate": 4.353020629556781e-08,
|
| 1562 |
+
"loss": 0.334,
|
| 1563 |
+
"step": 222
|
| 1564 |
+
},
|
| 1565 |
+
{
|
| 1566 |
+
"epoch": 4.9437585733882035,
|
| 1567 |
+
"grad_norm": 0.07250510923845269,
|
| 1568 |
+
"learning_rate": 1.934870832047686e-08,
|
| 1569 |
+
"loss": 0.3358,
|
| 1570 |
+
"step": 223
|
| 1571 |
+
},
|
| 1572 |
+
{
|
| 1573 |
+
"epoch": 4.965706447187928,
|
| 1574 |
+
"grad_norm": 0.07367572989174814,
|
| 1575 |
+
"learning_rate": 4.837469594018984e-09,
|
| 1576 |
+
"loss": 0.3436,
|
| 1577 |
+
"step": 224
|
| 1578 |
+
},
|
| 1579 |
+
{
|
| 1580 |
+
"epoch": 4.987654320987654,
|
| 1581 |
+
"grad_norm": 0.07408156451999265,
|
| 1582 |
+
"learning_rate": 0.0,
|
| 1583 |
+
"loss": 0.337,
|
| 1584 |
+
"step": 225
|
| 1585 |
+
},
|
| 1586 |
+
{
|
| 1587 |
+
"epoch": 4.987654320987654,
|
| 1588 |
+
"step": 225,
|
| 1589 |
+
"total_flos": 5.796779967165497e+18,
|
| 1590 |
+
"train_loss": 0.06433244016435412,
|
| 1591 |
+
"train_runtime": 4471.4422,
|
| 1592 |
+
"train_samples_per_second": 26.061,
|
| 1593 |
+
"train_steps_per_second": 0.05
|
| 1594 |
+
}
|
| 1595 |
+
],
|
| 1596 |
+
"logging_steps": 1,
|
| 1597 |
+
"max_steps": 225,
|
| 1598 |
+
"num_input_tokens_seen": 0,
|
| 1599 |
+
"num_train_epochs": 5,
|
| 1600 |
+
"save_steps": 500,
|
| 1601 |
+
"stateful_callbacks": {
|
| 1602 |
+
"TrainerControl": {
|
| 1603 |
+
"args": {
|
| 1604 |
+
"should_epoch_stop": false,
|
| 1605 |
+
"should_evaluate": false,
|
| 1606 |
+
"should_log": false,
|
| 1607 |
+
"should_save": true,
|
| 1608 |
+
"should_training_stop": true
|
| 1609 |
+
},
|
| 1610 |
+
"attributes": {}
|
| 1611 |
+
}
|
| 1612 |
+
},
|
| 1613 |
+
"total_flos": 5.796779967165497e+18,
|
| 1614 |
+
"train_batch_size": 1,
|
| 1615 |
+
"trial_name": null,
|
| 1616 |
+
"trial_params": null
|
| 1617 |
+
}
|
training_loss.png
ADDED
|