End of training
Browse files- README.md +2 -1
- all_results.json +8 -0
- train_results.json +8 -0
- trainer_state.json +1722 -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_100k
|
|
@@ -15,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 15 |
|
| 16 |
# no_pipeline_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_100k
|
|
|
|
| 16 |
|
| 17 |
# no_pipeline_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_100k dataset.
|
| 20 |
|
| 21 |
## Model description
|
| 22 |
|
all_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 4.961240310077519,
|
| 3 |
+
"total_flos": 6.175913847438705e+18,
|
| 4 |
+
"train_loss": 0.5455046107371648,
|
| 5 |
+
"train_runtime": 24740.3159,
|
| 6 |
+
"train_samples_per_second": 5.001,
|
| 7 |
+
"train_steps_per_second": 0.01
|
| 8 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 4.961240310077519,
|
| 3 |
+
"total_flos": 6.175913847438705e+18,
|
| 4 |
+
"train_loss": 0.5455046107371648,
|
| 5 |
+
"train_runtime": 24740.3159,
|
| 6 |
+
"train_samples_per_second": 5.001,
|
| 7 |
+
"train_steps_per_second": 0.01
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1722 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 4.961240310077519,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 240,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.020671834625323,
|
| 13 |
+
"grad_norm": 6.081340391876518,
|
| 14 |
+
"learning_rate": 3.3333333333333333e-06,
|
| 15 |
+
"loss": 1.0264,
|
| 16 |
+
"step": 1
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.041343669250646,
|
| 20 |
+
"grad_norm": 6.049219703890414,
|
| 21 |
+
"learning_rate": 6.666666666666667e-06,
|
| 22 |
+
"loss": 1.0229,
|
| 23 |
+
"step": 2
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.06201550387596899,
|
| 27 |
+
"grad_norm": 5.56528854905757,
|
| 28 |
+
"learning_rate": 1e-05,
|
| 29 |
+
"loss": 1.0092,
|
| 30 |
+
"step": 3
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.082687338501292,
|
| 34 |
+
"grad_norm": 2.3257360601962604,
|
| 35 |
+
"learning_rate": 1.3333333333333333e-05,
|
| 36 |
+
"loss": 0.9238,
|
| 37 |
+
"step": 4
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.10335917312661498,
|
| 41 |
+
"grad_norm": 4.521772675442624,
|
| 42 |
+
"learning_rate": 1.6666666666666667e-05,
|
| 43 |
+
"loss": 0.9198,
|
| 44 |
+
"step": 5
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.12403100775193798,
|
| 48 |
+
"grad_norm": 4.37427183723672,
|
| 49 |
+
"learning_rate": 2e-05,
|
| 50 |
+
"loss": 0.8775,
|
| 51 |
+
"step": 6
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.14470284237726097,
|
| 55 |
+
"grad_norm": 4.802783799029155,
|
| 56 |
+
"learning_rate": 2.3333333333333336e-05,
|
| 57 |
+
"loss": 0.8945,
|
| 58 |
+
"step": 7
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.165374677002584,
|
| 62 |
+
"grad_norm": 2.7669877595590453,
|
| 63 |
+
"learning_rate": 2.6666666666666667e-05,
|
| 64 |
+
"loss": 0.8358,
|
| 65 |
+
"step": 8
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.18604651162790697,
|
| 69 |
+
"grad_norm": 2.1697313693618114,
|
| 70 |
+
"learning_rate": 3.0000000000000004e-05,
|
| 71 |
+
"loss": 0.8016,
|
| 72 |
+
"step": 9
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.20671834625322996,
|
| 76 |
+
"grad_norm": 2.1242452729167947,
|
| 77 |
+
"learning_rate": 3.3333333333333335e-05,
|
| 78 |
+
"loss": 0.7892,
|
| 79 |
+
"step": 10
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.22739018087855298,
|
| 83 |
+
"grad_norm": 1.1871789145674065,
|
| 84 |
+
"learning_rate": 3.6666666666666666e-05,
|
| 85 |
+
"loss": 0.7626,
|
| 86 |
+
"step": 11
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.24806201550387597,
|
| 90 |
+
"grad_norm": 1.0307527389142968,
|
| 91 |
+
"learning_rate": 4e-05,
|
| 92 |
+
"loss": 0.7528,
|
| 93 |
+
"step": 12
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.268733850129199,
|
| 97 |
+
"grad_norm": 1.5380877865625096,
|
| 98 |
+
"learning_rate": 4.3333333333333334e-05,
|
| 99 |
+
"loss": 0.753,
|
| 100 |
+
"step": 13
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.28940568475452194,
|
| 104 |
+
"grad_norm": 1.3357996535784058,
|
| 105 |
+
"learning_rate": 4.666666666666667e-05,
|
| 106 |
+
"loss": 0.729,
|
| 107 |
+
"step": 14
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.31007751937984496,
|
| 111 |
+
"grad_norm": 2.108841055054808,
|
| 112 |
+
"learning_rate": 5e-05,
|
| 113 |
+
"loss": 0.7279,
|
| 114 |
+
"step": 15
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.330749354005168,
|
| 118 |
+
"grad_norm": 1.2363629258807027,
|
| 119 |
+
"learning_rate": 5.333333333333333e-05,
|
| 120 |
+
"loss": 0.7206,
|
| 121 |
+
"step": 16
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.35142118863049093,
|
| 125 |
+
"grad_norm": 1.6541735271205023,
|
| 126 |
+
"learning_rate": 5.666666666666668e-05,
|
| 127 |
+
"loss": 0.7111,
|
| 128 |
+
"step": 17
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.37209302325581395,
|
| 132 |
+
"grad_norm": 1.5865298239006316,
|
| 133 |
+
"learning_rate": 6.000000000000001e-05,
|
| 134 |
+
"loss": 0.7001,
|
| 135 |
+
"step": 18
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.39276485788113696,
|
| 139 |
+
"grad_norm": 1.4698126356680155,
|
| 140 |
+
"learning_rate": 6.333333333333333e-05,
|
| 141 |
+
"loss": 0.698,
|
| 142 |
+
"step": 19
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.4134366925064599,
|
| 146 |
+
"grad_norm": 1.7121395213894774,
|
| 147 |
+
"learning_rate": 6.666666666666667e-05,
|
| 148 |
+
"loss": 0.6963,
|
| 149 |
+
"step": 20
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.43410852713178294,
|
| 153 |
+
"grad_norm": 0.9852727154161406,
|
| 154 |
+
"learning_rate": 7.000000000000001e-05,
|
| 155 |
+
"loss": 0.696,
|
| 156 |
+
"step": 21
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.45478036175710596,
|
| 160 |
+
"grad_norm": 1.8968179606949396,
|
| 161 |
+
"learning_rate": 7.333333333333333e-05,
|
| 162 |
+
"loss": 0.7065,
|
| 163 |
+
"step": 22
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.4754521963824289,
|
| 167 |
+
"grad_norm": 146.52610473652015,
|
| 168 |
+
"learning_rate": 7.666666666666668e-05,
|
| 169 |
+
"loss": 0.8531,
|
| 170 |
+
"step": 23
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.49612403100775193,
|
| 174 |
+
"grad_norm": 10.069502663543679,
|
| 175 |
+
"learning_rate": 8e-05,
|
| 176 |
+
"loss": 0.774,
|
| 177 |
+
"step": 24
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.5167958656330749,
|
| 181 |
+
"grad_norm": 5.36227046437084,
|
| 182 |
+
"learning_rate": 7.99957692770843e-05,
|
| 183 |
+
"loss": 0.7956,
|
| 184 |
+
"step": 25
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.537467700258398,
|
| 188 |
+
"grad_norm": 1.4783577981990808,
|
| 189 |
+
"learning_rate": 7.998307800328803e-05,
|
| 190 |
+
"loss": 0.7182,
|
| 191 |
+
"step": 26
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.5581395348837209,
|
| 195 |
+
"grad_norm": 3.5766991353684947,
|
| 196 |
+
"learning_rate": 7.996192886327432e-05,
|
| 197 |
+
"loss": 0.7221,
|
| 198 |
+
"step": 27
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.5788113695090439,
|
| 202 |
+
"grad_norm": 4.003034876125958,
|
| 203 |
+
"learning_rate": 7.993232633085074e-05,
|
| 204 |
+
"loss": 0.7587,
|
| 205 |
+
"step": 28
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.599483204134367,
|
| 209 |
+
"grad_norm": 1.9492914484886623,
|
| 210 |
+
"learning_rate": 7.98942766680229e-05,
|
| 211 |
+
"loss": 0.7379,
|
| 212 |
+
"step": 29
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.6201550387596899,
|
| 216 |
+
"grad_norm": 2.6045364704629095,
|
| 217 |
+
"learning_rate": 7.984778792366983e-05,
|
| 218 |
+
"loss": 0.7131,
|
| 219 |
+
"step": 30
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.6408268733850129,
|
| 223 |
+
"grad_norm": 2.319907888607313,
|
| 224 |
+
"learning_rate": 7.979286993184134e-05,
|
| 225 |
+
"loss": 0.72,
|
| 226 |
+
"step": 31
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.661498708010336,
|
| 230 |
+
"grad_norm": 1.537213547697121,
|
| 231 |
+
"learning_rate": 7.972953430967773e-05,
|
| 232 |
+
"loss": 0.7132,
|
| 233 |
+
"step": 32
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.6821705426356589,
|
| 237 |
+
"grad_norm": 1.4610313174686893,
|
| 238 |
+
"learning_rate": 7.965779445495243e-05,
|
| 239 |
+
"loss": 0.6869,
|
| 240 |
+
"step": 33
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.7028423772609819,
|
| 244 |
+
"grad_norm": 1.4532833908707627,
|
| 245 |
+
"learning_rate": 7.957766554323778e-05,
|
| 246 |
+
"loss": 0.6989,
|
| 247 |
+
"step": 34
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.7235142118863049,
|
| 251 |
+
"grad_norm": 1.37784440284931,
|
| 252 |
+
"learning_rate": 7.948916452469497e-05,
|
| 253 |
+
"loss": 0.6758,
|
| 254 |
+
"step": 35
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.7441860465116279,
|
| 258 |
+
"grad_norm": 1.0775809244985184,
|
| 259 |
+
"learning_rate": 7.939231012048833e-05,
|
| 260 |
+
"loss": 0.6655,
|
| 261 |
+
"step": 36
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.7648578811369509,
|
| 265 |
+
"grad_norm": 0.9458264188781523,
|
| 266 |
+
"learning_rate": 7.928712281882523e-05,
|
| 267 |
+
"loss": 0.6772,
|
| 268 |
+
"step": 37
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.7855297157622739,
|
| 272 |
+
"grad_norm": 1.2667853755824652,
|
| 273 |
+
"learning_rate": 7.917362487062207e-05,
|
| 274 |
+
"loss": 0.6665,
|
| 275 |
+
"step": 38
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.8062015503875969,
|
| 279 |
+
"grad_norm": 1.101160214042413,
|
| 280 |
+
"learning_rate": 7.905184028479734e-05,
|
| 281 |
+
"loss": 0.6695,
|
| 282 |
+
"step": 39
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.8268733850129198,
|
| 286 |
+
"grad_norm": 0.7127541861054243,
|
| 287 |
+
"learning_rate": 7.892179482319297e-05,
|
| 288 |
+
"loss": 0.6581,
|
| 289 |
+
"step": 40
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.8475452196382429,
|
| 293 |
+
"grad_norm": 0.9635032822917181,
|
| 294 |
+
"learning_rate": 7.878351599512465e-05,
|
| 295 |
+
"loss": 0.6551,
|
| 296 |
+
"step": 41
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.8682170542635659,
|
| 300 |
+
"grad_norm": 1.1027366432137207,
|
| 301 |
+
"learning_rate": 7.863703305156273e-05,
|
| 302 |
+
"loss": 0.6461,
|
| 303 |
+
"step": 42
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.8888888888888888,
|
| 307 |
+
"grad_norm": 1.0436642791344313,
|
| 308 |
+
"learning_rate": 7.848237697894453e-05,
|
| 309 |
+
"loss": 0.6518,
|
| 310 |
+
"step": 43
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.9095607235142119,
|
| 314 |
+
"grad_norm": 1.056091207907409,
|
| 315 |
+
"learning_rate": 7.831958049261956e-05,
|
| 316 |
+
"loss": 0.654,
|
| 317 |
+
"step": 44
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.9302325581395349,
|
| 321 |
+
"grad_norm": 1.2384607035063486,
|
| 322 |
+
"learning_rate": 7.814867802992907e-05,
|
| 323 |
+
"loss": 0.6576,
|
| 324 |
+
"step": 45
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.9509043927648578,
|
| 328 |
+
"grad_norm": 0.7018322803784657,
|
| 329 |
+
"learning_rate": 7.796970574292136e-05,
|
| 330 |
+
"loss": 0.6481,
|
| 331 |
+
"step": 46
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.9715762273901809,
|
| 335 |
+
"grad_norm": 0.9251162844464464,
|
| 336 |
+
"learning_rate": 7.77827014907042e-05,
|
| 337 |
+
"loss": 0.6389,
|
| 338 |
+
"step": 47
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.9922480620155039,
|
| 342 |
+
"grad_norm": 1.2613456027945058,
|
| 343 |
+
"learning_rate": 7.758770483143634e-05,
|
| 344 |
+
"loss": 0.6467,
|
| 345 |
+
"step": 48
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 1.0129198966408268,
|
| 349 |
+
"grad_norm": 0.5671677774550341,
|
| 350 |
+
"learning_rate": 7.738475701395955e-05,
|
| 351 |
+
"loss": 0.6234,
|
| 352 |
+
"step": 49
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 1.0335917312661498,
|
| 356 |
+
"grad_norm": 0.9160751747799357,
|
| 357 |
+
"learning_rate": 7.71739009690729e-05,
|
| 358 |
+
"loss": 0.6109,
|
| 359 |
+
"step": 50
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 1.054263565891473,
|
| 363 |
+
"grad_norm": 1.2875726947500998,
|
| 364 |
+
"learning_rate": 7.695518130045147e-05,
|
| 365 |
+
"loss": 0.6093,
|
| 366 |
+
"step": 51
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 1.074935400516796,
|
| 370 |
+
"grad_norm": 0.554040187879577,
|
| 371 |
+
"learning_rate": 7.672864427521097e-05,
|
| 372 |
+
"loss": 0.5999,
|
| 373 |
+
"step": 52
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 1.0956072351421189,
|
| 377 |
+
"grad_norm": 1.296094506886022,
|
| 378 |
+
"learning_rate": 7.649433781412058e-05,
|
| 379 |
+
"loss": 0.6134,
|
| 380 |
+
"step": 53
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 1.1162790697674418,
|
| 384 |
+
"grad_norm": 0.8155821827606181,
|
| 385 |
+
"learning_rate": 7.625231148146601e-05,
|
| 386 |
+
"loss": 0.5979,
|
| 387 |
+
"step": 54
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 1.1369509043927648,
|
| 391 |
+
"grad_norm": 0.8752566192167226,
|
| 392 |
+
"learning_rate": 7.600261647456485e-05,
|
| 393 |
+
"loss": 0.6091,
|
| 394 |
+
"step": 55
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 1.1576227390180878,
|
| 398 |
+
"grad_norm": 0.7919804255607983,
|
| 399 |
+
"learning_rate": 7.57453056129365e-05,
|
| 400 |
+
"loss": 0.5972,
|
| 401 |
+
"step": 56
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 1.178294573643411,
|
| 405 |
+
"grad_norm": 0.7770985679546745,
|
| 406 |
+
"learning_rate": 7.548043332712887e-05,
|
| 407 |
+
"loss": 0.5933,
|
| 408 |
+
"step": 57
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 1.198966408268734,
|
| 412 |
+
"grad_norm": 0.9248530699486545,
|
| 413 |
+
"learning_rate": 7.520805564720444e-05,
|
| 414 |
+
"loss": 0.5982,
|
| 415 |
+
"step": 58
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 1.2196382428940569,
|
| 419 |
+
"grad_norm": 0.6869794809786043,
|
| 420 |
+
"learning_rate": 7.492823019088785e-05,
|
| 421 |
+
"loss": 0.589,
|
| 422 |
+
"step": 59
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 1.2403100775193798,
|
| 426 |
+
"grad_norm": 0.9105425085377974,
|
| 427 |
+
"learning_rate": 7.464101615137756e-05,
|
| 428 |
+
"loss": 0.5942,
|
| 429 |
+
"step": 60
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 1.2609819121447028,
|
| 433 |
+
"grad_norm": 0.6121169268536212,
|
| 434 |
+
"learning_rate": 7.434647428482453e-05,
|
| 435 |
+
"loss": 0.5855,
|
| 436 |
+
"step": 61
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 1.2816537467700257,
|
| 440 |
+
"grad_norm": 0.5083941774190959,
|
| 441 |
+
"learning_rate": 7.404466689747999e-05,
|
| 442 |
+
"loss": 0.585,
|
| 443 |
+
"step": 62
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 1.302325581395349,
|
| 447 |
+
"grad_norm": 0.7768425888786246,
|
| 448 |
+
"learning_rate": 7.373565783251544e-05,
|
| 449 |
+
"loss": 0.5885,
|
| 450 |
+
"step": 63
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 1.322997416020672,
|
| 454 |
+
"grad_norm": 0.638420735763997,
|
| 455 |
+
"learning_rate": 7.341951245651747e-05,
|
| 456 |
+
"loss": 0.5831,
|
| 457 |
+
"step": 64
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 1.3436692506459949,
|
| 461 |
+
"grad_norm": 0.6541167240196839,
|
| 462 |
+
"learning_rate": 7.309629764566042e-05,
|
| 463 |
+
"loss": 0.5896,
|
| 464 |
+
"step": 65
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 1.3643410852713178,
|
| 468 |
+
"grad_norm": 0.41660519240697114,
|
| 469 |
+
"learning_rate": 7.276608177155968e-05,
|
| 470 |
+
"loss": 0.5821,
|
| 471 |
+
"step": 66
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 1.3850129198966408,
|
| 475 |
+
"grad_norm": 0.5838901498646191,
|
| 476 |
+
"learning_rate": 7.242893468680849e-05,
|
| 477 |
+
"loss": 0.5806,
|
| 478 |
+
"step": 67
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 1.405684754521964,
|
| 482 |
+
"grad_norm": 0.6108345483440157,
|
| 483 |
+
"learning_rate": 7.208492771020176e-05,
|
| 484 |
+
"loss": 0.5722,
|
| 485 |
+
"step": 68
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 1.4263565891472867,
|
| 489 |
+
"grad_norm": 0.5874821613065601,
|
| 490 |
+
"learning_rate": 7.173413361164941e-05,
|
| 491 |
+
"loss": 0.5752,
|
| 492 |
+
"step": 69
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 1.4470284237726099,
|
| 496 |
+
"grad_norm": 0.4082209348015132,
|
| 497 |
+
"learning_rate": 7.137662659678303e-05,
|
| 498 |
+
"loss": 0.5807,
|
| 499 |
+
"step": 70
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 1.4677002583979328,
|
| 503 |
+
"grad_norm": 0.5381365402558759,
|
| 504 |
+
"learning_rate": 7.101248229125864e-05,
|
| 505 |
+
"loss": 0.5746,
|
| 506 |
+
"step": 71
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 1.4883720930232558,
|
| 510 |
+
"grad_norm": 0.45302323926539034,
|
| 511 |
+
"learning_rate": 7.064177772475912e-05,
|
| 512 |
+
"loss": 0.5847,
|
| 513 |
+
"step": 72
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 1.509043927648579,
|
| 517 |
+
"grad_norm": 0.48557743985677326,
|
| 518 |
+
"learning_rate": 7.026459131469972e-05,
|
| 519 |
+
"loss": 0.5813,
|
| 520 |
+
"step": 73
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 1.5297157622739017,
|
| 524 |
+
"grad_norm": 0.5262378389194146,
|
| 525 |
+
"learning_rate": 6.988100284963985e-05,
|
| 526 |
+
"loss": 0.5742,
|
| 527 |
+
"step": 74
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 1.550387596899225,
|
| 531 |
+
"grad_norm": 0.39828389103953854,
|
| 532 |
+
"learning_rate": 6.949109347240496e-05,
|
| 533 |
+
"loss": 0.5774,
|
| 534 |
+
"step": 75
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 1.5710594315245479,
|
| 538 |
+
"grad_norm": 0.42597294685589654,
|
| 539 |
+
"learning_rate": 6.909494566292195e-05,
|
| 540 |
+
"loss": 0.5788,
|
| 541 |
+
"step": 76
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 1.5917312661498708,
|
| 545 |
+
"grad_norm": 0.4935762794884097,
|
| 546 |
+
"learning_rate": 6.869264322077158e-05,
|
| 547 |
+
"loss": 0.5733,
|
| 548 |
+
"step": 77
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 1.6124031007751938,
|
| 552 |
+
"grad_norm": 0.3991446509521148,
|
| 553 |
+
"learning_rate": 6.828427124746191e-05,
|
| 554 |
+
"loss": 0.5689,
|
| 555 |
+
"step": 78
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 1.6330749354005167,
|
| 559 |
+
"grad_norm": 0.3892417895271408,
|
| 560 |
+
"learning_rate": 6.786991612842621e-05,
|
| 561 |
+
"loss": 0.577,
|
| 562 |
+
"step": 79
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 1.65374677002584,
|
| 566 |
+
"grad_norm": 0.5339994357651601,
|
| 567 |
+
"learning_rate": 6.744966551474936e-05,
|
| 568 |
+
"loss": 0.5762,
|
| 569 |
+
"step": 80
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 1.6744186046511627,
|
| 573 |
+
"grad_norm": 0.5475519212078531,
|
| 574 |
+
"learning_rate": 6.702360830462642e-05,
|
| 575 |
+
"loss": 0.5748,
|
| 576 |
+
"step": 81
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 1.6950904392764858,
|
| 580 |
+
"grad_norm": 0.5395921798190672,
|
| 581 |
+
"learning_rate": 6.659183462455751e-05,
|
| 582 |
+
"loss": 0.5749,
|
| 583 |
+
"step": 82
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 1.7157622739018088,
|
| 587 |
+
"grad_norm": 0.4491381596059805,
|
| 588 |
+
"learning_rate": 6.615443581028279e-05,
|
| 589 |
+
"loss": 0.577,
|
| 590 |
+
"step": 83
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 1.7364341085271318,
|
| 594 |
+
"grad_norm": 0.36466491876407947,
|
| 595 |
+
"learning_rate": 6.571150438746157e-05,
|
| 596 |
+
"loss": 0.5641,
|
| 597 |
+
"step": 84
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 1.757105943152455,
|
| 601 |
+
"grad_norm": 0.3701537836538638,
|
| 602 |
+
"learning_rate": 6.526313405209991e-05,
|
| 603 |
+
"loss": 0.5639,
|
| 604 |
+
"step": 85
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 1.7777777777777777,
|
| 608 |
+
"grad_norm": 0.4434538791567632,
|
| 609 |
+
"learning_rate": 6.480941965073041e-05,
|
| 610 |
+
"loss": 0.5717,
|
| 611 |
+
"step": 86
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 1.7984496124031009,
|
| 615 |
+
"grad_norm": 0.5051497502235356,
|
| 616 |
+
"learning_rate": 6.435045716034883e-05,
|
| 617 |
+
"loss": 0.5713,
|
| 618 |
+
"step": 87
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 1.8191214470284238,
|
| 622 |
+
"grad_norm": 0.4304045333495718,
|
| 623 |
+
"learning_rate": 6.388634366811146e-05,
|
| 624 |
+
"loss": 0.5664,
|
| 625 |
+
"step": 88
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 1.8397932816537468,
|
| 629 |
+
"grad_norm": 0.34693482042494633,
|
| 630 |
+
"learning_rate": 6.341717735079763e-05,
|
| 631 |
+
"loss": 0.572,
|
| 632 |
+
"step": 89
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 1.8604651162790697,
|
| 636 |
+
"grad_norm": 0.4252774495384975,
|
| 637 |
+
"learning_rate": 6.294305745404185e-05,
|
| 638 |
+
"loss": 0.5707,
|
| 639 |
+
"step": 90
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 1.8811369509043927,
|
| 643 |
+
"grad_norm": 0.5232979390151309,
|
| 644 |
+
"learning_rate": 6.246408427133972e-05,
|
| 645 |
+
"loss": 0.5758,
|
| 646 |
+
"step": 91
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 1.9018087855297159,
|
| 650 |
+
"grad_norm": 0.4377269590744326,
|
| 651 |
+
"learning_rate": 6.198035912283225e-05,
|
| 652 |
+
"loss": 0.5788,
|
| 653 |
+
"step": 92
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 1.9224806201550386,
|
| 657 |
+
"grad_norm": 0.24592685566286257,
|
| 658 |
+
"learning_rate": 6.149198433387297e-05,
|
| 659 |
+
"loss": 0.5667,
|
| 660 |
+
"step": 93
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 1.9431524547803618,
|
| 664 |
+
"grad_norm": 0.3027013998334105,
|
| 665 |
+
"learning_rate": 6.099906321338241e-05,
|
| 666 |
+
"loss": 0.5698,
|
| 667 |
+
"step": 94
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 1.9638242894056848,
|
| 671 |
+
"grad_norm": 0.3870500290875651,
|
| 672 |
+
"learning_rate": 6.0501700031994613e-05,
|
| 673 |
+
"loss": 0.5674,
|
| 674 |
+
"step": 95
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 1.9844961240310077,
|
| 678 |
+
"grad_norm": 0.3619909951438359,
|
| 679 |
+
"learning_rate": 6.000000000000001e-05,
|
| 680 |
+
"loss": 0.5659,
|
| 681 |
+
"step": 96
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 2.005167958656331,
|
| 685 |
+
"grad_norm": 0.2928009412756042,
|
| 686 |
+
"learning_rate": 5.94940692450897e-05,
|
| 687 |
+
"loss": 0.5571,
|
| 688 |
+
"step": 97
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 2.0258397932816536,
|
| 692 |
+
"grad_norm": 0.3182447769585121,
|
| 693 |
+
"learning_rate": 5.8984014789905625e-05,
|
| 694 |
+
"loss": 0.5155,
|
| 695 |
+
"step": 98
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 2.046511627906977,
|
| 699 |
+
"grad_norm": 0.31756241951749276,
|
| 700 |
+
"learning_rate": 5.846994452940137e-05,
|
| 701 |
+
"loss": 0.5168,
|
| 702 |
+
"step": 99
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 2.0671834625322996,
|
| 706 |
+
"grad_norm": 0.4319233302339089,
|
| 707 |
+
"learning_rate": 5.79519672080185e-05,
|
| 708 |
+
"loss": 0.5182,
|
| 709 |
+
"step": 100
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 2.0878552971576227,
|
| 713 |
+
"grad_norm": 0.6857605688800891,
|
| 714 |
+
"learning_rate": 5.743019239668318e-05,
|
| 715 |
+
"loss": 0.5262,
|
| 716 |
+
"step": 101
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 2.108527131782946,
|
| 720 |
+
"grad_norm": 0.979897045058122,
|
| 721 |
+
"learning_rate": 5.6904730469627985e-05,
|
| 722 |
+
"loss": 0.5266,
|
| 723 |
+
"step": 102
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 2.1291989664082687,
|
| 727 |
+
"grad_norm": 1.0974403188788746,
|
| 728 |
+
"learning_rate": 5.6375692581043705e-05,
|
| 729 |
+
"loss": 0.5195,
|
| 730 |
+
"step": 103
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 2.149870801033592,
|
| 734 |
+
"grad_norm": 0.6853370217341208,
|
| 735 |
+
"learning_rate": 5.584319064156628e-05,
|
| 736 |
+
"loss": 0.5133,
|
| 737 |
+
"step": 104
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 2.1705426356589146,
|
| 741 |
+
"grad_norm": 0.7088375080993865,
|
| 742 |
+
"learning_rate": 5.5307337294603595e-05,
|
| 743 |
+
"loss": 0.5234,
|
| 744 |
+
"step": 105
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 2.1912144702842378,
|
| 748 |
+
"grad_norm": 0.7876062584364661,
|
| 749 |
+
"learning_rate": 5.476824589250738e-05,
|
| 750 |
+
"loss": 0.5136,
|
| 751 |
+
"step": 106
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 2.2118863049095605,
|
| 755 |
+
"grad_norm": 0.7094284990143945,
|
| 756 |
+
"learning_rate": 5.4226030472595075e-05,
|
| 757 |
+
"loss": 0.5164,
|
| 758 |
+
"step": 107
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 2.2325581395348837,
|
| 762 |
+
"grad_norm": 0.47633469205660306,
|
| 763 |
+
"learning_rate": 5.368080573302676e-05,
|
| 764 |
+
"loss": 0.5138,
|
| 765 |
+
"step": 108
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 2.253229974160207,
|
| 769 |
+
"grad_norm": 0.6259985560998415,
|
| 770 |
+
"learning_rate": 5.3132687008542454e-05,
|
| 771 |
+
"loss": 0.5141,
|
| 772 |
+
"step": 109
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 2.2739018087855296,
|
| 776 |
+
"grad_norm": 0.5909251717900875,
|
| 777 |
+
"learning_rate": 5.258179024606455e-05,
|
| 778 |
+
"loss": 0.5097,
|
| 779 |
+
"step": 110
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 2.294573643410853,
|
| 783 |
+
"grad_norm": 0.48237832073015746,
|
| 784 |
+
"learning_rate": 5.202823198017092e-05,
|
| 785 |
+
"loss": 0.508,
|
| 786 |
+
"step": 111
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 2.3152454780361755,
|
| 790 |
+
"grad_norm": 0.42961478701613043,
|
| 791 |
+
"learning_rate": 5.1472129308443616e-05,
|
| 792 |
+
"loss": 0.5143,
|
| 793 |
+
"step": 112
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 2.3359173126614987,
|
| 797 |
+
"grad_norm": 0.4701977876238467,
|
| 798 |
+
"learning_rate": 5.091359986669845e-05,
|
| 799 |
+
"loss": 0.5144,
|
| 800 |
+
"step": 113
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 2.356589147286822,
|
| 804 |
+
"grad_norm": 0.4042528212628491,
|
| 805 |
+
"learning_rate": 5.0352761804100835e-05,
|
| 806 |
+
"loss": 0.5124,
|
| 807 |
+
"step": 114
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 2.3772609819121446,
|
| 811 |
+
"grad_norm": 0.4772687144806787,
|
| 812 |
+
"learning_rate": 4.9789733758172956e-05,
|
| 813 |
+
"loss": 0.5078,
|
| 814 |
+
"step": 115
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 2.397932816537468,
|
| 818 |
+
"grad_norm": 0.32596594872115625,
|
| 819 |
+
"learning_rate": 4.922463482969761e-05,
|
| 820 |
+
"loss": 0.5121,
|
| 821 |
+
"step": 116
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 2.4186046511627906,
|
| 825 |
+
"grad_norm": 0.4519574467429711,
|
| 826 |
+
"learning_rate": 4.8657584557524116e-05,
|
| 827 |
+
"loss": 0.5096,
|
| 828 |
+
"step": 117
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 2.4392764857881137,
|
| 832 |
+
"grad_norm": 0.34741131326324054,
|
| 833 |
+
"learning_rate": 4.808870289328153e-05,
|
| 834 |
+
"loss": 0.5073,
|
| 835 |
+
"step": 118
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 2.459948320413437,
|
| 839 |
+
"grad_norm": 0.33213800514453856,
|
| 840 |
+
"learning_rate": 4.751811017600448e-05,
|
| 841 |
+
"loss": 0.5127,
|
| 842 |
+
"step": 119
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 2.4806201550387597,
|
| 846 |
+
"grad_norm": 0.3288342560594038,
|
| 847 |
+
"learning_rate": 4.694592710667723e-05,
|
| 848 |
+
"loss": 0.5005,
|
| 849 |
+
"step": 120
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 2.501291989664083,
|
| 853 |
+
"grad_norm": 0.27041702222249453,
|
| 854 |
+
"learning_rate": 4.637227472270091e-05,
|
| 855 |
+
"loss": 0.5049,
|
| 856 |
+
"step": 121
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 2.5219638242894056,
|
| 860 |
+
"grad_norm": 0.32270193058473734,
|
| 861 |
+
"learning_rate": 4.579727437228987e-05,
|
| 862 |
+
"loss": 0.5096,
|
| 863 |
+
"step": 122
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 2.5426356589147288,
|
| 867 |
+
"grad_norm": 0.23335079075041096,
|
| 868 |
+
"learning_rate": 4.522104768880208e-05,
|
| 869 |
+
"loss": 0.5081,
|
| 870 |
+
"step": 123
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 2.5633074935400515,
|
| 874 |
+
"grad_norm": 0.30434430711273436,
|
| 875 |
+
"learning_rate": 4.464371656500921e-05,
|
| 876 |
+
"loss": 0.4975,
|
| 877 |
+
"step": 124
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 2.5839793281653747,
|
| 881 |
+
"grad_norm": 0.2054339593632784,
|
| 882 |
+
"learning_rate": 4.406540312731208e-05,
|
| 883 |
+
"loss": 0.5082,
|
| 884 |
+
"step": 125
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 2.604651162790698,
|
| 888 |
+
"grad_norm": 0.30415797673920053,
|
| 889 |
+
"learning_rate": 4.348622970990634e-05,
|
| 890 |
+
"loss": 0.5082,
|
| 891 |
+
"step": 126
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 2.6253229974160206,
|
| 895 |
+
"grad_norm": 0.20081102013294175,
|
| 896 |
+
"learning_rate": 4.290631882890443e-05,
|
| 897 |
+
"loss": 0.5045,
|
| 898 |
+
"step": 127
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 2.645994832041344,
|
| 902 |
+
"grad_norm": 0.2645082788535957,
|
| 903 |
+
"learning_rate": 4.2325793156419035e-05,
|
| 904 |
+
"loss": 0.5056,
|
| 905 |
+
"step": 128
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 2.6666666666666665,
|
| 909 |
+
"grad_norm": 0.25332732415756865,
|
| 910 |
+
"learning_rate": 4.174477549461345e-05,
|
| 911 |
+
"loss": 0.5066,
|
| 912 |
+
"step": 129
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 2.6873385012919897,
|
| 916 |
+
"grad_norm": 0.2555288373182799,
|
| 917 |
+
"learning_rate": 4.116338874972446e-05,
|
| 918 |
+
"loss": 0.5069,
|
| 919 |
+
"step": 130
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 2.7080103359173124,
|
| 923 |
+
"grad_norm": 0.22166955351874776,
|
| 924 |
+
"learning_rate": 4.058175590606332e-05,
|
| 925 |
+
"loss": 0.505,
|
| 926 |
+
"step": 131
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 2.7286821705426356,
|
| 930 |
+
"grad_norm": 0.23388382116450784,
|
| 931 |
+
"learning_rate": 4e-05,
|
| 932 |
+
"loss": 0.508,
|
| 933 |
+
"step": 132
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 2.749354005167959,
|
| 937 |
+
"grad_norm": 0.18647341799659248,
|
| 938 |
+
"learning_rate": 3.9418244093936694e-05,
|
| 939 |
+
"loss": 0.503,
|
| 940 |
+
"step": 133
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 2.7700258397932815,
|
| 944 |
+
"grad_norm": 0.17565964391428604,
|
| 945 |
+
"learning_rate": 3.8836611250275546e-05,
|
| 946 |
+
"loss": 0.5077,
|
| 947 |
+
"step": 134
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 2.7906976744186047,
|
| 951 |
+
"grad_norm": 0.182642588422039,
|
| 952 |
+
"learning_rate": 3.825522450538657e-05,
|
| 953 |
+
"loss": 0.5112,
|
| 954 |
+
"step": 135
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 2.811369509043928,
|
| 958 |
+
"grad_norm": 0.1754033168983459,
|
| 959 |
+
"learning_rate": 3.767420684358097e-05,
|
| 960 |
+
"loss": 0.5052,
|
| 961 |
+
"step": 136
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 2.8320413436692506,
|
| 965 |
+
"grad_norm": 0.16719637478676044,
|
| 966 |
+
"learning_rate": 3.709368117109558e-05,
|
| 967 |
+
"loss": 0.5076,
|
| 968 |
+
"step": 137
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 2.8527131782945734,
|
| 972 |
+
"grad_norm": 0.1634444926991004,
|
| 973 |
+
"learning_rate": 3.6513770290093674e-05,
|
| 974 |
+
"loss": 0.5067,
|
| 975 |
+
"step": 138
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 2.8733850129198966,
|
| 979 |
+
"grad_norm": 0.16416897404434663,
|
| 980 |
+
"learning_rate": 3.5934596872687924e-05,
|
| 981 |
+
"loss": 0.5052,
|
| 982 |
+
"step": 139
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 2.8940568475452197,
|
| 986 |
+
"grad_norm": 0.17669642337405403,
|
| 987 |
+
"learning_rate": 3.535628343499079e-05,
|
| 988 |
+
"loss": 0.5026,
|
| 989 |
+
"step": 140
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 2.9147286821705425,
|
| 993 |
+
"grad_norm": 0.13268304958503394,
|
| 994 |
+
"learning_rate": 3.477895231119795e-05,
|
| 995 |
+
"loss": 0.5131,
|
| 996 |
+
"step": 141
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 2.9354005167958657,
|
| 1000 |
+
"grad_norm": 0.1537884430700156,
|
| 1001 |
+
"learning_rate": 3.4202725627710136e-05,
|
| 1002 |
+
"loss": 0.5079,
|
| 1003 |
+
"step": 142
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 2.956072351421189,
|
| 1007 |
+
"grad_norm": 0.13654370111076927,
|
| 1008 |
+
"learning_rate": 3.3627725277299103e-05,
|
| 1009 |
+
"loss": 0.4975,
|
| 1010 |
+
"step": 143
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 2.9767441860465116,
|
| 1014 |
+
"grad_norm": 0.13537885556344376,
|
| 1015 |
+
"learning_rate": 3.305407289332279e-05,
|
| 1016 |
+
"loss": 0.5012,
|
| 1017 |
+
"step": 144
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 2.9974160206718348,
|
| 1021 |
+
"grad_norm": 0.13844052388817138,
|
| 1022 |
+
"learning_rate": 3.248188982399553e-05,
|
| 1023 |
+
"loss": 0.5084,
|
| 1024 |
+
"step": 145
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 3.0180878552971575,
|
| 1028 |
+
"grad_norm": 0.2386176766219906,
|
| 1029 |
+
"learning_rate": 3.191129710671849e-05,
|
| 1030 |
+
"loss": 0.4725,
|
| 1031 |
+
"step": 146
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 3.0387596899224807,
|
| 1035 |
+
"grad_norm": 0.15836289006865195,
|
| 1036 |
+
"learning_rate": 3.134241544247589e-05,
|
| 1037 |
+
"loss": 0.4612,
|
| 1038 |
+
"step": 147
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 3.0594315245478034,
|
| 1042 |
+
"grad_norm": 0.20569358929816706,
|
| 1043 |
+
"learning_rate": 3.07753651703024e-05,
|
| 1044 |
+
"loss": 0.4639,
|
| 1045 |
+
"step": 148
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 3.0801033591731266,
|
| 1049 |
+
"grad_norm": 0.23860379567193107,
|
| 1050 |
+
"learning_rate": 3.0210266241827047e-05,
|
| 1051 |
+
"loss": 0.4558,
|
| 1052 |
+
"step": 149
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 3.10077519379845,
|
| 1056 |
+
"grad_norm": 0.1797563193581857,
|
| 1057 |
+
"learning_rate": 2.9647238195899168e-05,
|
| 1058 |
+
"loss": 0.4583,
|
| 1059 |
+
"step": 150
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 3.1214470284237725,
|
| 1063 |
+
"grad_norm": 0.2309034832579931,
|
| 1064 |
+
"learning_rate": 2.9086400133301573e-05,
|
| 1065 |
+
"loss": 0.4507,
|
| 1066 |
+
"step": 151
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 3.1421188630490957,
|
| 1070 |
+
"grad_norm": 0.19408398760783607,
|
| 1071 |
+
"learning_rate": 2.8527870691556404e-05,
|
| 1072 |
+
"loss": 0.4576,
|
| 1073 |
+
"step": 152
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 3.1627906976744184,
|
| 1077 |
+
"grad_norm": 0.19370010890946,
|
| 1078 |
+
"learning_rate": 2.7971768019829083e-05,
|
| 1079 |
+
"loss": 0.4564,
|
| 1080 |
+
"step": 153
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 3.1834625322997416,
|
| 1084 |
+
"grad_norm": 0.20393652257874442,
|
| 1085 |
+
"learning_rate": 2.7418209753935464e-05,
|
| 1086 |
+
"loss": 0.4543,
|
| 1087 |
+
"step": 154
|
| 1088 |
+
},
|
| 1089 |
+
{
|
| 1090 |
+
"epoch": 3.2041343669250644,
|
| 1091 |
+
"grad_norm": 0.2126260560212861,
|
| 1092 |
+
"learning_rate": 2.6867312991457563e-05,
|
| 1093 |
+
"loss": 0.4555,
|
| 1094 |
+
"step": 155
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 3.2248062015503876,
|
| 1098 |
+
"grad_norm": 0.18702883489538974,
|
| 1099 |
+
"learning_rate": 2.6319194266973256e-05,
|
| 1100 |
+
"loss": 0.4569,
|
| 1101 |
+
"step": 156
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 3.2454780361757107,
|
| 1105 |
+
"grad_norm": 0.19084262286386808,
|
| 1106 |
+
"learning_rate": 2.577396952740495e-05,
|
| 1107 |
+
"loss": 0.4578,
|
| 1108 |
+
"step": 157
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"epoch": 3.2661498708010335,
|
| 1112 |
+
"grad_norm": 0.19502294645464513,
|
| 1113 |
+
"learning_rate": 2.523175410749263e-05,
|
| 1114 |
+
"loss": 0.4545,
|
| 1115 |
+
"step": 158
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 3.2868217054263567,
|
| 1119 |
+
"grad_norm": 0.15690954605019553,
|
| 1120 |
+
"learning_rate": 2.4692662705396412e-05,
|
| 1121 |
+
"loss": 0.4567,
|
| 1122 |
+
"step": 159
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 3.3074935400516794,
|
| 1126 |
+
"grad_norm": 0.18411001542533528,
|
| 1127 |
+
"learning_rate": 2.4156809358433728e-05,
|
| 1128 |
+
"loss": 0.4573,
|
| 1129 |
+
"step": 160
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 3.3281653746770026,
|
| 1133 |
+
"grad_norm": 0.1543947271873081,
|
| 1134 |
+
"learning_rate": 2.3624307418956298e-05,
|
| 1135 |
+
"loss": 0.4609,
|
| 1136 |
+
"step": 161
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 3.3488372093023258,
|
| 1140 |
+
"grad_norm": 0.15457127226752593,
|
| 1141 |
+
"learning_rate": 2.3095269530372032e-05,
|
| 1142 |
+
"loss": 0.4602,
|
| 1143 |
+
"step": 162
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 3.3695090439276485,
|
| 1147 |
+
"grad_norm": 0.1589621747112874,
|
| 1148 |
+
"learning_rate": 2.2569807603316836e-05,
|
| 1149 |
+
"loss": 0.4551,
|
| 1150 |
+
"step": 163
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 3.3901808785529717,
|
| 1154 |
+
"grad_norm": 0.14774423791181068,
|
| 1155 |
+
"learning_rate": 2.2048032791981515e-05,
|
| 1156 |
+
"loss": 0.4567,
|
| 1157 |
+
"step": 164
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 3.4108527131782944,
|
| 1161 |
+
"grad_norm": 0.13770622206264438,
|
| 1162 |
+
"learning_rate": 2.1530055470598654e-05,
|
| 1163 |
+
"loss": 0.4572,
|
| 1164 |
+
"step": 165
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 3.4315245478036176,
|
| 1168 |
+
"grad_norm": 0.1483527066505866,
|
| 1169 |
+
"learning_rate": 2.1015985210094385e-05,
|
| 1170 |
+
"loss": 0.4592,
|
| 1171 |
+
"step": 166
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 3.452196382428941,
|
| 1175 |
+
"grad_norm": 0.14556831349153918,
|
| 1176 |
+
"learning_rate": 2.050593075491031e-05,
|
| 1177 |
+
"loss": 0.455,
|
| 1178 |
+
"step": 167
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 3.4728682170542635,
|
| 1182 |
+
"grad_norm": 0.1345956480118845,
|
| 1183 |
+
"learning_rate": 2.0000000000000012e-05,
|
| 1184 |
+
"loss": 0.457,
|
| 1185 |
+
"step": 168
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 3.4935400516795867,
|
| 1189 |
+
"grad_norm": 0.13731093687667645,
|
| 1190 |
+
"learning_rate": 1.9498299968005393e-05,
|
| 1191 |
+
"loss": 0.451,
|
| 1192 |
+
"step": 169
|
| 1193 |
+
},
|
| 1194 |
+
{
|
| 1195 |
+
"epoch": 3.5142118863049094,
|
| 1196 |
+
"grad_norm": 0.1323560903236194,
|
| 1197 |
+
"learning_rate": 1.90009367866176e-05,
|
| 1198 |
+
"loss": 0.4583,
|
| 1199 |
+
"step": 170
|
| 1200 |
+
},
|
| 1201 |
+
{
|
| 1202 |
+
"epoch": 3.5348837209302326,
|
| 1203 |
+
"grad_norm": 0.1316854061469003,
|
| 1204 |
+
"learning_rate": 1.8508015666127043e-05,
|
| 1205 |
+
"loss": 0.4578,
|
| 1206 |
+
"step": 171
|
| 1207 |
+
},
|
| 1208 |
+
{
|
| 1209 |
+
"epoch": 3.5555555555555554,
|
| 1210 |
+
"grad_norm": 0.12014461989377777,
|
| 1211 |
+
"learning_rate": 1.8019640877167763e-05,
|
| 1212 |
+
"loss": 0.461,
|
| 1213 |
+
"step": 172
|
| 1214 |
+
},
|
| 1215 |
+
{
|
| 1216 |
+
"epoch": 3.5762273901808785,
|
| 1217 |
+
"grad_norm": 0.12754443867422668,
|
| 1218 |
+
"learning_rate": 1.753591572866029e-05,
|
| 1219 |
+
"loss": 0.4501,
|
| 1220 |
+
"step": 173
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 3.5968992248062017,
|
| 1224 |
+
"grad_norm": 0.12687428874084813,
|
| 1225 |
+
"learning_rate": 1.7056942545958167e-05,
|
| 1226 |
+
"loss": 0.4566,
|
| 1227 |
+
"step": 174
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"epoch": 3.6175710594315245,
|
| 1231 |
+
"grad_norm": 0.11163617587345787,
|
| 1232 |
+
"learning_rate": 1.6582822649202382e-05,
|
| 1233 |
+
"loss": 0.4585,
|
| 1234 |
+
"step": 175
|
| 1235 |
+
},
|
| 1236 |
+
{
|
| 1237 |
+
"epoch": 3.6382428940568476,
|
| 1238 |
+
"grad_norm": 0.11921623439814481,
|
| 1239 |
+
"learning_rate": 1.6113656331888563e-05,
|
| 1240 |
+
"loss": 0.4562,
|
| 1241 |
+
"step": 176
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 3.6589147286821704,
|
| 1245 |
+
"grad_norm": 0.11653048811823409,
|
| 1246 |
+
"learning_rate": 1.5649542839651175e-05,
|
| 1247 |
+
"loss": 0.4531,
|
| 1248 |
+
"step": 177
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 3.6795865633074936,
|
| 1252 |
+
"grad_norm": 0.11313755320317455,
|
| 1253 |
+
"learning_rate": 1.5190580349269604e-05,
|
| 1254 |
+
"loss": 0.4566,
|
| 1255 |
+
"step": 178
|
| 1256 |
+
},
|
| 1257 |
+
{
|
| 1258 |
+
"epoch": 3.7002583979328163,
|
| 1259 |
+
"grad_norm": 0.11320756383512366,
|
| 1260 |
+
"learning_rate": 1.4736865947900106e-05,
|
| 1261 |
+
"loss": 0.4551,
|
| 1262 |
+
"step": 179
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 3.7209302325581395,
|
| 1266 |
+
"grad_norm": 0.11504844831356965,
|
| 1267 |
+
"learning_rate": 1.4288495612538427e-05,
|
| 1268 |
+
"loss": 0.4544,
|
| 1269 |
+
"step": 180
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 3.7416020671834627,
|
| 1273 |
+
"grad_norm": 0.11168064710945105,
|
| 1274 |
+
"learning_rate": 1.3845564189717218e-05,
|
| 1275 |
+
"loss": 0.4505,
|
| 1276 |
+
"step": 181
|
| 1277 |
+
},
|
| 1278 |
+
{
|
| 1279 |
+
"epoch": 3.7622739018087854,
|
| 1280 |
+
"grad_norm": 0.11214166691765776,
|
| 1281 |
+
"learning_rate": 1.3408165375442486e-05,
|
| 1282 |
+
"loss": 0.4579,
|
| 1283 |
+
"step": 182
|
| 1284 |
+
},
|
| 1285 |
+
{
|
| 1286 |
+
"epoch": 3.7829457364341086,
|
| 1287 |
+
"grad_norm": 0.11069259941050201,
|
| 1288 |
+
"learning_rate": 1.297639169537359e-05,
|
| 1289 |
+
"loss": 0.4551,
|
| 1290 |
+
"step": 183
|
| 1291 |
+
},
|
| 1292 |
+
{
|
| 1293 |
+
"epoch": 3.8036175710594318,
|
| 1294 |
+
"grad_norm": 0.11028480012302656,
|
| 1295 |
+
"learning_rate": 1.2550334485250661e-05,
|
| 1296 |
+
"loss": 0.4592,
|
| 1297 |
+
"step": 184
|
| 1298 |
+
},
|
| 1299 |
+
{
|
| 1300 |
+
"epoch": 3.8242894056847545,
|
| 1301 |
+
"grad_norm": 0.10560438491744331,
|
| 1302 |
+
"learning_rate": 1.2130083871573812e-05,
|
| 1303 |
+
"loss": 0.4575,
|
| 1304 |
+
"step": 185
|
| 1305 |
+
},
|
| 1306 |
+
{
|
| 1307 |
+
"epoch": 3.8449612403100772,
|
| 1308 |
+
"grad_norm": 0.10673216512452338,
|
| 1309 |
+
"learning_rate": 1.1715728752538103e-05,
|
| 1310 |
+
"loss": 0.4506,
|
| 1311 |
+
"step": 186
|
| 1312 |
+
},
|
| 1313 |
+
{
|
| 1314 |
+
"epoch": 3.8656330749354004,
|
| 1315 |
+
"grad_norm": 0.11303287344305396,
|
| 1316 |
+
"learning_rate": 1.130735677922842e-05,
|
| 1317 |
+
"loss": 0.4581,
|
| 1318 |
+
"step": 187
|
| 1319 |
+
},
|
| 1320 |
+
{
|
| 1321 |
+
"epoch": 3.8863049095607236,
|
| 1322 |
+
"grad_norm": 0.10773397802575571,
|
| 1323 |
+
"learning_rate": 1.0905054337078051e-05,
|
| 1324 |
+
"loss": 0.4577,
|
| 1325 |
+
"step": 188
|
| 1326 |
+
},
|
| 1327 |
+
{
|
| 1328 |
+
"epoch": 3.9069767441860463,
|
| 1329 |
+
"grad_norm": 0.11332549654143391,
|
| 1330 |
+
"learning_rate": 1.0508906527595042e-05,
|
| 1331 |
+
"loss": 0.4509,
|
| 1332 |
+
"step": 189
|
| 1333 |
+
},
|
| 1334 |
+
{
|
| 1335 |
+
"epoch": 3.9276485788113695,
|
| 1336 |
+
"grad_norm": 0.10027531207102439,
|
| 1337 |
+
"learning_rate": 1.0118997150360169e-05,
|
| 1338 |
+
"loss": 0.4581,
|
| 1339 |
+
"step": 190
|
| 1340 |
+
},
|
| 1341 |
+
{
|
| 1342 |
+
"epoch": 3.9483204134366927,
|
| 1343 |
+
"grad_norm": 0.09956668267248524,
|
| 1344 |
+
"learning_rate": 9.735408685300287e-06,
|
| 1345 |
+
"loss": 0.4578,
|
| 1346 |
+
"step": 191
|
| 1347 |
+
},
|
| 1348 |
+
{
|
| 1349 |
+
"epoch": 3.9689922480620154,
|
| 1350 |
+
"grad_norm": 0.1097625167507584,
|
| 1351 |
+
"learning_rate": 9.358222275240884e-06,
|
| 1352 |
+
"loss": 0.4583,
|
| 1353 |
+
"step": 192
|
| 1354 |
+
},
|
| 1355 |
+
{
|
| 1356 |
+
"epoch": 3.9896640826873386,
|
| 1357 |
+
"grad_norm": 0.11050922470985575,
|
| 1358 |
+
"learning_rate": 8.987517708741364e-06,
|
| 1359 |
+
"loss": 0.4527,
|
| 1360 |
+
"step": 193
|
| 1361 |
+
},
|
| 1362 |
+
{
|
| 1363 |
+
"epoch": 4.010335917312662,
|
| 1364 |
+
"grad_norm": 0.12409482900195296,
|
| 1365 |
+
"learning_rate": 8.623373403216972e-06,
|
| 1366 |
+
"loss": 0.4428,
|
| 1367 |
+
"step": 194
|
| 1368 |
+
},
|
| 1369 |
+
{
|
| 1370 |
+
"epoch": 4.0310077519379846,
|
| 1371 |
+
"grad_norm": 0.1584348169461383,
|
| 1372 |
+
"learning_rate": 8.265866388350598e-06,
|
| 1373 |
+
"loss": 0.434,
|
| 1374 |
+
"step": 195
|
| 1375 |
+
},
|
| 1376 |
+
{
|
| 1377 |
+
"epoch": 4.051679586563307,
|
| 1378 |
+
"grad_norm": 0.12067683132671836,
|
| 1379 |
+
"learning_rate": 7.915072289798247e-06,
|
| 1380 |
+
"loss": 0.4242,
|
| 1381 |
+
"step": 196
|
| 1382 |
+
},
|
| 1383 |
+
{
|
| 1384 |
+
"epoch": 4.072351421188631,
|
| 1385 |
+
"grad_norm": 0.10434827899570648,
|
| 1386 |
+
"learning_rate": 7.5710653131915125e-06,
|
| 1387 |
+
"loss": 0.4334,
|
| 1388 |
+
"step": 197
|
| 1389 |
+
},
|
| 1390 |
+
{
|
| 1391 |
+
"epoch": 4.093023255813954,
|
| 1392 |
+
"grad_norm": 0.12218371907141959,
|
| 1393 |
+
"learning_rate": 7.233918228440324e-06,
|
| 1394 |
+
"loss": 0.4357,
|
| 1395 |
+
"step": 198
|
| 1396 |
+
},
|
| 1397 |
+
{
|
| 1398 |
+
"epoch": 4.113695090439276,
|
| 1399 |
+
"grad_norm": 0.12849422284351983,
|
| 1400 |
+
"learning_rate": 6.903702354339578e-06,
|
| 1401 |
+
"loss": 0.4281,
|
| 1402 |
+
"step": 199
|
| 1403 |
+
},
|
| 1404 |
+
{
|
| 1405 |
+
"epoch": 4.134366925064599,
|
| 1406 |
+
"grad_norm": 0.13122817840071574,
|
| 1407 |
+
"learning_rate": 6.58048754348255e-06,
|
| 1408 |
+
"loss": 0.4372,
|
| 1409 |
+
"step": 200
|
| 1410 |
+
},
|
| 1411 |
+
{
|
| 1412 |
+
"epoch": 4.155038759689923,
|
| 1413 |
+
"grad_norm": 0.11785316994314132,
|
| 1414 |
+
"learning_rate": 6.26434216748458e-06,
|
| 1415 |
+
"loss": 0.4274,
|
| 1416 |
+
"step": 201
|
| 1417 |
+
},
|
| 1418 |
+
{
|
| 1419 |
+
"epoch": 4.1757105943152455,
|
| 1420 |
+
"grad_norm": 0.1067990798857551,
|
| 1421 |
+
"learning_rate": 5.955333102520011e-06,
|
| 1422 |
+
"loss": 0.4303,
|
| 1423 |
+
"step": 202
|
| 1424 |
+
},
|
| 1425 |
+
{
|
| 1426 |
+
"epoch": 4.196382428940568,
|
| 1427 |
+
"grad_norm": 0.10890923987639589,
|
| 1428 |
+
"learning_rate": 5.653525715175483e-06,
|
| 1429 |
+
"loss": 0.4281,
|
| 1430 |
+
"step": 203
|
| 1431 |
+
},
|
| 1432 |
+
{
|
| 1433 |
+
"epoch": 4.217054263565892,
|
| 1434 |
+
"grad_norm": 0.11422387357526066,
|
| 1435 |
+
"learning_rate": 5.358983848622452e-06,
|
| 1436 |
+
"loss": 0.4282,
|
| 1437 |
+
"step": 204
|
| 1438 |
+
},
|
| 1439 |
+
{
|
| 1440 |
+
"epoch": 4.237726098191215,
|
| 1441 |
+
"grad_norm": 0.11278248966322517,
|
| 1442 |
+
"learning_rate": 5.07176980911217e-06,
|
| 1443 |
+
"loss": 0.434,
|
| 1444 |
+
"step": 205
|
| 1445 |
+
},
|
| 1446 |
+
{
|
| 1447 |
+
"epoch": 4.258397932816537,
|
| 1448 |
+
"grad_norm": 0.109846806441368,
|
| 1449 |
+
"learning_rate": 4.791944352795561e-06,
|
| 1450 |
+
"loss": 0.4305,
|
| 1451 |
+
"step": 206
|
| 1452 |
+
},
|
| 1453 |
+
{
|
| 1454 |
+
"epoch": 4.27906976744186,
|
| 1455 |
+
"grad_norm": 0.10419397518664446,
|
| 1456 |
+
"learning_rate": 4.519566672871132e-06,
|
| 1457 |
+
"loss": 0.4314,
|
| 1458 |
+
"step": 207
|
| 1459 |
+
},
|
| 1460 |
+
{
|
| 1461 |
+
"epoch": 4.299741602067184,
|
| 1462 |
+
"grad_norm": 0.10628297764969862,
|
| 1463 |
+
"learning_rate": 4.254694387063514e-06,
|
| 1464 |
+
"loss": 0.4303,
|
| 1465 |
+
"step": 208
|
| 1466 |
+
},
|
| 1467 |
+
{
|
| 1468 |
+
"epoch": 4.320413436692506,
|
| 1469 |
+
"grad_norm": 0.09429412057246614,
|
| 1470 |
+
"learning_rate": 3.997383525435154e-06,
|
| 1471 |
+
"loss": 0.4244,
|
| 1472 |
+
"step": 209
|
| 1473 |
+
},
|
| 1474 |
+
{
|
| 1475 |
+
"epoch": 4.341085271317829,
|
| 1476 |
+
"grad_norm": 0.09792240998195167,
|
| 1477 |
+
"learning_rate": 3.747688518534003e-06,
|
| 1478 |
+
"loss": 0.4332,
|
| 1479 |
+
"step": 210
|
| 1480 |
+
},
|
| 1481 |
+
{
|
| 1482 |
+
"epoch": 4.361757105943153,
|
| 1483 |
+
"grad_norm": 0.09661718795036013,
|
| 1484 |
+
"learning_rate": 3.5056621858794393e-06,
|
| 1485 |
+
"loss": 0.4226,
|
| 1486 |
+
"step": 211
|
| 1487 |
+
},
|
| 1488 |
+
{
|
| 1489 |
+
"epoch": 4.3824289405684755,
|
| 1490 |
+
"grad_norm": 0.09841756719203582,
|
| 1491 |
+
"learning_rate": 3.2713557247890447e-06,
|
| 1492 |
+
"loss": 0.4325,
|
| 1493 |
+
"step": 212
|
| 1494 |
+
},
|
| 1495 |
+
{
|
| 1496 |
+
"epoch": 4.403100775193798,
|
| 1497 |
+
"grad_norm": 0.09994005338415966,
|
| 1498 |
+
"learning_rate": 3.0448186995485307e-06,
|
| 1499 |
+
"loss": 0.4327,
|
| 1500 |
+
"step": 213
|
| 1501 |
+
},
|
| 1502 |
+
{
|
| 1503 |
+
"epoch": 4.423772609819121,
|
| 1504 |
+
"grad_norm": 0.09639474253998631,
|
| 1505 |
+
"learning_rate": 2.8260990309270987e-06,
|
| 1506 |
+
"loss": 0.4192,
|
| 1507 |
+
"step": 214
|
| 1508 |
+
},
|
| 1509 |
+
{
|
| 1510 |
+
"epoch": 4.444444444444445,
|
| 1511 |
+
"grad_norm": 0.09767324854820987,
|
| 1512 |
+
"learning_rate": 2.6152429860404647e-06,
|
| 1513 |
+
"loss": 0.4335,
|
| 1514 |
+
"step": 215
|
| 1515 |
+
},
|
| 1516 |
+
{
|
| 1517 |
+
"epoch": 4.465116279069767,
|
| 1518 |
+
"grad_norm": 0.0938288148550861,
|
| 1519 |
+
"learning_rate": 2.4122951685636674e-06,
|
| 1520 |
+
"loss": 0.4281,
|
| 1521 |
+
"step": 216
|
| 1522 |
+
},
|
| 1523 |
+
{
|
| 1524 |
+
"epoch": 4.48578811369509,
|
| 1525 |
+
"grad_norm": 0.09105375718182585,
|
| 1526 |
+
"learning_rate": 2.217298509295813e-06,
|
| 1527 |
+
"loss": 0.4268,
|
| 1528 |
+
"step": 217
|
| 1529 |
+
},
|
| 1530 |
+
{
|
| 1531 |
+
"epoch": 4.506459948320414,
|
| 1532 |
+
"grad_norm": 0.0910060602550716,
|
| 1533 |
+
"learning_rate": 2.0302942570786446e-06,
|
| 1534 |
+
"loss": 0.4286,
|
| 1535 |
+
"step": 218
|
| 1536 |
+
},
|
| 1537 |
+
{
|
| 1538 |
+
"epoch": 4.5271317829457365,
|
| 1539 |
+
"grad_norm": 0.09036874784188015,
|
| 1540 |
+
"learning_rate": 1.8513219700709272e-06,
|
| 1541 |
+
"loss": 0.43,
|
| 1542 |
+
"step": 219
|
| 1543 |
+
},
|
| 1544 |
+
{
|
| 1545 |
+
"epoch": 4.547803617571059,
|
| 1546 |
+
"grad_norm": 0.08651164285431141,
|
| 1547 |
+
"learning_rate": 1.6804195073804442e-06,
|
| 1548 |
+
"loss": 0.4249,
|
| 1549 |
+
"step": 220
|
| 1550 |
+
},
|
| 1551 |
+
{
|
| 1552 |
+
"epoch": 4.568475452196383,
|
| 1553 |
+
"grad_norm": 0.0897008811173659,
|
| 1554 |
+
"learning_rate": 1.5176230210554744e-06,
|
| 1555 |
+
"loss": 0.4236,
|
| 1556 |
+
"step": 221
|
| 1557 |
+
},
|
| 1558 |
+
{
|
| 1559 |
+
"epoch": 4.589147286821706,
|
| 1560 |
+
"grad_norm": 0.09241291657095897,
|
| 1561 |
+
"learning_rate": 1.3629669484372722e-06,
|
| 1562 |
+
"loss": 0.4291,
|
| 1563 |
+
"step": 222
|
| 1564 |
+
},
|
| 1565 |
+
{
|
| 1566 |
+
"epoch": 4.609819121447028,
|
| 1567 |
+
"grad_norm": 0.09156015733759103,
|
| 1568 |
+
"learning_rate": 1.2164840048753602e-06,
|
| 1569 |
+
"loss": 0.4295,
|
| 1570 |
+
"step": 223
|
| 1571 |
+
},
|
| 1572 |
+
{
|
| 1573 |
+
"epoch": 4.630490956072351,
|
| 1574 |
+
"grad_norm": 0.09115293170102101,
|
| 1575 |
+
"learning_rate": 1.0782051768070477e-06,
|
| 1576 |
+
"loss": 0.4308,
|
| 1577 |
+
"step": 224
|
| 1578 |
+
},
|
| 1579 |
+
{
|
| 1580 |
+
"epoch": 4.651162790697675,
|
| 1581 |
+
"grad_norm": 0.08830219798218956,
|
| 1582 |
+
"learning_rate": 9.481597152026656e-07,
|
| 1583 |
+
"loss": 0.4211,
|
| 1584 |
+
"step": 225
|
| 1585 |
+
},
|
| 1586 |
+
{
|
| 1587 |
+
"epoch": 4.671834625322997,
|
| 1588 |
+
"grad_norm": 0.0860166395242222,
|
| 1589 |
+
"learning_rate": 8.263751293779409e-07,
|
| 1590 |
+
"loss": 0.4225,
|
| 1591 |
+
"step": 226
|
| 1592 |
+
},
|
| 1593 |
+
{
|
| 1594 |
+
"epoch": 4.69250645994832,
|
| 1595 |
+
"grad_norm": 0.08646489048791478,
|
| 1596 |
+
"learning_rate": 7.128771811747737e-07,
|
| 1597 |
+
"loss": 0.4245,
|
| 1598 |
+
"step": 227
|
| 1599 |
+
},
|
| 1600 |
+
{
|
| 1601 |
+
"epoch": 4.713178294573644,
|
| 1602 |
+
"grad_norm": 0.08482681845835482,
|
| 1603 |
+
"learning_rate": 6.076898795116792e-07,
|
| 1604 |
+
"loss": 0.4307,
|
| 1605 |
+
"step": 228
|
| 1606 |
+
},
|
| 1607 |
+
{
|
| 1608 |
+
"epoch": 4.7338501291989665,
|
| 1609 |
+
"grad_norm": 0.08379658608163532,
|
| 1610 |
+
"learning_rate": 5.108354753050381e-07,
|
| 1611 |
+
"loss": 0.4266,
|
| 1612 |
+
"step": 229
|
| 1613 |
+
},
|
| 1614 |
+
{
|
| 1615 |
+
"epoch": 4.754521963824289,
|
| 1616 |
+
"grad_norm": 0.08491038557735651,
|
| 1617 |
+
"learning_rate": 4.223344567622212e-07,
|
| 1618 |
+
"loss": 0.4219,
|
| 1619 |
+
"step": 230
|
| 1620 |
+
},
|
| 1621 |
+
{
|
| 1622 |
+
"epoch": 4.775193798449612,
|
| 1623 |
+
"grad_norm": 0.08636570293069676,
|
| 1624 |
+
"learning_rate": 3.4220554504758475e-07,
|
| 1625 |
+
"loss": 0.4294,
|
| 1626 |
+
"step": 231
|
| 1627 |
+
},
|
| 1628 |
+
{
|
| 1629 |
+
"epoch": 4.795865633074936,
|
| 1630 |
+
"grad_norm": 0.08723293532233117,
|
| 1631 |
+
"learning_rate": 2.704656903222791e-07,
|
| 1632 |
+
"loss": 0.4301,
|
| 1633 |
+
"step": 232
|
| 1634 |
+
},
|
| 1635 |
+
{
|
| 1636 |
+
"epoch": 4.816537467700258,
|
| 1637 |
+
"grad_norm": 0.08820979907519637,
|
| 1638 |
+
"learning_rate": 2.0713006815868075e-07,
|
| 1639 |
+
"loss": 0.4305,
|
| 1640 |
+
"step": 233
|
| 1641 |
+
},
|
| 1642 |
+
{
|
| 1643 |
+
"epoch": 4.837209302325581,
|
| 1644 |
+
"grad_norm": 0.0838340706295227,
|
| 1645 |
+
"learning_rate": 1.522120763301782e-07,
|
| 1646 |
+
"loss": 0.4303,
|
| 1647 |
+
"step": 234
|
| 1648 |
+
},
|
| 1649 |
+
{
|
| 1650 |
+
"epoch": 4.857881136950905,
|
| 1651 |
+
"grad_norm": 0.08328787967046157,
|
| 1652 |
+
"learning_rate": 1.0572333197711005e-07,
|
| 1653 |
+
"loss": 0.4249,
|
| 1654 |
+
"step": 235
|
| 1655 |
+
},
|
| 1656 |
+
{
|
| 1657 |
+
"epoch": 4.8785529715762275,
|
| 1658 |
+
"grad_norm": 0.08645136527121673,
|
| 1659 |
+
"learning_rate": 6.767366914927298e-08,
|
| 1660 |
+
"loss": 0.4279,
|
| 1661 |
+
"step": 236
|
| 1662 |
+
},
|
| 1663 |
+
{
|
| 1664 |
+
"epoch": 4.89922480620155,
|
| 1665 |
+
"grad_norm": 0.08535798938855207,
|
| 1666 |
+
"learning_rate": 3.8071136725688074e-08,
|
| 1667 |
+
"loss": 0.4264,
|
| 1668 |
+
"step": 237
|
| 1669 |
+
},
|
| 1670 |
+
{
|
| 1671 |
+
"epoch": 4.919896640826874,
|
| 1672 |
+
"grad_norm": 0.08615112748775564,
|
| 1673 |
+
"learning_rate": 1.6921996711976028e-08,
|
| 1674 |
+
"loss": 0.4218,
|
| 1675 |
+
"step": 238
|
| 1676 |
+
},
|
| 1677 |
+
{
|
| 1678 |
+
"epoch": 4.940568475452197,
|
| 1679 |
+
"grad_norm": 0.08586983817287319,
|
| 1680 |
+
"learning_rate": 4.230722915701257e-09,
|
| 1681 |
+
"loss": 0.4293,
|
| 1682 |
+
"step": 239
|
| 1683 |
+
},
|
| 1684 |
+
{
|
| 1685 |
+
"epoch": 4.961240310077519,
|
| 1686 |
+
"grad_norm": 0.08300238292892234,
|
| 1687 |
+
"learning_rate": 0.0,
|
| 1688 |
+
"loss": 0.4271,
|
| 1689 |
+
"step": 240
|
| 1690 |
+
},
|
| 1691 |
+
{
|
| 1692 |
+
"epoch": 4.961240310077519,
|
| 1693 |
+
"step": 240,
|
| 1694 |
+
"total_flos": 6.175913847438705e+18,
|
| 1695 |
+
"train_loss": 0.5455046107371648,
|
| 1696 |
+
"train_runtime": 24740.3159,
|
| 1697 |
+
"train_samples_per_second": 5.001,
|
| 1698 |
+
"train_steps_per_second": 0.01
|
| 1699 |
+
}
|
| 1700 |
+
],
|
| 1701 |
+
"logging_steps": 1,
|
| 1702 |
+
"max_steps": 240,
|
| 1703 |
+
"num_input_tokens_seen": 0,
|
| 1704 |
+
"num_train_epochs": 5,
|
| 1705 |
+
"save_steps": 500,
|
| 1706 |
+
"stateful_callbacks": {
|
| 1707 |
+
"TrainerControl": {
|
| 1708 |
+
"args": {
|
| 1709 |
+
"should_epoch_stop": false,
|
| 1710 |
+
"should_evaluate": false,
|
| 1711 |
+
"should_log": false,
|
| 1712 |
+
"should_save": true,
|
| 1713 |
+
"should_training_stop": true
|
| 1714 |
+
},
|
| 1715 |
+
"attributes": {}
|
| 1716 |
+
}
|
| 1717 |
+
},
|
| 1718 |
+
"total_flos": 6.175913847438705e+18,
|
| 1719 |
+
"train_batch_size": 1,
|
| 1720 |
+
"trial_name": null,
|
| 1721 |
+
"trial_params": null
|
| 1722 |
+
}
|
training_loss.png
ADDED
|