End of training
Browse files- README.md +2 -1
- all_results.json +8 -0
- train_results.json +8 -0
- trainer_state.json +1862 -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: 10k_globalbatchsize192_lr4e5_epochs5
|
|
@@ -15,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 15 |
|
| 16 |
# 10k_globalbatchsize192_lr4e5_epochs5
|
| 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: 10k_globalbatchsize192_lr4e5_epochs5
|
|
|
|
| 16 |
|
| 17 |
# 10k_globalbatchsize192_lr4e5_epochs5
|
| 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/openthoughts_10000 dataset.
|
| 20 |
|
| 21 |
## Model description
|
| 22 |
|
all_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 4.992,
|
| 3 |
+
"total_flos": 1.013794005747499e+18,
|
| 4 |
+
"train_loss": 0.39605270406374565,
|
| 5 |
+
"train_runtime": 28764.0375,
|
| 6 |
+
"train_samples_per_second": 1.738,
|
| 7 |
+
"train_steps_per_second": 0.009
|
| 8 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 4.992,
|
| 3 |
+
"total_flos": 1.013794005747499e+18,
|
| 4 |
+
"train_loss": 0.39605270406374565,
|
| 5 |
+
"train_runtime": 28764.0375,
|
| 6 |
+
"train_samples_per_second": 1.738,
|
| 7 |
+
"train_steps_per_second": 0.009
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1862 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 4.992,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 260,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.0192,
|
| 13 |
+
"grad_norm": 5.847007281913707,
|
| 14 |
+
"learning_rate": 1.5384615384615387e-06,
|
| 15 |
+
"loss": 0.8597,
|
| 16 |
+
"step": 1
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.0384,
|
| 20 |
+
"grad_norm": 5.931738062420693,
|
| 21 |
+
"learning_rate": 3.0769230769230774e-06,
|
| 22 |
+
"loss": 0.8763,
|
| 23 |
+
"step": 2
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.0576,
|
| 27 |
+
"grad_norm": 5.656406269355022,
|
| 28 |
+
"learning_rate": 4.615384615384616e-06,
|
| 29 |
+
"loss": 0.8534,
|
| 30 |
+
"step": 3
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.0768,
|
| 34 |
+
"grad_norm": 4.287541515348409,
|
| 35 |
+
"learning_rate": 6.153846153846155e-06,
|
| 36 |
+
"loss": 0.7935,
|
| 37 |
+
"step": 4
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.096,
|
| 41 |
+
"grad_norm": 2.3534068142647597,
|
| 42 |
+
"learning_rate": 7.692307692307694e-06,
|
| 43 |
+
"loss": 0.7946,
|
| 44 |
+
"step": 5
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.1152,
|
| 48 |
+
"grad_norm": 1.9543083149631972,
|
| 49 |
+
"learning_rate": 9.230769230769232e-06,
|
| 50 |
+
"loss": 0.7613,
|
| 51 |
+
"step": 6
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.1344,
|
| 55 |
+
"grad_norm": 3.9371469840680877,
|
| 56 |
+
"learning_rate": 1.076923076923077e-05,
|
| 57 |
+
"loss": 0.7641,
|
| 58 |
+
"step": 7
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.1536,
|
| 62 |
+
"grad_norm": 4.17184450682719,
|
| 63 |
+
"learning_rate": 1.230769230769231e-05,
|
| 64 |
+
"loss": 0.755,
|
| 65 |
+
"step": 8
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.1728,
|
| 69 |
+
"grad_norm": 4.727273546652897,
|
| 70 |
+
"learning_rate": 1.3846153846153847e-05,
|
| 71 |
+
"loss": 0.7501,
|
| 72 |
+
"step": 9
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.192,
|
| 76 |
+
"grad_norm": 3.158614416439187,
|
| 77 |
+
"learning_rate": 1.5384615384615387e-05,
|
| 78 |
+
"loss": 0.7147,
|
| 79 |
+
"step": 10
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.2112,
|
| 83 |
+
"grad_norm": 2.1353266419901678,
|
| 84 |
+
"learning_rate": 1.6923076923076924e-05,
|
| 85 |
+
"loss": 0.6831,
|
| 86 |
+
"step": 11
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.2304,
|
| 90 |
+
"grad_norm": 2.297454949298204,
|
| 91 |
+
"learning_rate": 1.8461538461538465e-05,
|
| 92 |
+
"loss": 0.667,
|
| 93 |
+
"step": 12
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.2496,
|
| 97 |
+
"grad_norm": 1.452362183878892,
|
| 98 |
+
"learning_rate": 2e-05,
|
| 99 |
+
"loss": 0.6296,
|
| 100 |
+
"step": 13
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.2688,
|
| 104 |
+
"grad_norm": 1.3292654053570208,
|
| 105 |
+
"learning_rate": 2.153846153846154e-05,
|
| 106 |
+
"loss": 0.6104,
|
| 107 |
+
"step": 14
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.288,
|
| 111 |
+
"grad_norm": 1.3119179199635396,
|
| 112 |
+
"learning_rate": 2.3076923076923076e-05,
|
| 113 |
+
"loss": 0.629,
|
| 114 |
+
"step": 15
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.3072,
|
| 118 |
+
"grad_norm": 0.8324165174493602,
|
| 119 |
+
"learning_rate": 2.461538461538462e-05,
|
| 120 |
+
"loss": 0.6211,
|
| 121 |
+
"step": 16
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.3264,
|
| 125 |
+
"grad_norm": 1.1093447453645293,
|
| 126 |
+
"learning_rate": 2.6153846153846157e-05,
|
| 127 |
+
"loss": 0.6119,
|
| 128 |
+
"step": 17
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.3456,
|
| 132 |
+
"grad_norm": 0.7072674647777949,
|
| 133 |
+
"learning_rate": 2.7692307692307694e-05,
|
| 134 |
+
"loss": 0.5978,
|
| 135 |
+
"step": 18
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.3648,
|
| 139 |
+
"grad_norm": 0.9595194925517859,
|
| 140 |
+
"learning_rate": 2.923076923076923e-05,
|
| 141 |
+
"loss": 0.6075,
|
| 142 |
+
"step": 19
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.384,
|
| 146 |
+
"grad_norm": 0.6983851639849895,
|
| 147 |
+
"learning_rate": 3.0769230769230774e-05,
|
| 148 |
+
"loss": 0.5617,
|
| 149 |
+
"step": 20
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.4032,
|
| 153 |
+
"grad_norm": 0.9587446260295661,
|
| 154 |
+
"learning_rate": 3.230769230769231e-05,
|
| 155 |
+
"loss": 0.6013,
|
| 156 |
+
"step": 21
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.4224,
|
| 160 |
+
"grad_norm": 0.9811812661539728,
|
| 161 |
+
"learning_rate": 3.384615384615385e-05,
|
| 162 |
+
"loss": 0.5886,
|
| 163 |
+
"step": 22
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.4416,
|
| 167 |
+
"grad_norm": 1.4687581854343579,
|
| 168 |
+
"learning_rate": 3.538461538461539e-05,
|
| 169 |
+
"loss": 0.5891,
|
| 170 |
+
"step": 23
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.4608,
|
| 174 |
+
"grad_norm": 0.8358762442823919,
|
| 175 |
+
"learning_rate": 3.692307692307693e-05,
|
| 176 |
+
"loss": 0.6009,
|
| 177 |
+
"step": 24
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.48,
|
| 181 |
+
"grad_norm": 1.1845418627015347,
|
| 182 |
+
"learning_rate": 3.846153846153846e-05,
|
| 183 |
+
"loss": 0.5845,
|
| 184 |
+
"step": 25
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.4992,
|
| 188 |
+
"grad_norm": 1.2146578276187556,
|
| 189 |
+
"learning_rate": 4e-05,
|
| 190 |
+
"loss": 0.5744,
|
| 191 |
+
"step": 26
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.5184,
|
| 195 |
+
"grad_norm": 1.1346244765506583,
|
| 196 |
+
"learning_rate": 3.999819755713442e-05,
|
| 197 |
+
"loss": 0.5492,
|
| 198 |
+
"step": 27
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.5376,
|
| 202 |
+
"grad_norm": 0.9428027704273113,
|
| 203 |
+
"learning_rate": 3.999279055341771e-05,
|
| 204 |
+
"loss": 0.5514,
|
| 205 |
+
"step": 28
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.5568,
|
| 209 |
+
"grad_norm": 1.4962321798872082,
|
| 210 |
+
"learning_rate": 3.998377996343139e-05,
|
| 211 |
+
"loss": 0.5726,
|
| 212 |
+
"step": 29
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.576,
|
| 216 |
+
"grad_norm": 0.9401890626311193,
|
| 217 |
+
"learning_rate": 3.9971167411282835e-05,
|
| 218 |
+
"loss": 0.5507,
|
| 219 |
+
"step": 30
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.5952,
|
| 223 |
+
"grad_norm": 1.1843710547290214,
|
| 224 |
+
"learning_rate": 3.9954955170312504e-05,
|
| 225 |
+
"loss": 0.5348,
|
| 226 |
+
"step": 31
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.6144,
|
| 230 |
+
"grad_norm": 0.8847867205599478,
|
| 231 |
+
"learning_rate": 3.9935146162684206e-05,
|
| 232 |
+
"loss": 0.5276,
|
| 233 |
+
"step": 32
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.6336,
|
| 237 |
+
"grad_norm": 0.9561628367044479,
|
| 238 |
+
"learning_rate": 3.9911743958858376e-05,
|
| 239 |
+
"loss": 0.549,
|
| 240 |
+
"step": 33
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.6528,
|
| 244 |
+
"grad_norm": 0.7547940224197549,
|
| 245 |
+
"learning_rate": 3.9884752776948564e-05,
|
| 246 |
+
"loss": 0.5428,
|
| 247 |
+
"step": 34
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.672,
|
| 251 |
+
"grad_norm": 0.8565582543058445,
|
| 252 |
+
"learning_rate": 3.985417748196108e-05,
|
| 253 |
+
"loss": 0.5356,
|
| 254 |
+
"step": 35
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.6912,
|
| 258 |
+
"grad_norm": 0.8865926824320747,
|
| 259 |
+
"learning_rate": 3.982002358491817e-05,
|
| 260 |
+
"loss": 0.5447,
|
| 261 |
+
"step": 36
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.7104,
|
| 265 |
+
"grad_norm": 0.9805971363245475,
|
| 266 |
+
"learning_rate": 3.978229724186464e-05,
|
| 267 |
+
"loss": 0.5609,
|
| 268 |
+
"step": 37
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.7296,
|
| 272 |
+
"grad_norm": 0.8920791032060812,
|
| 273 |
+
"learning_rate": 3.9741005252758255e-05,
|
| 274 |
+
"loss": 0.532,
|
| 275 |
+
"step": 38
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.7488,
|
| 279 |
+
"grad_norm": 0.8928066131097164,
|
| 280 |
+
"learning_rate": 3.9696155060244166e-05,
|
| 281 |
+
"loss": 0.5412,
|
| 282 |
+
"step": 39
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.768,
|
| 286 |
+
"grad_norm": 0.8442960283250895,
|
| 287 |
+
"learning_rate": 3.9647754748313294e-05,
|
| 288 |
+
"loss": 0.5283,
|
| 289 |
+
"step": 40
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.7872,
|
| 293 |
+
"grad_norm": 0.8185927987364593,
|
| 294 |
+
"learning_rate": 3.959581304084536e-05,
|
| 295 |
+
"loss": 0.5401,
|
| 296 |
+
"step": 41
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.8064,
|
| 300 |
+
"grad_norm": 0.9188760187541917,
|
| 301 |
+
"learning_rate": 3.954033930003634e-05,
|
| 302 |
+
"loss": 0.5335,
|
| 303 |
+
"step": 42
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.8256,
|
| 307 |
+
"grad_norm": 0.6931270007943641,
|
| 308 |
+
"learning_rate": 3.9481343524711095e-05,
|
| 309 |
+
"loss": 0.5216,
|
| 310 |
+
"step": 43
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.8448,
|
| 314 |
+
"grad_norm": 1.0702517172694637,
|
| 315 |
+
"learning_rate": 3.9418836348521045e-05,
|
| 316 |
+
"loss": 0.5402,
|
| 317 |
+
"step": 44
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.864,
|
| 321 |
+
"grad_norm": 0.9169465504930947,
|
| 322 |
+
"learning_rate": 3.9352829038027564e-05,
|
| 323 |
+
"loss": 0.5153,
|
| 324 |
+
"step": 45
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.8832,
|
| 328 |
+
"grad_norm": 0.7616594640489186,
|
| 329 |
+
"learning_rate": 3.928333349067125e-05,
|
| 330 |
+
"loss": 0.5354,
|
| 331 |
+
"step": 46
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.9024,
|
| 335 |
+
"grad_norm": 1.2942890577056538,
|
| 336 |
+
"learning_rate": 3.921036223262745e-05,
|
| 337 |
+
"loss": 0.5306,
|
| 338 |
+
"step": 47
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.9216,
|
| 342 |
+
"grad_norm": 0.6324450385447212,
|
| 343 |
+
"learning_rate": 3.913392841654851e-05,
|
| 344 |
+
"loss": 0.5148,
|
| 345 |
+
"step": 48
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.9408,
|
| 349 |
+
"grad_norm": 1.2008627280331907,
|
| 350 |
+
"learning_rate": 3.9054045819193074e-05,
|
| 351 |
+
"loss": 0.5128,
|
| 352 |
+
"step": 49
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.96,
|
| 356 |
+
"grad_norm": 0.645331797229742,
|
| 357 |
+
"learning_rate": 3.897072883894291e-05,
|
| 358 |
+
"loss": 0.5318,
|
| 359 |
+
"step": 50
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.9792,
|
| 363 |
+
"grad_norm": 0.925525728087652,
|
| 364 |
+
"learning_rate": 3.8883992493207696e-05,
|
| 365 |
+
"loss": 0.5059,
|
| 366 |
+
"step": 51
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.9984,
|
| 370 |
+
"grad_norm": 0.7835487902477336,
|
| 371 |
+
"learning_rate": 3.879385241571817e-05,
|
| 372 |
+
"loss": 0.5266,
|
| 373 |
+
"step": 52
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 1.0176,
|
| 377 |
+
"grad_norm": 1.7615120358797791,
|
| 378 |
+
"learning_rate": 3.8700324853708304e-05,
|
| 379 |
+
"loss": 0.978,
|
| 380 |
+
"step": 53
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 1.0368,
|
| 384 |
+
"grad_norm": 1.3736934842576367,
|
| 385 |
+
"learning_rate": 3.860342666498677e-05,
|
| 386 |
+
"loss": 0.4707,
|
| 387 |
+
"step": 54
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 1.056,
|
| 391 |
+
"grad_norm": 0.7091681423698334,
|
| 392 |
+
"learning_rate": 3.850317531489848e-05,
|
| 393 |
+
"loss": 0.4695,
|
| 394 |
+
"step": 55
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 1.0752,
|
| 398 |
+
"grad_norm": 0.8153962897510232,
|
| 399 |
+
"learning_rate": 3.839958887317649e-05,
|
| 400 |
+
"loss": 0.44,
|
| 401 |
+
"step": 56
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 1.0944,
|
| 405 |
+
"grad_norm": 1.0029372177048717,
|
| 406 |
+
"learning_rate": 3.829268601068509e-05,
|
| 407 |
+
"loss": 0.4833,
|
| 408 |
+
"step": 57
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 1.1136,
|
| 412 |
+
"grad_norm": 1.2260679933931073,
|
| 413 |
+
"learning_rate": 3.818248599605448e-05,
|
| 414 |
+
"loss": 0.4643,
|
| 415 |
+
"step": 58
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 1.1328,
|
| 419 |
+
"grad_norm": 0.8835191514318156,
|
| 420 |
+
"learning_rate": 3.806900869220765e-05,
|
| 421 |
+
"loss": 0.4487,
|
| 422 |
+
"step": 59
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 1.152,
|
| 426 |
+
"grad_norm": 0.7336539313865963,
|
| 427 |
+
"learning_rate": 3.795227455278029e-05,
|
| 428 |
+
"loss": 0.4659,
|
| 429 |
+
"step": 60
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 1.1712,
|
| 433 |
+
"grad_norm": 1.1738992224611373,
|
| 434 |
+
"learning_rate": 3.783230461843406e-05,
|
| 435 |
+
"loss": 0.4557,
|
| 436 |
+
"step": 61
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 1.1904,
|
| 440 |
+
"grad_norm": 0.8632649740716217,
|
| 441 |
+
"learning_rate": 3.7709120513064196e-05,
|
| 442 |
+
"loss": 0.4702,
|
| 443 |
+
"step": 62
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 1.2096,
|
| 447 |
+
"grad_norm": 1.0186561537681438,
|
| 448 |
+
"learning_rate": 3.75827444399019e-05,
|
| 449 |
+
"loss": 0.4706,
|
| 450 |
+
"step": 63
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 1.2288000000000001,
|
| 454 |
+
"grad_norm": 0.8470570448133808,
|
| 455 |
+
"learning_rate": 3.745319917751229e-05,
|
| 456 |
+
"loss": 0.4735,
|
| 457 |
+
"step": 64
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 1.248,
|
| 461 |
+
"grad_norm": 0.9454594361767897,
|
| 462 |
+
"learning_rate": 3.732050807568878e-05,
|
| 463 |
+
"loss": 0.4697,
|
| 464 |
+
"step": 65
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 1.2671999999999999,
|
| 468 |
+
"grad_norm": 0.8038599403084322,
|
| 469 |
+
"learning_rate": 3.718469505124434e-05,
|
| 470 |
+
"loss": 0.4684,
|
| 471 |
+
"step": 66
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 1.2864,
|
| 475 |
+
"grad_norm": 0.8018939895947839,
|
| 476 |
+
"learning_rate": 3.704578458370067e-05,
|
| 477 |
+
"loss": 0.4764,
|
| 478 |
+
"step": 67
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 1.3056,
|
| 482 |
+
"grad_norm": 1.04208128671029,
|
| 483 |
+
"learning_rate": 3.69038017108759e-05,
|
| 484 |
+
"loss": 0.4663,
|
| 485 |
+
"step": 68
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 1.3248,
|
| 489 |
+
"grad_norm": 0.6360539527447016,
|
| 490 |
+
"learning_rate": 3.6758772024371626e-05,
|
| 491 |
+
"loss": 0.4614,
|
| 492 |
+
"step": 69
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 1.3439999999999999,
|
| 496 |
+
"grad_norm": 0.939451130170268,
|
| 497 |
+
"learning_rate": 3.6610721664960236e-05,
|
| 498 |
+
"loss": 0.4614,
|
| 499 |
+
"step": 70
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 1.3632,
|
| 503 |
+
"grad_norm": 0.840406307921151,
|
| 504 |
+
"learning_rate": 3.645967731787313e-05,
|
| 505 |
+
"loss": 0.476,
|
| 506 |
+
"step": 71
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 1.3824,
|
| 510 |
+
"grad_norm": 0.6919958541728773,
|
| 511 |
+
"learning_rate": 3.6305666207990886e-05,
|
| 512 |
+
"loss": 0.4431,
|
| 513 |
+
"step": 72
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 1.4016,
|
| 517 |
+
"grad_norm": 0.9404542712534945,
|
| 518 |
+
"learning_rate": 3.614871609493614e-05,
|
| 519 |
+
"loss": 0.4522,
|
| 520 |
+
"step": 73
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 1.4208,
|
| 524 |
+
"grad_norm": 0.6990229016962196,
|
| 525 |
+
"learning_rate": 3.598885526807003e-05,
|
| 526 |
+
"loss": 0.4632,
|
| 527 |
+
"step": 74
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 1.44,
|
| 531 |
+
"grad_norm": 0.9016548613546168,
|
| 532 |
+
"learning_rate": 3.582611254139324e-05,
|
| 533 |
+
"loss": 0.4634,
|
| 534 |
+
"step": 75
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 1.4592,
|
| 538 |
+
"grad_norm": 0.8488017334507152,
|
| 539 |
+
"learning_rate": 3.566051724835245e-05,
|
| 540 |
+
"loss": 0.442,
|
| 541 |
+
"step": 76
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 1.4784,
|
| 545 |
+
"grad_norm": 0.44826911117643364,
|
| 546 |
+
"learning_rate": 3.549209923655309e-05,
|
| 547 |
+
"loss": 0.4286,
|
| 548 |
+
"step": 77
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 1.4976,
|
| 552 |
+
"grad_norm": 0.9127605859032204,
|
| 553 |
+
"learning_rate": 3.532088886237956e-05,
|
| 554 |
+
"loss": 0.4633,
|
| 555 |
+
"step": 78
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 1.5168,
|
| 559 |
+
"grad_norm": 0.8211427452159049,
|
| 560 |
+
"learning_rate": 3.5146916985523604e-05,
|
| 561 |
+
"loss": 0.4421,
|
| 562 |
+
"step": 79
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 1.536,
|
| 566 |
+
"grad_norm": 0.4705140388341284,
|
| 567 |
+
"learning_rate": 3.497021496342203e-05,
|
| 568 |
+
"loss": 0.4305,
|
| 569 |
+
"step": 80
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 1.5552000000000001,
|
| 573 |
+
"grad_norm": 0.698717576934082,
|
| 574 |
+
"learning_rate": 3.479081464560475e-05,
|
| 575 |
+
"loss": 0.4399,
|
| 576 |
+
"step": 81
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 1.5744,
|
| 580 |
+
"grad_norm": 0.6025383968097947,
|
| 581 |
+
"learning_rate": 3.4608748367954064e-05,
|
| 582 |
+
"loss": 0.4458,
|
| 583 |
+
"step": 82
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 1.5936,
|
| 587 |
+
"grad_norm": 0.6268417039855624,
|
| 588 |
+
"learning_rate": 3.442404894687629e-05,
|
| 589 |
+
"loss": 0.4464,
|
| 590 |
+
"step": 83
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 1.6128,
|
| 594 |
+
"grad_norm": 0.533020603233409,
|
| 595 |
+
"learning_rate": 3.423674967338681e-05,
|
| 596 |
+
"loss": 0.4565,
|
| 597 |
+
"step": 84
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 1.6320000000000001,
|
| 601 |
+
"grad_norm": 0.5677250200555859,
|
| 602 |
+
"learning_rate": 3.404688430710955e-05,
|
| 603 |
+
"loss": 0.4427,
|
| 604 |
+
"step": 85
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 1.6512,
|
| 608 |
+
"grad_norm": 0.6660871725178085,
|
| 609 |
+
"learning_rate": 3.385448707019199e-05,
|
| 610 |
+
"loss": 0.456,
|
| 611 |
+
"step": 86
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 1.6703999999999999,
|
| 615 |
+
"grad_norm": 0.6596910185879423,
|
| 616 |
+
"learning_rate": 3.365959264113683e-05,
|
| 617 |
+
"loss": 0.4417,
|
| 618 |
+
"step": 87
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 1.6896,
|
| 622 |
+
"grad_norm": 0.516235855421342,
|
| 623 |
+
"learning_rate": 3.34622361485514e-05,
|
| 624 |
+
"loss": 0.4425,
|
| 625 |
+
"step": 88
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 1.7088,
|
| 629 |
+
"grad_norm": 0.7298989810328028,
|
| 630 |
+
"learning_rate": 3.326245316481591e-05,
|
| 631 |
+
"loss": 0.4636,
|
| 632 |
+
"step": 89
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 1.728,
|
| 636 |
+
"grad_norm": 0.7852267190513897,
|
| 637 |
+
"learning_rate": 3.30602796996717e-05,
|
| 638 |
+
"loss": 0.4406,
|
| 639 |
+
"step": 90
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 1.7471999999999999,
|
| 643 |
+
"grad_norm": 0.4679131697373689,
|
| 644 |
+
"learning_rate": 3.285575219373079e-05,
|
| 645 |
+
"loss": 0.4752,
|
| 646 |
+
"step": 91
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 1.7664,
|
| 650 |
+
"grad_norm": 0.7151718172882814,
|
| 651 |
+
"learning_rate": 3.2648907511907544e-05,
|
| 652 |
+
"loss": 0.4581,
|
| 653 |
+
"step": 92
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 1.7856,
|
| 657 |
+
"grad_norm": 0.702218276711869,
|
| 658 |
+
"learning_rate": 3.243978293677408e-05,
|
| 659 |
+
"loss": 0.4461,
|
| 660 |
+
"step": 93
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 1.8048,
|
| 664 |
+
"grad_norm": 0.47386792815227347,
|
| 665 |
+
"learning_rate": 3.222841616184025e-05,
|
| 666 |
+
"loss": 0.4232,
|
| 667 |
+
"step": 94
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 1.8239999999999998,
|
| 671 |
+
"grad_norm": 0.762720027225129,
|
| 672 |
+
"learning_rate": 3.201484528475958e-05,
|
| 673 |
+
"loss": 0.4463,
|
| 674 |
+
"step": 95
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 1.8432,
|
| 678 |
+
"grad_norm": 0.8423811597579547,
|
| 679 |
+
"learning_rate": 3.1799108800462466e-05,
|
| 680 |
+
"loss": 0.4686,
|
| 681 |
+
"step": 96
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 1.8624,
|
| 685 |
+
"grad_norm": 0.6074259910205732,
|
| 686 |
+
"learning_rate": 3.158124559421758e-05,
|
| 687 |
+
"loss": 0.445,
|
| 688 |
+
"step": 97
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 1.8816000000000002,
|
| 692 |
+
"grad_norm": 0.6199366317033334,
|
| 693 |
+
"learning_rate": 3.136129493462312e-05,
|
| 694 |
+
"loss": 0.4254,
|
| 695 |
+
"step": 98
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 1.9008,
|
| 699 |
+
"grad_norm": 0.9875902978935551,
|
| 700 |
+
"learning_rate": 3.113929646652879e-05,
|
| 701 |
+
"loss": 0.4636,
|
| 702 |
+
"step": 99
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 1.92,
|
| 706 |
+
"grad_norm": 0.615162427842402,
|
| 707 |
+
"learning_rate": 3.091529020389009e-05,
|
| 708 |
+
"loss": 0.4511,
|
| 709 |
+
"step": 100
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 1.9392,
|
| 713 |
+
"grad_norm": 0.6333989781465663,
|
| 714 |
+
"learning_rate": 3.0689316522556026e-05,
|
| 715 |
+
"loss": 0.4421,
|
| 716 |
+
"step": 101
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 1.9584000000000001,
|
| 720 |
+
"grad_norm": 0.8261498458791247,
|
| 721 |
+
"learning_rate": 3.0461416152991555e-05,
|
| 722 |
+
"loss": 0.4624,
|
| 723 |
+
"step": 102
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 1.9776,
|
| 727 |
+
"grad_norm": 0.6411717910358062,
|
| 728 |
+
"learning_rate": 3.0231630172936205e-05,
|
| 729 |
+
"loss": 0.4472,
|
| 730 |
+
"step": 103
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 1.9968,
|
| 734 |
+
"grad_norm": 0.5238750862157695,
|
| 735 |
+
"learning_rate": 3.0000000000000004e-05,
|
| 736 |
+
"loss": 0.4563,
|
| 737 |
+
"step": 104
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 2.016,
|
| 741 |
+
"grad_norm": 1.7980054051757128,
|
| 742 |
+
"learning_rate": 2.9766567384198228e-05,
|
| 743 |
+
"loss": 0.7451,
|
| 744 |
+
"step": 105
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 2.0352,
|
| 748 |
+
"grad_norm": 0.7276020848755101,
|
| 749 |
+
"learning_rate": 2.9531374400426158e-05,
|
| 750 |
+
"loss": 0.3699,
|
| 751 |
+
"step": 106
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 2.0544,
|
| 755 |
+
"grad_norm": 0.727367023844261,
|
| 756 |
+
"learning_rate": 2.9294463440875375e-05,
|
| 757 |
+
"loss": 0.3901,
|
| 758 |
+
"step": 107
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 2.0736,
|
| 762 |
+
"grad_norm": 0.8829445142802318,
|
| 763 |
+
"learning_rate": 2.9055877207392752e-05,
|
| 764 |
+
"loss": 0.3826,
|
| 765 |
+
"step": 108
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 2.0928,
|
| 769 |
+
"grad_norm": 0.7521098230762986,
|
| 770 |
+
"learning_rate": 2.8815658703783715e-05,
|
| 771 |
+
"loss": 0.3784,
|
| 772 |
+
"step": 109
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 2.112,
|
| 776 |
+
"grad_norm": 0.6040228065537577,
|
| 777 |
+
"learning_rate": 2.8573851228061084e-05,
|
| 778 |
+
"loss": 0.3539,
|
| 779 |
+
"step": 110
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 2.1312,
|
| 783 |
+
"grad_norm": 0.8162209469019132,
|
| 784 |
+
"learning_rate": 2.8330498364640803e-05,
|
| 785 |
+
"loss": 0.3512,
|
| 786 |
+
"step": 111
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 2.1504,
|
| 790 |
+
"grad_norm": 0.5692656163513317,
|
| 791 |
+
"learning_rate": 2.80856439764861e-05,
|
| 792 |
+
"loss": 0.3634,
|
| 793 |
+
"step": 112
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 2.1696,
|
| 797 |
+
"grad_norm": 0.7118480225612055,
|
| 798 |
+
"learning_rate": 2.7839332197201505e-05,
|
| 799 |
+
"loss": 0.3753,
|
| 800 |
+
"step": 113
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 2.1888,
|
| 804 |
+
"grad_norm": 0.6055440353173442,
|
| 805 |
+
"learning_rate": 2.7591607423077932e-05,
|
| 806 |
+
"loss": 0.3648,
|
| 807 |
+
"step": 114
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 2.208,
|
| 811 |
+
"grad_norm": 0.5656720068814893,
|
| 812 |
+
"learning_rate": 2.7342514305090554e-05,
|
| 813 |
+
"loss": 0.3741,
|
| 814 |
+
"step": 115
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 2.2272,
|
| 818 |
+
"grad_norm": 0.653308490898562,
|
| 819 |
+
"learning_rate": 2.7092097740850712e-05,
|
| 820 |
+
"loss": 0.361,
|
| 821 |
+
"step": 116
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 2.2464,
|
| 825 |
+
"grad_norm": 0.4576378388124488,
|
| 826 |
+
"learning_rate": 2.684040286651338e-05,
|
| 827 |
+
"loss": 0.3705,
|
| 828 |
+
"step": 117
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 2.2656,
|
| 832 |
+
"grad_norm": 0.5215105047727198,
|
| 833 |
+
"learning_rate": 2.6587475048641596e-05,
|
| 834 |
+
"loss": 0.3427,
|
| 835 |
+
"step": 118
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 2.2848,
|
| 839 |
+
"grad_norm": 0.41463913533313007,
|
| 840 |
+
"learning_rate": 2.6333359876029455e-05,
|
| 841 |
+
"loss": 0.3616,
|
| 842 |
+
"step": 119
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 2.304,
|
| 846 |
+
"grad_norm": 0.5282531910010791,
|
| 847 |
+
"learning_rate": 2.607810315148494e-05,
|
| 848 |
+
"loss": 0.3552,
|
| 849 |
+
"step": 120
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 2.3232,
|
| 853 |
+
"grad_norm": 0.5306632941377617,
|
| 854 |
+
"learning_rate": 2.582175088357426e-05,
|
| 855 |
+
"loss": 0.3448,
|
| 856 |
+
"step": 121
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 2.3424,
|
| 860 |
+
"grad_norm": 0.4051924220318362,
|
| 861 |
+
"learning_rate": 2.5564349278329056e-05,
|
| 862 |
+
"loss": 0.3452,
|
| 863 |
+
"step": 122
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 2.3616,
|
| 867 |
+
"grad_norm": 0.508616160412741,
|
| 868 |
+
"learning_rate": 2.5305944730918017e-05,
|
| 869 |
+
"loss": 0.3837,
|
| 870 |
+
"step": 123
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 2.3808,
|
| 874 |
+
"grad_norm": 0.3378524403238181,
|
| 875 |
+
"learning_rate": 2.5046583817284437e-05,
|
| 876 |
+
"loss": 0.3537,
|
| 877 |
+
"step": 124
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 2.4,
|
| 881 |
+
"grad_norm": 0.48775469762217477,
|
| 882 |
+
"learning_rate": 2.4786313285751158e-05,
|
| 883 |
+
"loss": 0.3605,
|
| 884 |
+
"step": 125
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 2.4192,
|
| 888 |
+
"grad_norm": 0.3451033020473491,
|
| 889 |
+
"learning_rate": 2.4525180048594452e-05,
|
| 890 |
+
"loss": 0.3726,
|
| 891 |
+
"step": 126
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 2.4384,
|
| 895 |
+
"grad_norm": 0.45308817090579745,
|
| 896 |
+
"learning_rate": 2.4263231173588323e-05,
|
| 897 |
+
"loss": 0.3543,
|
| 898 |
+
"step": 127
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 2.4576000000000002,
|
| 902 |
+
"grad_norm": 0.33853826482562827,
|
| 903 |
+
"learning_rate": 2.4000513875520892e-05,
|
| 904 |
+
"loss": 0.3587,
|
| 905 |
+
"step": 128
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 2.4768,
|
| 909 |
+
"grad_norm": 0.47330873339431967,
|
| 910 |
+
"learning_rate": 2.3737075507684103e-05,
|
| 911 |
+
"loss": 0.3693,
|
| 912 |
+
"step": 129
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 2.496,
|
| 916 |
+
"grad_norm": 0.373965136690713,
|
| 917 |
+
"learning_rate": 2.3472963553338614e-05,
|
| 918 |
+
"loss": 0.3639,
|
| 919 |
+
"step": 130
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 2.5152,
|
| 923 |
+
"grad_norm": 0.41340998047103167,
|
| 924 |
+
"learning_rate": 2.3208225617155206e-05,
|
| 925 |
+
"loss": 0.3599,
|
| 926 |
+
"step": 131
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 2.5343999999999998,
|
| 930 |
+
"grad_norm": 0.40367459643441794,
|
| 931 |
+
"learning_rate": 2.2942909416634326e-05,
|
| 932 |
+
"loss": 0.3574,
|
| 933 |
+
"step": 132
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 2.5536,
|
| 937 |
+
"grad_norm": 0.3638389343120581,
|
| 938 |
+
"learning_rate": 2.2677062773505235e-05,
|
| 939 |
+
"loss": 0.3706,
|
| 940 |
+
"step": 133
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 2.5728,
|
| 944 |
+
"grad_norm": 0.3980869313223262,
|
| 945 |
+
"learning_rate": 2.2410733605106462e-05,
|
| 946 |
+
"loss": 0.3497,
|
| 947 |
+
"step": 134
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 2.592,
|
| 951 |
+
"grad_norm": 0.35795198642986037,
|
| 952 |
+
"learning_rate": 2.214396991574896e-05,
|
| 953 |
+
"loss": 0.3549,
|
| 954 |
+
"step": 135
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 2.6112,
|
| 958 |
+
"grad_norm": 0.340140481057047,
|
| 959 |
+
"learning_rate": 2.1876819788063586e-05,
|
| 960 |
+
"loss": 0.3451,
|
| 961 |
+
"step": 136
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 2.6304,
|
| 965 |
+
"grad_norm": 0.30874903712848367,
|
| 966 |
+
"learning_rate": 2.1609331374334523e-05,
|
| 967 |
+
"loss": 0.3845,
|
| 968 |
+
"step": 137
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 2.6496,
|
| 972 |
+
"grad_norm": 0.34454659636746066,
|
| 973 |
+
"learning_rate": 2.1341552887820048e-05,
|
| 974 |
+
"loss": 0.3661,
|
| 975 |
+
"step": 138
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 2.6688,
|
| 979 |
+
"grad_norm": 0.3016655303090405,
|
| 980 |
+
"learning_rate": 2.1073532594062432e-05,
|
| 981 |
+
"loss": 0.3559,
|
| 982 |
+
"step": 139
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 2.6879999999999997,
|
| 986 |
+
"grad_norm": 0.35646418433985827,
|
| 987 |
+
"learning_rate": 2.0805318802188307e-05,
|
| 988 |
+
"loss": 0.3631,
|
| 989 |
+
"step": 140
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 2.7072000000000003,
|
| 993 |
+
"grad_norm": 0.317241905396711,
|
| 994 |
+
"learning_rate": 2.053695985620123e-05,
|
| 995 |
+
"loss": 0.3511,
|
| 996 |
+
"step": 141
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 2.7264,
|
| 1000 |
+
"grad_norm": 0.30386353019207646,
|
| 1001 |
+
"learning_rate": 2.0268504126267952e-05,
|
| 1002 |
+
"loss": 0.3435,
|
| 1003 |
+
"step": 142
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 2.7456,
|
| 1007 |
+
"grad_norm": 0.3292894480712393,
|
| 1008 |
+
"learning_rate": 2e-05,
|
| 1009 |
+
"loss": 0.3653,
|
| 1010 |
+
"step": 143
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 2.7648,
|
| 1014 |
+
"grad_norm": 0.32346177330376885,
|
| 1015 |
+
"learning_rate": 1.9731495873732055e-05,
|
| 1016 |
+
"loss": 0.3668,
|
| 1017 |
+
"step": 144
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 2.784,
|
| 1021 |
+
"grad_norm": 0.3769316147989722,
|
| 1022 |
+
"learning_rate": 1.946304014379878e-05,
|
| 1023 |
+
"loss": 0.3866,
|
| 1024 |
+
"step": 145
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 2.8032,
|
| 1028 |
+
"grad_norm": 0.3145062658156527,
|
| 1029 |
+
"learning_rate": 1.9194681197811703e-05,
|
| 1030 |
+
"loss": 0.357,
|
| 1031 |
+
"step": 146
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 2.8224,
|
| 1035 |
+
"grad_norm": 0.3042611587089968,
|
| 1036 |
+
"learning_rate": 1.8926467405937567e-05,
|
| 1037 |
+
"loss": 0.3421,
|
| 1038 |
+
"step": 147
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 2.8416,
|
| 1042 |
+
"grad_norm": 0.3318042133827233,
|
| 1043 |
+
"learning_rate": 1.8658447112179952e-05,
|
| 1044 |
+
"loss": 0.4035,
|
| 1045 |
+
"step": 148
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 2.8608000000000002,
|
| 1049 |
+
"grad_norm": 0.3491886102217581,
|
| 1050 |
+
"learning_rate": 1.8390668625665483e-05,
|
| 1051 |
+
"loss": 0.3729,
|
| 1052 |
+
"step": 149
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 2.88,
|
| 1056 |
+
"grad_norm": 0.30230937174058775,
|
| 1057 |
+
"learning_rate": 1.8123180211936417e-05,
|
| 1058 |
+
"loss": 0.349,
|
| 1059 |
+
"step": 150
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 2.8992,
|
| 1063 |
+
"grad_norm": 0.32697046584785133,
|
| 1064 |
+
"learning_rate": 1.7856030084251045e-05,
|
| 1065 |
+
"loss": 0.3776,
|
| 1066 |
+
"step": 151
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 2.9184,
|
| 1070 |
+
"grad_norm": 0.39727532096773555,
|
| 1071 |
+
"learning_rate": 1.758926639489354e-05,
|
| 1072 |
+
"loss": 0.3965,
|
| 1073 |
+
"step": 152
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 2.9375999999999998,
|
| 1077 |
+
"grad_norm": 0.3243977706295201,
|
| 1078 |
+
"learning_rate": 1.7322937226494775e-05,
|
| 1079 |
+
"loss": 0.3497,
|
| 1080 |
+
"step": 153
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 2.9568,
|
| 1084 |
+
"grad_norm": 0.3906432277375083,
|
| 1085 |
+
"learning_rate": 1.7057090583365678e-05,
|
| 1086 |
+
"loss": 0.379,
|
| 1087 |
+
"step": 154
|
| 1088 |
+
},
|
| 1089 |
+
{
|
| 1090 |
+
"epoch": 2.976,
|
| 1091 |
+
"grad_norm": 0.30560806509852734,
|
| 1092 |
+
"learning_rate": 1.6791774382844794e-05,
|
| 1093 |
+
"loss": 0.3462,
|
| 1094 |
+
"step": 155
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 2.9952,
|
| 1098 |
+
"grad_norm": 0.3289075039433621,
|
| 1099 |
+
"learning_rate": 1.6527036446661396e-05,
|
| 1100 |
+
"loss": 0.3689,
|
| 1101 |
+
"step": 156
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 3.0144,
|
| 1105 |
+
"grad_norm": 1.0744244522661688,
|
| 1106 |
+
"learning_rate": 1.6262924492315907e-05,
|
| 1107 |
+
"loss": 0.6347,
|
| 1108 |
+
"step": 157
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"epoch": 3.0336,
|
| 1112 |
+
"grad_norm": 0.45002810316657843,
|
| 1113 |
+
"learning_rate": 1.5999486124479115e-05,
|
| 1114 |
+
"loss": 0.3047,
|
| 1115 |
+
"step": 158
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 3.0528,
|
| 1119 |
+
"grad_norm": 0.5578965378664995,
|
| 1120 |
+
"learning_rate": 1.5736768826411683e-05,
|
| 1121 |
+
"loss": 0.2932,
|
| 1122 |
+
"step": 159
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 3.072,
|
| 1126 |
+
"grad_norm": 0.5266309071534713,
|
| 1127 |
+
"learning_rate": 1.547481995140556e-05,
|
| 1128 |
+
"loss": 0.3001,
|
| 1129 |
+
"step": 160
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 3.0912,
|
| 1133 |
+
"grad_norm": 0.4396590372688216,
|
| 1134 |
+
"learning_rate": 1.5213686714248852e-05,
|
| 1135 |
+
"loss": 0.2756,
|
| 1136 |
+
"step": 161
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 3.1104,
|
| 1140 |
+
"grad_norm": 0.5071431892234814,
|
| 1141 |
+
"learning_rate": 1.4953416182715566e-05,
|
| 1142 |
+
"loss": 0.3178,
|
| 1143 |
+
"step": 162
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 3.1296,
|
| 1147 |
+
"grad_norm": 0.42903567761053213,
|
| 1148 |
+
"learning_rate": 1.4694055269081987e-05,
|
| 1149 |
+
"loss": 0.2826,
|
| 1150 |
+
"step": 163
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 3.1488,
|
| 1154 |
+
"grad_norm": 0.36593038049111837,
|
| 1155 |
+
"learning_rate": 1.443565072167095e-05,
|
| 1156 |
+
"loss": 0.2696,
|
| 1157 |
+
"step": 164
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 3.168,
|
| 1161 |
+
"grad_norm": 0.4220071295447752,
|
| 1162 |
+
"learning_rate": 1.4178249116425744e-05,
|
| 1163 |
+
"loss": 0.3022,
|
| 1164 |
+
"step": 165
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 3.1872,
|
| 1168 |
+
"grad_norm": 0.43925847720899547,
|
| 1169 |
+
"learning_rate": 1.3921896848515064e-05,
|
| 1170 |
+
"loss": 0.2916,
|
| 1171 |
+
"step": 166
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 3.2064,
|
| 1175 |
+
"grad_norm": 0.3322572214747103,
|
| 1176 |
+
"learning_rate": 1.3666640123970556e-05,
|
| 1177 |
+
"loss": 0.2669,
|
| 1178 |
+
"step": 167
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 3.2256,
|
| 1182 |
+
"grad_norm": 0.37286662756436534,
|
| 1183 |
+
"learning_rate": 1.341252495135841e-05,
|
| 1184 |
+
"loss": 0.2971,
|
| 1185 |
+
"step": 168
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 3.2448,
|
| 1189 |
+
"grad_norm": 0.3433861317101218,
|
| 1190 |
+
"learning_rate": 1.3159597133486628e-05,
|
| 1191 |
+
"loss": 0.281,
|
| 1192 |
+
"step": 169
|
| 1193 |
+
},
|
| 1194 |
+
{
|
| 1195 |
+
"epoch": 3.2640000000000002,
|
| 1196 |
+
"grad_norm": 0.3115476999005927,
|
| 1197 |
+
"learning_rate": 1.2907902259149287e-05,
|
| 1198 |
+
"loss": 0.2636,
|
| 1199 |
+
"step": 170
|
| 1200 |
+
},
|
| 1201 |
+
{
|
| 1202 |
+
"epoch": 3.2832,
|
| 1203 |
+
"grad_norm": 0.3420536300923057,
|
| 1204 |
+
"learning_rate": 1.2657485694909451e-05,
|
| 1205 |
+
"loss": 0.2983,
|
| 1206 |
+
"step": 171
|
| 1207 |
+
},
|
| 1208 |
+
{
|
| 1209 |
+
"epoch": 3.3024,
|
| 1210 |
+
"grad_norm": 0.32836155358706837,
|
| 1211 |
+
"learning_rate": 1.2408392576922075e-05,
|
| 1212 |
+
"loss": 0.2875,
|
| 1213 |
+
"step": 172
|
| 1214 |
+
},
|
| 1215 |
+
{
|
| 1216 |
+
"epoch": 3.3216,
|
| 1217 |
+
"grad_norm": 0.31152168259460605,
|
| 1218 |
+
"learning_rate": 1.2160667802798503e-05,
|
| 1219 |
+
"loss": 0.2806,
|
| 1220 |
+
"step": 173
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 3.3407999999999998,
|
| 1224 |
+
"grad_norm": 0.34671769878819064,
|
| 1225 |
+
"learning_rate": 1.1914356023513904e-05,
|
| 1226 |
+
"loss": 0.2913,
|
| 1227 |
+
"step": 174
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"epoch": 3.36,
|
| 1231 |
+
"grad_norm": 0.30502064747125873,
|
| 1232 |
+
"learning_rate": 1.1669501635359212e-05,
|
| 1233 |
+
"loss": 0.2681,
|
| 1234 |
+
"step": 175
|
| 1235 |
+
},
|
| 1236 |
+
{
|
| 1237 |
+
"epoch": 3.3792,
|
| 1238 |
+
"grad_norm": 0.3010224113476069,
|
| 1239 |
+
"learning_rate": 1.1426148771938915e-05,
|
| 1240 |
+
"loss": 0.2722,
|
| 1241 |
+
"step": 176
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 3.3984,
|
| 1245 |
+
"grad_norm": 0.3255475578114101,
|
| 1246 |
+
"learning_rate": 1.1184341296216281e-05,
|
| 1247 |
+
"loss": 0.2879,
|
| 1248 |
+
"step": 177
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 3.4176,
|
| 1252 |
+
"grad_norm": 0.2976000653733745,
|
| 1253 |
+
"learning_rate": 1.094412279260726e-05,
|
| 1254 |
+
"loss": 0.3085,
|
| 1255 |
+
"step": 178
|
| 1256 |
+
},
|
| 1257 |
+
{
|
| 1258 |
+
"epoch": 3.4368,
|
| 1259 |
+
"grad_norm": 0.28898862229483296,
|
| 1260 |
+
"learning_rate": 1.070553655912463e-05,
|
| 1261 |
+
"loss": 0.2646,
|
| 1262 |
+
"step": 179
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 3.456,
|
| 1266 |
+
"grad_norm": 0.30613876148246727,
|
| 1267 |
+
"learning_rate": 1.0468625599573842e-05,
|
| 1268 |
+
"loss": 0.3029,
|
| 1269 |
+
"step": 180
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 3.4752,
|
| 1273 |
+
"grad_norm": 0.2867455330514505,
|
| 1274 |
+
"learning_rate": 1.0233432615801786e-05,
|
| 1275 |
+
"loss": 0.2662,
|
| 1276 |
+
"step": 181
|
| 1277 |
+
},
|
| 1278 |
+
{
|
| 1279 |
+
"epoch": 3.4944,
|
| 1280 |
+
"grad_norm": 0.321535584872095,
|
| 1281 |
+
"learning_rate": 1.0000000000000006e-05,
|
| 1282 |
+
"loss": 0.3194,
|
| 1283 |
+
"step": 182
|
| 1284 |
+
},
|
| 1285 |
+
{
|
| 1286 |
+
"epoch": 3.5136,
|
| 1287 |
+
"grad_norm": 0.2799539422012591,
|
| 1288 |
+
"learning_rate": 9.768369827063803e-06,
|
| 1289 |
+
"loss": 0.2869,
|
| 1290 |
+
"step": 183
|
| 1291 |
+
},
|
| 1292 |
+
{
|
| 1293 |
+
"epoch": 3.5328,
|
| 1294 |
+
"grad_norm": 0.32692548189893855,
|
| 1295 |
+
"learning_rate": 9.538583847008452e-06,
|
| 1296 |
+
"loss": 0.272,
|
| 1297 |
+
"step": 184
|
| 1298 |
+
},
|
| 1299 |
+
{
|
| 1300 |
+
"epoch": 3.552,
|
| 1301 |
+
"grad_norm": 0.3016574029313282,
|
| 1302 |
+
"learning_rate": 9.310683477443979e-06,
|
| 1303 |
+
"loss": 0.2873,
|
| 1304 |
+
"step": 185
|
| 1305 |
+
},
|
| 1306 |
+
{
|
| 1307 |
+
"epoch": 3.5712,
|
| 1308 |
+
"grad_norm": 0.2842885247117643,
|
| 1309 |
+
"learning_rate": 9.084709796109907e-06,
|
| 1310 |
+
"loss": 0.2851,
|
| 1311 |
+
"step": 186
|
| 1312 |
+
},
|
| 1313 |
+
{
|
| 1314 |
+
"epoch": 3.5904,
|
| 1315 |
+
"grad_norm": 0.34904163312850944,
|
| 1316 |
+
"learning_rate": 8.860703533471218e-06,
|
| 1317 |
+
"loss": 0.2801,
|
| 1318 |
+
"step": 187
|
| 1319 |
+
},
|
| 1320 |
+
{
|
| 1321 |
+
"epoch": 3.6096,
|
| 1322 |
+
"grad_norm": 0.2714443358558226,
|
| 1323 |
+
"learning_rate": 8.638705065376887e-06,
|
| 1324 |
+
"loss": 0.2822,
|
| 1325 |
+
"step": 188
|
| 1326 |
+
},
|
| 1327 |
+
{
|
| 1328 |
+
"epoch": 3.6288,
|
| 1329 |
+
"grad_norm": 0.27180553837932087,
|
| 1330 |
+
"learning_rate": 8.418754405782423e-06,
|
| 1331 |
+
"loss": 0.2721,
|
| 1332 |
+
"step": 189
|
| 1333 |
+
},
|
| 1334 |
+
{
|
| 1335 |
+
"epoch": 3.648,
|
| 1336 |
+
"grad_norm": 0.3185134908684331,
|
| 1337 |
+
"learning_rate": 8.200891199537549e-06,
|
| 1338 |
+
"loss": 0.2843,
|
| 1339 |
+
"step": 190
|
| 1340 |
+
},
|
| 1341 |
+
{
|
| 1342 |
+
"epoch": 3.6672000000000002,
|
| 1343 |
+
"grad_norm": 0.26988818834945916,
|
| 1344 |
+
"learning_rate": 7.98515471524042e-06,
|
| 1345 |
+
"loss": 0.2812,
|
| 1346 |
+
"step": 191
|
| 1347 |
+
},
|
| 1348 |
+
{
|
| 1349 |
+
"epoch": 3.6864,
|
| 1350 |
+
"grad_norm": 0.2760707580517701,
|
| 1351 |
+
"learning_rate": 7.771583838159756e-06,
|
| 1352 |
+
"loss": 0.2867,
|
| 1353 |
+
"step": 192
|
| 1354 |
+
},
|
| 1355 |
+
{
|
| 1356 |
+
"epoch": 3.7056,
|
| 1357 |
+
"grad_norm": 0.28425199451328026,
|
| 1358 |
+
"learning_rate": 7.560217063225923e-06,
|
| 1359 |
+
"loss": 0.2815,
|
| 1360 |
+
"step": 193
|
| 1361 |
+
},
|
| 1362 |
+
{
|
| 1363 |
+
"epoch": 3.7248,
|
| 1364 |
+
"grad_norm": 5.801284476847544,
|
| 1365 |
+
"learning_rate": 7.3510924880924575e-06,
|
| 1366 |
+
"loss": 0.2941,
|
| 1367 |
+
"step": 194
|
| 1368 |
+
},
|
| 1369 |
+
{
|
| 1370 |
+
"epoch": 3.7439999999999998,
|
| 1371 |
+
"grad_norm": 0.30776733736644724,
|
| 1372 |
+
"learning_rate": 7.1442478062692135e-06,
|
| 1373 |
+
"loss": 0.3114,
|
| 1374 |
+
"step": 195
|
| 1375 |
+
},
|
| 1376 |
+
{
|
| 1377 |
+
"epoch": 3.7632,
|
| 1378 |
+
"grad_norm": 0.27592085553763634,
|
| 1379 |
+
"learning_rate": 6.939720300328303e-06,
|
| 1380 |
+
"loss": 0.296,
|
| 1381 |
+
"step": 196
|
| 1382 |
+
},
|
| 1383 |
+
{
|
| 1384 |
+
"epoch": 3.7824,
|
| 1385 |
+
"grad_norm": 0.24165223127467064,
|
| 1386 |
+
"learning_rate": 6.737546835184101e-06,
|
| 1387 |
+
"loss": 0.2802,
|
| 1388 |
+
"step": 197
|
| 1389 |
+
},
|
| 1390 |
+
{
|
| 1391 |
+
"epoch": 3.8016,
|
| 1392 |
+
"grad_norm": 0.29879081329577717,
|
| 1393 |
+
"learning_rate": 6.537763851448593e-06,
|
| 1394 |
+
"loss": 0.2863,
|
| 1395 |
+
"step": 198
|
| 1396 |
+
},
|
| 1397 |
+
{
|
| 1398 |
+
"epoch": 3.8208,
|
| 1399 |
+
"grad_norm": 0.2937852483961078,
|
| 1400 |
+
"learning_rate": 6.340407358863167e-06,
|
| 1401 |
+
"loss": 0.3099,
|
| 1402 |
+
"step": 199
|
| 1403 |
+
},
|
| 1404 |
+
{
|
| 1405 |
+
"epoch": 3.84,
|
| 1406 |
+
"grad_norm": 0.24617700051309252,
|
| 1407 |
+
"learning_rate": 6.145512929808013e-06,
|
| 1408 |
+
"loss": 0.2737,
|
| 1409 |
+
"step": 200
|
| 1410 |
+
},
|
| 1411 |
+
{
|
| 1412 |
+
"epoch": 3.8592,
|
| 1413 |
+
"grad_norm": 0.24909560771302497,
|
| 1414 |
+
"learning_rate": 5.95311569289045e-06,
|
| 1415 |
+
"loss": 0.2692,
|
| 1416 |
+
"step": 201
|
| 1417 |
+
},
|
| 1418 |
+
{
|
| 1419 |
+
"epoch": 3.8784,
|
| 1420 |
+
"grad_norm": 0.26304961869462845,
|
| 1421 |
+
"learning_rate": 5.7632503266131925e-06,
|
| 1422 |
+
"loss": 0.2714,
|
| 1423 |
+
"step": 202
|
| 1424 |
+
},
|
| 1425 |
+
{
|
| 1426 |
+
"epoch": 3.8975999999999997,
|
| 1427 |
+
"grad_norm": 0.24326253567701014,
|
| 1428 |
+
"learning_rate": 5.5759510531237116e-06,
|
| 1429 |
+
"loss": 0.2955,
|
| 1430 |
+
"step": 203
|
| 1431 |
+
},
|
| 1432 |
+
{
|
| 1433 |
+
"epoch": 3.9168,
|
| 1434 |
+
"grad_norm": 0.2641176309816874,
|
| 1435 |
+
"learning_rate": 5.39125163204594e-06,
|
| 1436 |
+
"loss": 0.2883,
|
| 1437 |
+
"step": 204
|
| 1438 |
+
},
|
| 1439 |
+
{
|
| 1440 |
+
"epoch": 3.936,
|
| 1441 |
+
"grad_norm": 0.26774132672842993,
|
| 1442 |
+
"learning_rate": 5.209185354395261e-06,
|
| 1443 |
+
"loss": 0.2761,
|
| 1444 |
+
"step": 205
|
| 1445 |
+
},
|
| 1446 |
+
{
|
| 1447 |
+
"epoch": 3.9552,
|
| 1448 |
+
"grad_norm": 0.24278770999257318,
|
| 1449 |
+
"learning_rate": 5.029785036577976e-06,
|
| 1450 |
+
"loss": 0.2927,
|
| 1451 |
+
"step": 206
|
| 1452 |
+
},
|
| 1453 |
+
{
|
| 1454 |
+
"epoch": 3.9744,
|
| 1455 |
+
"grad_norm": 0.23635317757839228,
|
| 1456 |
+
"learning_rate": 4.853083014476403e-06,
|
| 1457 |
+
"loss": 0.2867,
|
| 1458 |
+
"step": 207
|
| 1459 |
+
},
|
| 1460 |
+
{
|
| 1461 |
+
"epoch": 3.9936,
|
| 1462 |
+
"grad_norm": 0.2612939333895405,
|
| 1463 |
+
"learning_rate": 4.679111137620442e-06,
|
| 1464 |
+
"loss": 0.2953,
|
| 1465 |
+
"step": 208
|
| 1466 |
+
},
|
| 1467 |
+
{
|
| 1468 |
+
"epoch": 4.0128,
|
| 1469 |
+
"grad_norm": 0.7973649914801277,
|
| 1470 |
+
"learning_rate": 4.507900763446911e-06,
|
| 1471 |
+
"loss": 0.4388,
|
| 1472 |
+
"step": 209
|
| 1473 |
+
},
|
| 1474 |
+
{
|
| 1475 |
+
"epoch": 4.032,
|
| 1476 |
+
"grad_norm": 0.36905208951857915,
|
| 1477 |
+
"learning_rate": 4.339482751647557e-06,
|
| 1478 |
+
"loss": 0.2546,
|
| 1479 |
+
"step": 210
|
| 1480 |
+
},
|
| 1481 |
+
{
|
| 1482 |
+
"epoch": 4.0512,
|
| 1483 |
+
"grad_norm": 0.30553766630132745,
|
| 1484 |
+
"learning_rate": 4.173887458606767e-06,
|
| 1485 |
+
"loss": 0.2682,
|
| 1486 |
+
"step": 211
|
| 1487 |
+
},
|
| 1488 |
+
{
|
| 1489 |
+
"epoch": 4.0704,
|
| 1490 |
+
"grad_norm": 0.2641701403552184,
|
| 1491 |
+
"learning_rate": 4.011144731929981e-06,
|
| 1492 |
+
"loss": 0.2389,
|
| 1493 |
+
"step": 212
|
| 1494 |
+
},
|
| 1495 |
+
{
|
| 1496 |
+
"epoch": 4.0896,
|
| 1497 |
+
"grad_norm": 0.3435676691721713,
|
| 1498 |
+
"learning_rate": 3.851283905063863e-06,
|
| 1499 |
+
"loss": 0.2501,
|
| 1500 |
+
"step": 213
|
| 1501 |
+
},
|
| 1502 |
+
{
|
| 1503 |
+
"epoch": 4.1088,
|
| 1504 |
+
"grad_norm": 0.3667853839975115,
|
| 1505 |
+
"learning_rate": 3.694333792009115e-06,
|
| 1506 |
+
"loss": 0.2393,
|
| 1507 |
+
"step": 214
|
| 1508 |
+
},
|
| 1509 |
+
{
|
| 1510 |
+
"epoch": 4.128,
|
| 1511 |
+
"grad_norm": 0.32811110252365916,
|
| 1512 |
+
"learning_rate": 3.5403226821268734e-06,
|
| 1513 |
+
"loss": 0.2447,
|
| 1514 |
+
"step": 215
|
| 1515 |
+
},
|
| 1516 |
+
{
|
| 1517 |
+
"epoch": 4.1472,
|
| 1518 |
+
"grad_norm": 0.2862447952562706,
|
| 1519 |
+
"learning_rate": 3.3892783350397675e-06,
|
| 1520 |
+
"loss": 0.2335,
|
| 1521 |
+
"step": 216
|
| 1522 |
+
},
|
| 1523 |
+
{
|
| 1524 |
+
"epoch": 4.1664,
|
| 1525 |
+
"grad_norm": 1.0330289368038168,
|
| 1526 |
+
"learning_rate": 3.241227975628378e-06,
|
| 1527 |
+
"loss": 0.255,
|
| 1528 |
+
"step": 217
|
| 1529 |
+
},
|
| 1530 |
+
{
|
| 1531 |
+
"epoch": 4.1856,
|
| 1532 |
+
"grad_norm": 0.2922923146170167,
|
| 1533 |
+
"learning_rate": 3.0961982891241083e-06,
|
| 1534 |
+
"loss": 0.255,
|
| 1535 |
+
"step": 218
|
| 1536 |
+
},
|
| 1537 |
+
{
|
| 1538 |
+
"epoch": 4.2048,
|
| 1539 |
+
"grad_norm": 0.2916869066279048,
|
| 1540 |
+
"learning_rate": 2.954215416299331e-06,
|
| 1541 |
+
"loss": 0.2471,
|
| 1542 |
+
"step": 219
|
| 1543 |
+
},
|
| 1544 |
+
{
|
| 1545 |
+
"epoch": 4.224,
|
| 1546 |
+
"grad_norm": 0.24996405877435382,
|
| 1547 |
+
"learning_rate": 2.815304948755664e-06,
|
| 1548 |
+
"loss": 0.2252,
|
| 1549 |
+
"step": 220
|
| 1550 |
+
},
|
| 1551 |
+
{
|
| 1552 |
+
"epoch": 4.2432,
|
| 1553 |
+
"grad_norm": 0.267247536941504,
|
| 1554 |
+
"learning_rate": 2.679491924311226e-06,
|
| 1555 |
+
"loss": 0.2389,
|
| 1556 |
+
"step": 221
|
| 1557 |
+
},
|
| 1558 |
+
{
|
| 1559 |
+
"epoch": 4.2624,
|
| 1560 |
+
"grad_norm": 0.2393212636544996,
|
| 1561 |
+
"learning_rate": 2.546800822487714e-06,
|
| 1562 |
+
"loss": 0.2036,
|
| 1563 |
+
"step": 222
|
| 1564 |
+
},
|
| 1565 |
+
{
|
| 1566 |
+
"epoch": 4.2816,
|
| 1567 |
+
"grad_norm": 0.256538641400182,
|
| 1568 |
+
"learning_rate": 2.417255560098111e-06,
|
| 1569 |
+
"loss": 0.2346,
|
| 1570 |
+
"step": 223
|
| 1571 |
+
},
|
| 1572 |
+
{
|
| 1573 |
+
"epoch": 4.3008,
|
| 1574 |
+
"grad_norm": 0.24647477806407794,
|
| 1575 |
+
"learning_rate": 2.2908794869358044e-06,
|
| 1576 |
+
"loss": 0.2403,
|
| 1577 |
+
"step": 224
|
| 1578 |
+
},
|
| 1579 |
+
{
|
| 1580 |
+
"epoch": 4.32,
|
| 1581 |
+
"grad_norm": 0.2513252195848202,
|
| 1582 |
+
"learning_rate": 2.167695381565944e-06,
|
| 1583 |
+
"loss": 0.251,
|
| 1584 |
+
"step": 225
|
| 1585 |
+
},
|
| 1586 |
+
{
|
| 1587 |
+
"epoch": 4.3392,
|
| 1588 |
+
"grad_norm": 0.23573305852654786,
|
| 1589 |
+
"learning_rate": 2.0477254472197237e-06,
|
| 1590 |
+
"loss": 0.2469,
|
| 1591 |
+
"step": 226
|
| 1592 |
+
},
|
| 1593 |
+
{
|
| 1594 |
+
"epoch": 4.3584,
|
| 1595 |
+
"grad_norm": 0.2343709834019784,
|
| 1596 |
+
"learning_rate": 1.930991307792358e-06,
|
| 1597 |
+
"loss": 0.2449,
|
| 1598 |
+
"step": 227
|
| 1599 |
+
},
|
| 1600 |
+
{
|
| 1601 |
+
"epoch": 4.3776,
|
| 1602 |
+
"grad_norm": 0.23524387147763087,
|
| 1603 |
+
"learning_rate": 1.817514003945524e-06,
|
| 1604 |
+
"loss": 0.223,
|
| 1605 |
+
"step": 228
|
| 1606 |
+
},
|
| 1607 |
+
{
|
| 1608 |
+
"epoch": 4.3968,
|
| 1609 |
+
"grad_norm": 0.2630623239373731,
|
| 1610 |
+
"learning_rate": 1.7073139893149092e-06,
|
| 1611 |
+
"loss": 0.2797,
|
| 1612 |
+
"step": 229
|
| 1613 |
+
},
|
| 1614 |
+
{
|
| 1615 |
+
"epoch": 4.416,
|
| 1616 |
+
"grad_norm": 0.21357907994461814,
|
| 1617 |
+
"learning_rate": 1.6004111268235156e-06,
|
| 1618 |
+
"loss": 0.2254,
|
| 1619 |
+
"step": 230
|
| 1620 |
+
},
|
| 1621 |
+
{
|
| 1622 |
+
"epoch": 4.4352,
|
| 1623 |
+
"grad_norm": 0.24114367628100736,
|
| 1624 |
+
"learning_rate": 1.49682468510153e-06,
|
| 1625 |
+
"loss": 0.2532,
|
| 1626 |
+
"step": 231
|
| 1627 |
+
},
|
| 1628 |
+
{
|
| 1629 |
+
"epoch": 4.4544,
|
| 1630 |
+
"grad_norm": 0.21661732233231876,
|
| 1631 |
+
"learning_rate": 1.396573335013236e-06,
|
| 1632 |
+
"loss": 0.2157,
|
| 1633 |
+
"step": 232
|
| 1634 |
+
},
|
| 1635 |
+
{
|
| 1636 |
+
"epoch": 4.4736,
|
| 1637 |
+
"grad_norm": 0.23150889229425836,
|
| 1638 |
+
"learning_rate": 1.2996751462917057e-06,
|
| 1639 |
+
"loss": 0.2566,
|
| 1640 |
+
"step": 233
|
| 1641 |
+
},
|
| 1642 |
+
{
|
| 1643 |
+
"epoch": 4.4928,
|
| 1644 |
+
"grad_norm": 0.2379212292063147,
|
| 1645 |
+
"learning_rate": 1.2061475842818337e-06,
|
| 1646 |
+
"loss": 0.2367,
|
| 1647 |
+
"step": 234
|
| 1648 |
+
},
|
| 1649 |
+
{
|
| 1650 |
+
"epoch": 4.5120000000000005,
|
| 1651 |
+
"grad_norm": 0.22007544548694274,
|
| 1652 |
+
"learning_rate": 1.1160075067923092e-06,
|
| 1653 |
+
"loss": 0.2324,
|
| 1654 |
+
"step": 235
|
| 1655 |
+
},
|
| 1656 |
+
{
|
| 1657 |
+
"epoch": 4.5312,
|
| 1658 |
+
"grad_norm": 0.21549828002229496,
|
| 1659 |
+
"learning_rate": 1.0292711610570904e-06,
|
| 1660 |
+
"loss": 0.2231,
|
| 1661 |
+
"step": 236
|
| 1662 |
+
},
|
| 1663 |
+
{
|
| 1664 |
+
"epoch": 4.5504,
|
| 1665 |
+
"grad_norm": 0.23118095418384038,
|
| 1666 |
+
"learning_rate": 9.459541808069295e-07,
|
| 1667 |
+
"loss": 0.2477,
|
| 1668 |
+
"step": 237
|
| 1669 |
+
},
|
| 1670 |
+
{
|
| 1671 |
+
"epoch": 4.5696,
|
| 1672 |
+
"grad_norm": 0.21690542858450784,
|
| 1673 |
+
"learning_rate": 8.660715834514977e-07,
|
| 1674 |
+
"loss": 0.2235,
|
| 1675 |
+
"step": 238
|
| 1676 |
+
},
|
| 1677 |
+
{
|
| 1678 |
+
"epoch": 4.5888,
|
| 1679 |
+
"grad_norm": 0.21555826489685279,
|
| 1680 |
+
"learning_rate": 7.896377673725553e-07,
|
| 1681 |
+
"loss": 0.2245,
|
| 1682 |
+
"step": 239
|
| 1683 |
+
},
|
| 1684 |
+
{
|
| 1685 |
+
"epoch": 4.608,
|
| 1686 |
+
"grad_norm": 0.22910791259612318,
|
| 1687 |
+
"learning_rate": 7.166665093287539e-07,
|
| 1688 |
+
"loss": 0.2288,
|
| 1689 |
+
"step": 240
|
| 1690 |
+
},
|
| 1691 |
+
{
|
| 1692 |
+
"epoch": 4.6272,
|
| 1693 |
+
"grad_norm": 0.23348263601941588,
|
| 1694 |
+
"learning_rate": 6.471709619724386e-07,
|
| 1695 |
+
"loss": 0.2575,
|
| 1696 |
+
"step": 241
|
| 1697 |
+
},
|
| 1698 |
+
{
|
| 1699 |
+
"epoch": 4.6464,
|
| 1700 |
+
"grad_norm": 0.21483845228914805,
|
| 1701 |
+
"learning_rate": 5.811636514789598e-07,
|
| 1702 |
+
"loss": 0.2354,
|
| 1703 |
+
"step": 242
|
| 1704 |
+
},
|
| 1705 |
+
{
|
| 1706 |
+
"epoch": 4.6655999999999995,
|
| 1707 |
+
"grad_norm": 0.20939600576946568,
|
| 1708 |
+
"learning_rate": 5.186564752889078e-07,
|
| 1709 |
+
"loss": 0.2353,
|
| 1710 |
+
"step": 243
|
| 1711 |
+
},
|
| 1712 |
+
{
|
| 1713 |
+
"epoch": 4.6848,
|
| 1714 |
+
"grad_norm": 0.21575112450307377,
|
| 1715 |
+
"learning_rate": 4.5966069996365993e-07,
|
| 1716 |
+
"loss": 0.2326,
|
| 1717 |
+
"step": 244
|
| 1718 |
+
},
|
| 1719 |
+
{
|
| 1720 |
+
"epoch": 4.704,
|
| 1721 |
+
"grad_norm": 0.21964522917124985,
|
| 1722 |
+
"learning_rate": 4.0418695915464657e-07,
|
| 1723 |
+
"loss": 0.2087,
|
| 1724 |
+
"step": 245
|
| 1725 |
+
},
|
| 1726 |
+
{
|
| 1727 |
+
"epoch": 4.7232,
|
| 1728 |
+
"grad_norm": 0.21691011006961272,
|
| 1729 |
+
"learning_rate": 3.522452516867048e-07,
|
| 1730 |
+
"loss": 0.2451,
|
| 1731 |
+
"step": 246
|
| 1732 |
+
},
|
| 1733 |
+
{
|
| 1734 |
+
"epoch": 4.7424,
|
| 1735 |
+
"grad_norm": 0.20917292131866752,
|
| 1736 |
+
"learning_rate": 3.038449397558396e-07,
|
| 1737 |
+
"loss": 0.2459,
|
| 1738 |
+
"step": 247
|
| 1739 |
+
},
|
| 1740 |
+
{
|
| 1741 |
+
"epoch": 4.7616,
|
| 1742 |
+
"grad_norm": 0.2236515502703497,
|
| 1743 |
+
"learning_rate": 2.5899474724174313e-07,
|
| 1744 |
+
"loss": 0.2416,
|
| 1745 |
+
"step": 248
|
| 1746 |
+
},
|
| 1747 |
+
{
|
| 1748 |
+
"epoch": 4.7808,
|
| 1749 |
+
"grad_norm": 0.23000987027877715,
|
| 1750 |
+
"learning_rate": 2.1770275813536746e-07,
|
| 1751 |
+
"loss": 0.2367,
|
| 1752 |
+
"step": 249
|
| 1753 |
+
},
|
| 1754 |
+
{
|
| 1755 |
+
"epoch": 4.8,
|
| 1756 |
+
"grad_norm": 0.2174658218472906,
|
| 1757 |
+
"learning_rate": 1.799764150818306e-07,
|
| 1758 |
+
"loss": 0.2471,
|
| 1759 |
+
"step": 250
|
| 1760 |
+
},
|
| 1761 |
+
{
|
| 1762 |
+
"epoch": 4.8192,
|
| 1763 |
+
"grad_norm": 0.20869570837977713,
|
| 1764 |
+
"learning_rate": 1.4582251803892055e-07,
|
| 1765 |
+
"loss": 0.208,
|
| 1766 |
+
"step": 251
|
| 1767 |
+
},
|
| 1768 |
+
{
|
| 1769 |
+
"epoch": 4.8384,
|
| 1770 |
+
"grad_norm": 0.20843555819911957,
|
| 1771 |
+
"learning_rate": 1.1524722305144231e-07,
|
| 1772 |
+
"loss": 0.2358,
|
| 1773 |
+
"step": 252
|
| 1774 |
+
},
|
| 1775 |
+
{
|
| 1776 |
+
"epoch": 4.8576,
|
| 1777 |
+
"grad_norm": 0.21189978049987293,
|
| 1778 |
+
"learning_rate": 8.825604114162556e-08,
|
| 1779 |
+
"loss": 0.2384,
|
| 1780 |
+
"step": 253
|
| 1781 |
+
},
|
| 1782 |
+
{
|
| 1783 |
+
"epoch": 4.8768,
|
| 1784 |
+
"grad_norm": 0.20060464814134882,
|
| 1785 |
+
"learning_rate": 6.485383731580142e-08,
|
| 1786 |
+
"loss": 0.2169,
|
| 1787 |
+
"step": 254
|
| 1788 |
+
},
|
| 1789 |
+
{
|
| 1790 |
+
"epoch": 4.896,
|
| 1791 |
+
"grad_norm": 0.2140648074010658,
|
| 1792 |
+
"learning_rate": 4.5044829687497236e-08,
|
| 1793 |
+
"loss": 0.2407,
|
| 1794 |
+
"step": 255
|
| 1795 |
+
},
|
| 1796 |
+
{
|
| 1797 |
+
"epoch": 4.9152000000000005,
|
| 1798 |
+
"grad_norm": 0.21391107187688388,
|
| 1799 |
+
"learning_rate": 2.8832588717164766e-08,
|
| 1800 |
+
"loss": 0.2365,
|
| 1801 |
+
"step": 256
|
| 1802 |
+
},
|
| 1803 |
+
{
|
| 1804 |
+
"epoch": 4.9344,
|
| 1805 |
+
"grad_norm": 0.20631520508588339,
|
| 1806 |
+
"learning_rate": 1.6220036568608265e-08,
|
| 1807 |
+
"loss": 0.2311,
|
| 1808 |
+
"step": 257
|
| 1809 |
+
},
|
| 1810 |
+
{
|
| 1811 |
+
"epoch": 4.9536,
|
| 1812 |
+
"grad_norm": 0.21724546755525317,
|
| 1813 |
+
"learning_rate": 7.209446582292501e-09,
|
| 1814 |
+
"loss": 0.2428,
|
| 1815 |
+
"step": 258
|
| 1816 |
+
},
|
| 1817 |
+
{
|
| 1818 |
+
"epoch": 4.9728,
|
| 1819 |
+
"grad_norm": 0.20895998801297003,
|
| 1820 |
+
"learning_rate": 1.8024428655794012e-09,
|
| 1821 |
+
"loss": 0.2479,
|
| 1822 |
+
"step": 259
|
| 1823 |
+
},
|
| 1824 |
+
{
|
| 1825 |
+
"epoch": 4.992,
|
| 1826 |
+
"grad_norm": 0.21787569434125945,
|
| 1827 |
+
"learning_rate": 0.0,
|
| 1828 |
+
"loss": 0.2504,
|
| 1829 |
+
"step": 260
|
| 1830 |
+
},
|
| 1831 |
+
{
|
| 1832 |
+
"epoch": 4.992,
|
| 1833 |
+
"step": 260,
|
| 1834 |
+
"total_flos": 1.013794005747499e+18,
|
| 1835 |
+
"train_loss": 0.39605270406374565,
|
| 1836 |
+
"train_runtime": 28764.0375,
|
| 1837 |
+
"train_samples_per_second": 1.738,
|
| 1838 |
+
"train_steps_per_second": 0.009
|
| 1839 |
+
}
|
| 1840 |
+
],
|
| 1841 |
+
"logging_steps": 1.0,
|
| 1842 |
+
"max_steps": 260,
|
| 1843 |
+
"num_input_tokens_seen": 0,
|
| 1844 |
+
"num_train_epochs": 5,
|
| 1845 |
+
"save_steps": 500,
|
| 1846 |
+
"stateful_callbacks": {
|
| 1847 |
+
"TrainerControl": {
|
| 1848 |
+
"args": {
|
| 1849 |
+
"should_epoch_stop": false,
|
| 1850 |
+
"should_evaluate": false,
|
| 1851 |
+
"should_log": false,
|
| 1852 |
+
"should_save": true,
|
| 1853 |
+
"should_training_stop": true
|
| 1854 |
+
},
|
| 1855 |
+
"attributes": {}
|
| 1856 |
+
}
|
| 1857 |
+
},
|
| 1858 |
+
"total_flos": 1.013794005747499e+18,
|
| 1859 |
+
"train_batch_size": 1,
|
| 1860 |
+
"trial_name": null,
|
| 1861 |
+
"trial_params": null
|
| 1862 |
+
}
|
training_loss.png
ADDED
|