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_lr2e5_epochs5
|
|
@@ -15,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 15 |
|
| 16 |
# 10k_globalbatchsize192_lr2e5_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_lr2e5_epochs5
|
|
|
|
| 16 |
|
| 17 |
# 10k_globalbatchsize192_lr2e5_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": 0.45938570224321806,
|
| 5 |
+
"train_runtime": 28676.658,
|
| 6 |
+
"train_samples_per_second": 1.743,
|
| 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": 0.45938570224321806,
|
| 5 |
+
"train_runtime": 28676.658,
|
| 6 |
+
"train_samples_per_second": 1.743,
|
| 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": 5.846937880044676,
|
| 14 |
+
"learning_rate": 7.692307692307694e-07,
|
| 15 |
+
"loss": 0.8597,
|
| 16 |
+
"step": 1
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.0384,
|
| 20 |
+
"grad_norm": 5.931659414307684,
|
| 21 |
+
"learning_rate": 1.5384615384615387e-06,
|
| 22 |
+
"loss": 0.8763,
|
| 23 |
+
"step": 2
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.0576,
|
| 27 |
+
"grad_norm": 5.746179244995965,
|
| 28 |
+
"learning_rate": 2.307692307692308e-06,
|
| 29 |
+
"loss": 0.8566,
|
| 30 |
+
"step": 3
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.0768,
|
| 34 |
+
"grad_norm": 5.369485667884124,
|
| 35 |
+
"learning_rate": 3.0769230769230774e-06,
|
| 36 |
+
"loss": 0.8251,
|
| 37 |
+
"step": 4
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.096,
|
| 41 |
+
"grad_norm": 4.410590931795538,
|
| 42 |
+
"learning_rate": 3.846153846153847e-06,
|
| 43 |
+
"loss": 0.8395,
|
| 44 |
+
"step": 5
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.1152,
|
| 48 |
+
"grad_norm": 2.7310819010168697,
|
| 49 |
+
"learning_rate": 4.615384615384616e-06,
|
| 50 |
+
"loss": 0.787,
|
| 51 |
+
"step": 6
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.1344,
|
| 55 |
+
"grad_norm": 2.1515635314288937,
|
| 56 |
+
"learning_rate": 5.384615384615385e-06,
|
| 57 |
+
"loss": 0.7624,
|
| 58 |
+
"step": 7
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.1536,
|
| 62 |
+
"grad_norm": 3.3989509860999205,
|
| 63 |
+
"learning_rate": 6.153846153846155e-06,
|
| 64 |
+
"loss": 0.7759,
|
| 65 |
+
"step": 8
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.1728,
|
| 69 |
+
"grad_norm": 4.034792205908373,
|
| 70 |
+
"learning_rate": 6.923076923076923e-06,
|
| 71 |
+
"loss": 0.7899,
|
| 72 |
+
"step": 9
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.192,
|
| 76 |
+
"grad_norm": 3.907817772672124,
|
| 77 |
+
"learning_rate": 7.692307692307694e-06,
|
| 78 |
+
"loss": 0.7632,
|
| 79 |
+
"step": 10
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.2112,
|
| 83 |
+
"grad_norm": 3.470870669030584,
|
| 84 |
+
"learning_rate": 8.461538461538462e-06,
|
| 85 |
+
"loss": 0.724,
|
| 86 |
+
"step": 11
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.2304,
|
| 90 |
+
"grad_norm": 3.1402907073737953,
|
| 91 |
+
"learning_rate": 9.230769230769232e-06,
|
| 92 |
+
"loss": 0.7111,
|
| 93 |
+
"step": 12
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.2496,
|
| 97 |
+
"grad_norm": 2.1731006224629033,
|
| 98 |
+
"learning_rate": 1e-05,
|
| 99 |
+
"loss": 0.6717,
|
| 100 |
+
"step": 13
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.2688,
|
| 104 |
+
"grad_norm": 1.5326151017825498,
|
| 105 |
+
"learning_rate": 1.076923076923077e-05,
|
| 106 |
+
"loss": 0.6485,
|
| 107 |
+
"step": 14
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.288,
|
| 111 |
+
"grad_norm": 1.7662820057428061,
|
| 112 |
+
"learning_rate": 1.1538461538461538e-05,
|
| 113 |
+
"loss": 0.662,
|
| 114 |
+
"step": 15
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.3072,
|
| 118 |
+
"grad_norm": 1.795523541787461,
|
| 119 |
+
"learning_rate": 1.230769230769231e-05,
|
| 120 |
+
"loss": 0.6578,
|
| 121 |
+
"step": 16
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.3264,
|
| 125 |
+
"grad_norm": 1.2466042393289243,
|
| 126 |
+
"learning_rate": 1.3076923076923078e-05,
|
| 127 |
+
"loss": 0.6433,
|
| 128 |
+
"step": 17
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.3456,
|
| 132 |
+
"grad_norm": 1.1989614953646517,
|
| 133 |
+
"learning_rate": 1.3846153846153847e-05,
|
| 134 |
+
"loss": 0.6281,
|
| 135 |
+
"step": 18
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.3648,
|
| 139 |
+
"grad_norm": 1.1222706184557578,
|
| 140 |
+
"learning_rate": 1.4615384615384615e-05,
|
| 141 |
+
"loss": 0.6341,
|
| 142 |
+
"step": 19
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.384,
|
| 146 |
+
"grad_norm": 0.860111856911054,
|
| 147 |
+
"learning_rate": 1.5384615384615387e-05,
|
| 148 |
+
"loss": 0.5856,
|
| 149 |
+
"step": 20
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.4032,
|
| 153 |
+
"grad_norm": 1.0918732757957865,
|
| 154 |
+
"learning_rate": 1.6153846153846154e-05,
|
| 155 |
+
"loss": 0.6245,
|
| 156 |
+
"step": 21
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.4224,
|
| 160 |
+
"grad_norm": 0.9053521122130429,
|
| 161 |
+
"learning_rate": 1.6923076923076924e-05,
|
| 162 |
+
"loss": 0.6087,
|
| 163 |
+
"step": 22
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.4416,
|
| 167 |
+
"grad_norm": 0.8985496052053772,
|
| 168 |
+
"learning_rate": 1.7692307692307694e-05,
|
| 169 |
+
"loss": 0.6063,
|
| 170 |
+
"step": 23
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.4608,
|
| 174 |
+
"grad_norm": 0.8628626200658147,
|
| 175 |
+
"learning_rate": 1.8461538461538465e-05,
|
| 176 |
+
"loss": 0.6189,
|
| 177 |
+
"step": 24
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.48,
|
| 181 |
+
"grad_norm": 0.7211709915468527,
|
| 182 |
+
"learning_rate": 1.923076923076923e-05,
|
| 183 |
+
"loss": 0.6001,
|
| 184 |
+
"step": 25
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.4992,
|
| 188 |
+
"grad_norm": 0.872604906635497,
|
| 189 |
+
"learning_rate": 2e-05,
|
| 190 |
+
"loss": 0.5884,
|
| 191 |
+
"step": 26
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.5184,
|
| 195 |
+
"grad_norm": 0.6786157993953024,
|
| 196 |
+
"learning_rate": 1.999909877856721e-05,
|
| 197 |
+
"loss": 0.5634,
|
| 198 |
+
"step": 27
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.5376,
|
| 202 |
+
"grad_norm": 0.7781689321266382,
|
| 203 |
+
"learning_rate": 1.9996395276708856e-05,
|
| 204 |
+
"loss": 0.5638,
|
| 205 |
+
"step": 28
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.5568,
|
| 209 |
+
"grad_norm": 0.674895896297859,
|
| 210 |
+
"learning_rate": 1.9991889981715696e-05,
|
| 211 |
+
"loss": 0.5834,
|
| 212 |
+
"step": 29
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.576,
|
| 216 |
+
"grad_norm": 0.6058222670625754,
|
| 217 |
+
"learning_rate": 1.9985583705641418e-05,
|
| 218 |
+
"loss": 0.5603,
|
| 219 |
+
"step": 30
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.5952,
|
| 223 |
+
"grad_norm": 0.636543754341517,
|
| 224 |
+
"learning_rate": 1.9977477585156252e-05,
|
| 225 |
+
"loss": 0.5444,
|
| 226 |
+
"step": 31
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.6144,
|
| 230 |
+
"grad_norm": 0.6346830226930983,
|
| 231 |
+
"learning_rate": 1.9967573081342103e-05,
|
| 232 |
+
"loss": 0.5372,
|
| 233 |
+
"step": 32
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.6336,
|
| 237 |
+
"grad_norm": 0.582179976007427,
|
| 238 |
+
"learning_rate": 1.9955871979429188e-05,
|
| 239 |
+
"loss": 0.5604,
|
| 240 |
+
"step": 33
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.6528,
|
| 244 |
+
"grad_norm": 0.5918951226210546,
|
| 245 |
+
"learning_rate": 1.9942376388474282e-05,
|
| 246 |
+
"loss": 0.5522,
|
| 247 |
+
"step": 34
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.672,
|
| 251 |
+
"grad_norm": 0.8175533184760783,
|
| 252 |
+
"learning_rate": 1.992708874098054e-05,
|
| 253 |
+
"loss": 0.5458,
|
| 254 |
+
"step": 35
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.6912,
|
| 258 |
+
"grad_norm": 0.46598853743982027,
|
| 259 |
+
"learning_rate": 1.9910011792459086e-05,
|
| 260 |
+
"loss": 0.5529,
|
| 261 |
+
"step": 36
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.7104,
|
| 265 |
+
"grad_norm": 0.8069415543739152,
|
| 266 |
+
"learning_rate": 1.989114862093232e-05,
|
| 267 |
+
"loss": 0.5692,
|
| 268 |
+
"step": 37
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.7296,
|
| 272 |
+
"grad_norm": 0.5302114192267839,
|
| 273 |
+
"learning_rate": 1.9870502626379127e-05,
|
| 274 |
+
"loss": 0.5415,
|
| 275 |
+
"step": 38
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.7488,
|
| 279 |
+
"grad_norm": 0.6881031584281583,
|
| 280 |
+
"learning_rate": 1.9848077530122083e-05,
|
| 281 |
+
"loss": 0.549,
|
| 282 |
+
"step": 39
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.768,
|
| 286 |
+
"grad_norm": 0.5187653158582829,
|
| 287 |
+
"learning_rate": 1.9823877374156647e-05,
|
| 288 |
+
"loss": 0.5362,
|
| 289 |
+
"step": 40
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.7872,
|
| 293 |
+
"grad_norm": 0.41585341807745396,
|
| 294 |
+
"learning_rate": 1.979790652042268e-05,
|
| 295 |
+
"loss": 0.5478,
|
| 296 |
+
"step": 41
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.8064,
|
| 300 |
+
"grad_norm": 0.5012261322596314,
|
| 301 |
+
"learning_rate": 1.977016965001817e-05,
|
| 302 |
+
"loss": 0.5409,
|
| 303 |
+
"step": 42
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.8256,
|
| 307 |
+
"grad_norm": 0.4590613765647621,
|
| 308 |
+
"learning_rate": 1.9740671762355548e-05,
|
| 309 |
+
"loss": 0.5297,
|
| 310 |
+
"step": 43
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.8448,
|
| 314 |
+
"grad_norm": 0.48130826973555896,
|
| 315 |
+
"learning_rate": 1.9709418174260523e-05,
|
| 316 |
+
"loss": 0.5483,
|
| 317 |
+
"step": 44
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.864,
|
| 321 |
+
"grad_norm": 0.5860586957089473,
|
| 322 |
+
"learning_rate": 1.9676414519013782e-05,
|
| 323 |
+
"loss": 0.5225,
|
| 324 |
+
"step": 45
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.8832,
|
| 328 |
+
"grad_norm": 0.6267782085363393,
|
| 329 |
+
"learning_rate": 1.9641666745335626e-05,
|
| 330 |
+
"loss": 0.5434,
|
| 331 |
+
"step": 46
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.9024,
|
| 335 |
+
"grad_norm": 0.6722375547807576,
|
| 336 |
+
"learning_rate": 1.9605181116313725e-05,
|
| 337 |
+
"loss": 0.5378,
|
| 338 |
+
"step": 47
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.9216,
|
| 342 |
+
"grad_norm": 0.5923333515035274,
|
| 343 |
+
"learning_rate": 1.9566964208274254e-05,
|
| 344 |
+
"loss": 0.5216,
|
| 345 |
+
"step": 48
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.9408,
|
| 349 |
+
"grad_norm": 0.4197906838618975,
|
| 350 |
+
"learning_rate": 1.9527022909596537e-05,
|
| 351 |
+
"loss": 0.5189,
|
| 352 |
+
"step": 49
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.96,
|
| 356 |
+
"grad_norm": 0.5842990870167546,
|
| 357 |
+
"learning_rate": 1.9485364419471454e-05,
|
| 358 |
+
"loss": 0.5389,
|
| 359 |
+
"step": 50
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.9792,
|
| 363 |
+
"grad_norm": 0.4788736266707313,
|
| 364 |
+
"learning_rate": 1.9441996246603848e-05,
|
| 365 |
+
"loss": 0.5121,
|
| 366 |
+
"step": 51
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.9984,
|
| 370 |
+
"grad_norm": 0.528762383097365,
|
| 371 |
+
"learning_rate": 1.9396926207859085e-05,
|
| 372 |
+
"loss": 0.5328,
|
| 373 |
+
"step": 52
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 1.0176,
|
| 377 |
+
"grad_norm": 1.323838799774278,
|
| 378 |
+
"learning_rate": 1.9350162426854152e-05,
|
| 379 |
+
"loss": 1.0087,
|
| 380 |
+
"step": 53
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 1.0368,
|
| 384 |
+
"grad_norm": 0.8199572730249561,
|
| 385 |
+
"learning_rate": 1.9301713332493386e-05,
|
| 386 |
+
"loss": 0.4953,
|
| 387 |
+
"step": 54
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 1.056,
|
| 391 |
+
"grad_norm": 0.7257224873865245,
|
| 392 |
+
"learning_rate": 1.925158765744924e-05,
|
| 393 |
+
"loss": 0.4954,
|
| 394 |
+
"step": 55
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 1.0752,
|
| 398 |
+
"grad_norm": 0.6247086914341693,
|
| 399 |
+
"learning_rate": 1.9199794436588244e-05,
|
| 400 |
+
"loss": 0.4647,
|
| 401 |
+
"step": 56
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 1.0944,
|
| 405 |
+
"grad_norm": 0.8518556393259077,
|
| 406 |
+
"learning_rate": 1.9146343005342546e-05,
|
| 407 |
+
"loss": 0.5076,
|
| 408 |
+
"step": 57
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 1.1136,
|
| 412 |
+
"grad_norm": 0.5633237693853321,
|
| 413 |
+
"learning_rate": 1.909124299802724e-05,
|
| 414 |
+
"loss": 0.4897,
|
| 415 |
+
"step": 58
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 1.1328,
|
| 419 |
+
"grad_norm": 0.5333731297850538,
|
| 420 |
+
"learning_rate": 1.9034504346103825e-05,
|
| 421 |
+
"loss": 0.4741,
|
| 422 |
+
"step": 59
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 1.152,
|
| 426 |
+
"grad_norm": 0.6392263927216528,
|
| 427 |
+
"learning_rate": 1.8976137276390145e-05,
|
| 428 |
+
"loss": 0.4919,
|
| 429 |
+
"step": 60
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 1.1712,
|
| 433 |
+
"grad_norm": 0.47603767476389836,
|
| 434 |
+
"learning_rate": 1.891615230921703e-05,
|
| 435 |
+
"loss": 0.4821,
|
| 436 |
+
"step": 61
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 1.1904,
|
| 440 |
+
"grad_norm": 0.6311303310972034,
|
| 441 |
+
"learning_rate": 1.8854560256532098e-05,
|
| 442 |
+
"loss": 0.4956,
|
| 443 |
+
"step": 62
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 1.2096,
|
| 447 |
+
"grad_norm": 0.5601141015421137,
|
| 448 |
+
"learning_rate": 1.879137221995095e-05,
|
| 449 |
+
"loss": 0.5015,
|
| 450 |
+
"step": 63
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 1.2288000000000001,
|
| 454 |
+
"grad_norm": 0.5455236127294453,
|
| 455 |
+
"learning_rate": 1.8726599588756144e-05,
|
| 456 |
+
"loss": 0.5027,
|
| 457 |
+
"step": 64
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 1.248,
|
| 461 |
+
"grad_norm": 0.5462827376903222,
|
| 462 |
+
"learning_rate": 1.866025403784439e-05,
|
| 463 |
+
"loss": 0.4988,
|
| 464 |
+
"step": 65
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 1.2671999999999999,
|
| 468 |
+
"grad_norm": 0.44863862709342234,
|
| 469 |
+
"learning_rate": 1.859234752562217e-05,
|
| 470 |
+
"loss": 0.4989,
|
| 471 |
+
"step": 66
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 1.2864,
|
| 475 |
+
"grad_norm": 0.580784132822074,
|
| 476 |
+
"learning_rate": 1.8522892291850335e-05,
|
| 477 |
+
"loss": 0.5044,
|
| 478 |
+
"step": 67
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 1.3056,
|
| 482 |
+
"grad_norm": 0.46331820609096697,
|
| 483 |
+
"learning_rate": 1.845190085543795e-05,
|
| 484 |
+
"loss": 0.4917,
|
| 485 |
+
"step": 68
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 1.3248,
|
| 489 |
+
"grad_norm": 0.5395563609012295,
|
| 490 |
+
"learning_rate": 1.8379386012185813e-05,
|
| 491 |
+
"loss": 0.4871,
|
| 492 |
+
"step": 69
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 1.3439999999999999,
|
| 496 |
+
"grad_norm": 0.3788766704105627,
|
| 497 |
+
"learning_rate": 1.8305360832480118e-05,
|
| 498 |
+
"loss": 0.4879,
|
| 499 |
+
"step": 70
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 1.3632,
|
| 503 |
+
"grad_norm": 0.6424447426894108,
|
| 504 |
+
"learning_rate": 1.8229838658936566e-05,
|
| 505 |
+
"loss": 0.504,
|
| 506 |
+
"step": 71
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 1.3824,
|
| 510 |
+
"grad_norm": 0.3967766552277642,
|
| 511 |
+
"learning_rate": 1.8152833103995443e-05,
|
| 512 |
+
"loss": 0.4689,
|
| 513 |
+
"step": 72
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 1.4016,
|
| 517 |
+
"grad_norm": 0.5303038940270486,
|
| 518 |
+
"learning_rate": 1.807435804746807e-05,
|
| 519 |
+
"loss": 0.4773,
|
| 520 |
+
"step": 73
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 1.4208,
|
| 524 |
+
"grad_norm": 0.4448757461924023,
|
| 525 |
+
"learning_rate": 1.7994427634035016e-05,
|
| 526 |
+
"loss": 0.4896,
|
| 527 |
+
"step": 74
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 1.44,
|
| 531 |
+
"grad_norm": 0.5154783916174432,
|
| 532 |
+
"learning_rate": 1.791305627069662e-05,
|
| 533 |
+
"loss": 0.4923,
|
| 534 |
+
"step": 75
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 1.4592,
|
| 538 |
+
"grad_norm": 0.458113659714897,
|
| 539 |
+
"learning_rate": 1.7830258624176224e-05,
|
| 540 |
+
"loss": 0.4667,
|
| 541 |
+
"step": 76
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 1.4784,
|
| 545 |
+
"grad_norm": 0.4356408968867487,
|
| 546 |
+
"learning_rate": 1.7746049618276545e-05,
|
| 547 |
+
"loss": 0.4541,
|
| 548 |
+
"step": 77
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 1.4976,
|
| 552 |
+
"grad_norm": 0.5681525567691887,
|
| 553 |
+
"learning_rate": 1.766044443118978e-05,
|
| 554 |
+
"loss": 0.4922,
|
| 555 |
+
"step": 78
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 1.5168,
|
| 559 |
+
"grad_norm": 0.4005629875464663,
|
| 560 |
+
"learning_rate": 1.7573458492761802e-05,
|
| 561 |
+
"loss": 0.468,
|
| 562 |
+
"step": 79
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 1.536,
|
| 566 |
+
"grad_norm": 0.46717196382869475,
|
| 567 |
+
"learning_rate": 1.7485107481711014e-05,
|
| 568 |
+
"loss": 0.4567,
|
| 569 |
+
"step": 80
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 1.5552000000000001,
|
| 573 |
+
"grad_norm": 0.4467898707928224,
|
| 574 |
+
"learning_rate": 1.7395407322802374e-05,
|
| 575 |
+
"loss": 0.4671,
|
| 576 |
+
"step": 81
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 1.5744,
|
| 580 |
+
"grad_norm": 0.39293767611983454,
|
| 581 |
+
"learning_rate": 1.7304374183977032e-05,
|
| 582 |
+
"loss": 0.4718,
|
| 583 |
+
"step": 82
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 1.5936,
|
| 587 |
+
"grad_norm": 0.415240105051738,
|
| 588 |
+
"learning_rate": 1.7212024473438145e-05,
|
| 589 |
+
"loss": 0.4746,
|
| 590 |
+
"step": 83
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 1.6128,
|
| 594 |
+
"grad_norm": 0.3764419656074339,
|
| 595 |
+
"learning_rate": 1.7118374836693407e-05,
|
| 596 |
+
"loss": 0.4817,
|
| 597 |
+
"step": 84
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 1.6320000000000001,
|
| 601 |
+
"grad_norm": 0.36398721604761547,
|
| 602 |
+
"learning_rate": 1.7023442153554776e-05,
|
| 603 |
+
"loss": 0.4687,
|
| 604 |
+
"step": 85
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 1.6512,
|
| 608 |
+
"grad_norm": 0.42047910574520564,
|
| 609 |
+
"learning_rate": 1.6927243535095995e-05,
|
| 610 |
+
"loss": 0.4823,
|
| 611 |
+
"step": 86
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 1.6703999999999999,
|
| 615 |
+
"grad_norm": 0.3685182579381418,
|
| 616 |
+
"learning_rate": 1.6829796320568416e-05,
|
| 617 |
+
"loss": 0.4672,
|
| 618 |
+
"step": 87
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 1.6896,
|
| 622 |
+
"grad_norm": 0.3927315022309727,
|
| 623 |
+
"learning_rate": 1.67311180742757e-05,
|
| 624 |
+
"loss": 0.4697,
|
| 625 |
+
"step": 88
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 1.7088,
|
| 629 |
+
"grad_norm": 0.39958088629722804,
|
| 630 |
+
"learning_rate": 1.6631226582407954e-05,
|
| 631 |
+
"loss": 0.4901,
|
| 632 |
+
"step": 89
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 1.728,
|
| 636 |
+
"grad_norm": 0.4077663143906098,
|
| 637 |
+
"learning_rate": 1.653013984983585e-05,
|
| 638 |
+
"loss": 0.4658,
|
| 639 |
+
"step": 90
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 1.7471999999999999,
|
| 643 |
+
"grad_norm": 0.4764370302897352,
|
| 644 |
+
"learning_rate": 1.6427876096865394e-05,
|
| 645 |
+
"loss": 0.501,
|
| 646 |
+
"step": 91
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 1.7664,
|
| 650 |
+
"grad_norm": 0.4538599957502835,
|
| 651 |
+
"learning_rate": 1.6324453755953772e-05,
|
| 652 |
+
"loss": 0.4824,
|
| 653 |
+
"step": 92
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 1.7856,
|
| 657 |
+
"grad_norm": 0.4147232946184169,
|
| 658 |
+
"learning_rate": 1.621989146838704e-05,
|
| 659 |
+
"loss": 0.4709,
|
| 660 |
+
"step": 93
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 1.8048,
|
| 664 |
+
"grad_norm": 0.4151599994292114,
|
| 665 |
+
"learning_rate": 1.6114208080920125e-05,
|
| 666 |
+
"loss": 0.4487,
|
| 667 |
+
"step": 94
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 1.8239999999999998,
|
| 671 |
+
"grad_norm": 0.453059578263153,
|
| 672 |
+
"learning_rate": 1.600742264237979e-05,
|
| 673 |
+
"loss": 0.4712,
|
| 674 |
+
"step": 95
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 1.8432,
|
| 678 |
+
"grad_norm": 0.4669464128087692,
|
| 679 |
+
"learning_rate": 1.5899554400231233e-05,
|
| 680 |
+
"loss": 0.4942,
|
| 681 |
+
"step": 96
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 1.8624,
|
| 685 |
+
"grad_norm": 0.507769964802943,
|
| 686 |
+
"learning_rate": 1.579062279710879e-05,
|
| 687 |
+
"loss": 0.4703,
|
| 688 |
+
"step": 97
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 1.8816000000000002,
|
| 692 |
+
"grad_norm": 0.46040360287823795,
|
| 693 |
+
"learning_rate": 1.568064746731156e-05,
|
| 694 |
+
"loss": 0.4495,
|
| 695 |
+
"step": 98
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 1.9008,
|
| 699 |
+
"grad_norm": 0.40150741395369927,
|
| 700 |
+
"learning_rate": 1.5569648233264395e-05,
|
| 701 |
+
"loss": 0.489,
|
| 702 |
+
"step": 99
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 1.92,
|
| 706 |
+
"grad_norm": 0.547344855769592,
|
| 707 |
+
"learning_rate": 1.5457645101945046e-05,
|
| 708 |
+
"loss": 0.4778,
|
| 709 |
+
"step": 100
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 1.9392,
|
| 713 |
+
"grad_norm": 0.353681783992314,
|
| 714 |
+
"learning_rate": 1.5344658261278013e-05,
|
| 715 |
+
"loss": 0.4663,
|
| 716 |
+
"step": 101
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 1.9584000000000001,
|
| 720 |
+
"grad_norm": 0.49069375937784326,
|
| 721 |
+
"learning_rate": 1.5230708076495777e-05,
|
| 722 |
+
"loss": 0.488,
|
| 723 |
+
"step": 102
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 1.9776,
|
| 727 |
+
"grad_norm": 0.4470183844615094,
|
| 728 |
+
"learning_rate": 1.5115815086468103e-05,
|
| 729 |
+
"loss": 0.4742,
|
| 730 |
+
"step": 103
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 1.9968,
|
| 734 |
+
"grad_norm": 0.4409431058823664,
|
| 735 |
+
"learning_rate": 1.5000000000000002e-05,
|
| 736 |
+
"loss": 0.4837,
|
| 737 |
+
"step": 104
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 2.016,
|
| 741 |
+
"grad_norm": 1.0882108441429175,
|
| 742 |
+
"learning_rate": 1.4883283692099114e-05,
|
| 743 |
+
"loss": 0.8134,
|
| 744 |
+
"step": 105
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 2.0352,
|
| 748 |
+
"grad_norm": 0.7435815630951839,
|
| 749 |
+
"learning_rate": 1.4765687200213079e-05,
|
| 750 |
+
"loss": 0.4272,
|
| 751 |
+
"step": 106
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 2.0544,
|
| 755 |
+
"grad_norm": 0.6140584304876984,
|
| 756 |
+
"learning_rate": 1.4647231720437687e-05,
|
| 757 |
+
"loss": 0.4489,
|
| 758 |
+
"step": 107
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 2.0736,
|
| 762 |
+
"grad_norm": 0.5887853366619927,
|
| 763 |
+
"learning_rate": 1.4527938603696376e-05,
|
| 764 |
+
"loss": 0.44,
|
| 765 |
+
"step": 108
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 2.0928,
|
| 769 |
+
"grad_norm": 0.7595049900887504,
|
| 770 |
+
"learning_rate": 1.4407829351891858e-05,
|
| 771 |
+
"loss": 0.441,
|
| 772 |
+
"step": 109
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 2.112,
|
| 776 |
+
"grad_norm": 0.47995539022631356,
|
| 777 |
+
"learning_rate": 1.4286925614030542e-05,
|
| 778 |
+
"loss": 0.4105,
|
| 779 |
+
"step": 110
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 2.1312,
|
| 783 |
+
"grad_norm": 0.6893426018303987,
|
| 784 |
+
"learning_rate": 1.4165249182320401e-05,
|
| 785 |
+
"loss": 0.4084,
|
| 786 |
+
"step": 111
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 2.1504,
|
| 790 |
+
"grad_norm": 0.5054111558477574,
|
| 791 |
+
"learning_rate": 1.404282198824305e-05,
|
| 792 |
+
"loss": 0.4233,
|
| 793 |
+
"step": 112
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 2.1696,
|
| 797 |
+
"grad_norm": 0.5907935238290536,
|
| 798 |
+
"learning_rate": 1.3919666098600753e-05,
|
| 799 |
+
"loss": 0.4352,
|
| 800 |
+
"step": 113
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 2.1888,
|
| 804 |
+
"grad_norm": 0.6273598134006475,
|
| 805 |
+
"learning_rate": 1.3795803711538966e-05,
|
| 806 |
+
"loss": 0.4224,
|
| 807 |
+
"step": 114
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 2.208,
|
| 811 |
+
"grad_norm": 0.5366725428498582,
|
| 812 |
+
"learning_rate": 1.3671257152545277e-05,
|
| 813 |
+
"loss": 0.4349,
|
| 814 |
+
"step": 115
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 2.2272,
|
| 818 |
+
"grad_norm": 0.4928816412600711,
|
| 819 |
+
"learning_rate": 1.3546048870425356e-05,
|
| 820 |
+
"loss": 0.4186,
|
| 821 |
+
"step": 116
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 2.2464,
|
| 825 |
+
"grad_norm": 0.5146973809995227,
|
| 826 |
+
"learning_rate": 1.342020143325669e-05,
|
| 827 |
+
"loss": 0.4298,
|
| 828 |
+
"step": 117
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 2.2656,
|
| 832 |
+
"grad_norm": 0.5556751587428149,
|
| 833 |
+
"learning_rate": 1.3293737524320798e-05,
|
| 834 |
+
"loss": 0.3997,
|
| 835 |
+
"step": 118
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 2.2848,
|
| 839 |
+
"grad_norm": 0.42527696851476915,
|
| 840 |
+
"learning_rate": 1.3166679938014728e-05,
|
| 841 |
+
"loss": 0.4218,
|
| 842 |
+
"step": 119
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 2.304,
|
| 846 |
+
"grad_norm": 0.5036645214489491,
|
| 847 |
+
"learning_rate": 1.303905157574247e-05,
|
| 848 |
+
"loss": 0.4148,
|
| 849 |
+
"step": 120
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 2.3232,
|
| 853 |
+
"grad_norm": 0.4873585280528824,
|
| 854 |
+
"learning_rate": 1.291087544178713e-05,
|
| 855 |
+
"loss": 0.4002,
|
| 856 |
+
"step": 121
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 2.3424,
|
| 860 |
+
"grad_norm": 0.5252450814695536,
|
| 861 |
+
"learning_rate": 1.2782174639164528e-05,
|
| 862 |
+
"loss": 0.4009,
|
| 863 |
+
"step": 122
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 2.3616,
|
| 867 |
+
"grad_norm": 0.4460681328816807,
|
| 868 |
+
"learning_rate": 1.2652972365459008e-05,
|
| 869 |
+
"loss": 0.4443,
|
| 870 |
+
"step": 123
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 2.3808,
|
| 874 |
+
"grad_norm": 0.4453898566110727,
|
| 875 |
+
"learning_rate": 1.2523291908642219e-05,
|
| 876 |
+
"loss": 0.4108,
|
| 877 |
+
"step": 124
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 2.4,
|
| 881 |
+
"grad_norm": 0.44364897863496705,
|
| 882 |
+
"learning_rate": 1.2393156642875579e-05,
|
| 883 |
+
"loss": 0.4181,
|
| 884 |
+
"step": 125
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 2.4192,
|
| 888 |
+
"grad_norm": 0.4164405975098455,
|
| 889 |
+
"learning_rate": 1.2262590024297226e-05,
|
| 890 |
+
"loss": 0.4319,
|
| 891 |
+
"step": 126
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 2.4384,
|
| 895 |
+
"grad_norm": 0.43923104408591046,
|
| 896 |
+
"learning_rate": 1.2131615586794162e-05,
|
| 897 |
+
"loss": 0.4113,
|
| 898 |
+
"step": 127
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 2.4576000000000002,
|
| 902 |
+
"grad_norm": 0.3446487510196942,
|
| 903 |
+
"learning_rate": 1.2000256937760446e-05,
|
| 904 |
+
"loss": 0.4167,
|
| 905 |
+
"step": 128
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 2.4768,
|
| 909 |
+
"grad_norm": 0.4030196278219182,
|
| 910 |
+
"learning_rate": 1.1868537753842052e-05,
|
| 911 |
+
"loss": 0.429,
|
| 912 |
+
"step": 129
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 2.496,
|
| 916 |
+
"grad_norm": 0.37984372539007827,
|
| 917 |
+
"learning_rate": 1.1736481776669307e-05,
|
| 918 |
+
"loss": 0.425,
|
| 919 |
+
"step": 130
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 2.5152,
|
| 923 |
+
"grad_norm": 0.3954712252628794,
|
| 924 |
+
"learning_rate": 1.1604112808577603e-05,
|
| 925 |
+
"loss": 0.4209,
|
| 926 |
+
"step": 131
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 2.5343999999999998,
|
| 930 |
+
"grad_norm": 0.4087335368950779,
|
| 931 |
+
"learning_rate": 1.1471454708317163e-05,
|
| 932 |
+
"loss": 0.4184,
|
| 933 |
+
"step": 132
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 2.5536,
|
| 937 |
+
"grad_norm": 0.4005302069999065,
|
| 938 |
+
"learning_rate": 1.1338531386752618e-05,
|
| 939 |
+
"loss": 0.4295,
|
| 940 |
+
"step": 133
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 2.5728,
|
| 944 |
+
"grad_norm": 0.41903995722966514,
|
| 945 |
+
"learning_rate": 1.1205366802553231e-05,
|
| 946 |
+
"loss": 0.4053,
|
| 947 |
+
"step": 134
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 2.592,
|
| 951 |
+
"grad_norm": 0.34543264101270166,
|
| 952 |
+
"learning_rate": 1.107198495787448e-05,
|
| 953 |
+
"loss": 0.4122,
|
| 954 |
+
"step": 135
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 2.6112,
|
| 958 |
+
"grad_norm": 0.35088482777238417,
|
| 959 |
+
"learning_rate": 1.0938409894031793e-05,
|
| 960 |
+
"loss": 0.4,
|
| 961 |
+
"step": 136
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 2.6304,
|
| 965 |
+
"grad_norm": 0.3830825201136949,
|
| 966 |
+
"learning_rate": 1.0804665687167262e-05,
|
| 967 |
+
"loss": 0.4452,
|
| 968 |
+
"step": 137
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 2.6496,
|
| 972 |
+
"grad_norm": 0.2987260548692871,
|
| 973 |
+
"learning_rate": 1.0670776443910024e-05,
|
| 974 |
+
"loss": 0.4246,
|
| 975 |
+
"step": 138
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 2.6688,
|
| 979 |
+
"grad_norm": 0.39721449587146224,
|
| 980 |
+
"learning_rate": 1.0536766297031216e-05,
|
| 981 |
+
"loss": 0.4131,
|
| 982 |
+
"step": 139
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 2.6879999999999997,
|
| 986 |
+
"grad_norm": 0.31933394327588016,
|
| 987 |
+
"learning_rate": 1.0402659401094154e-05,
|
| 988 |
+
"loss": 0.4192,
|
| 989 |
+
"step": 140
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 2.7072000000000003,
|
| 993 |
+
"grad_norm": 0.3205758647710288,
|
| 994 |
+
"learning_rate": 1.0268479928100615e-05,
|
| 995 |
+
"loss": 0.4063,
|
| 996 |
+
"step": 141
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 2.7264,
|
| 1000 |
+
"grad_norm": 0.3375206996764473,
|
| 1001 |
+
"learning_rate": 1.0134252063133976e-05,
|
| 1002 |
+
"loss": 0.3979,
|
| 1003 |
+
"step": 142
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 2.7456,
|
| 1007 |
+
"grad_norm": 0.347961457448101,
|
| 1008 |
+
"learning_rate": 1e-05,
|
| 1009 |
+
"loss": 0.4245,
|
| 1010 |
+
"step": 143
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 2.7648,
|
| 1014 |
+
"grad_norm": 0.3390144156435501,
|
| 1015 |
+
"learning_rate": 9.865747936866027e-06,
|
| 1016 |
+
"loss": 0.4277,
|
| 1017 |
+
"step": 144
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 2.784,
|
| 1021 |
+
"grad_norm": 0.33326035893729095,
|
| 1022 |
+
"learning_rate": 9.73152007189939e-06,
|
| 1023 |
+
"loss": 0.4454,
|
| 1024 |
+
"step": 145
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 2.8032,
|
| 1028 |
+
"grad_norm": 0.31677007135038937,
|
| 1029 |
+
"learning_rate": 9.597340598905851e-06,
|
| 1030 |
+
"loss": 0.4147,
|
| 1031 |
+
"step": 146
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 2.8224,
|
| 1035 |
+
"grad_norm": 0.3108862587740751,
|
| 1036 |
+
"learning_rate": 9.463233702968784e-06,
|
| 1037 |
+
"loss": 0.3999,
|
| 1038 |
+
"step": 147
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 2.8416,
|
| 1042 |
+
"grad_norm": 0.34523790102395624,
|
| 1043 |
+
"learning_rate": 9.329223556089976e-06,
|
| 1044 |
+
"loss": 0.464,
|
| 1045 |
+
"step": 148
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 2.8608000000000002,
|
| 1049 |
+
"grad_norm": 0.3165498325744382,
|
| 1050 |
+
"learning_rate": 9.195334312832742e-06,
|
| 1051 |
+
"loss": 0.4285,
|
| 1052 |
+
"step": 149
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 2.88,
|
| 1056 |
+
"grad_norm": 0.2910881126817616,
|
| 1057 |
+
"learning_rate": 9.061590105968208e-06,
|
| 1058 |
+
"loss": 0.4037,
|
| 1059 |
+
"step": 150
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 2.8992,
|
| 1063 |
+
"grad_norm": 0.3037728807050414,
|
| 1064 |
+
"learning_rate": 8.928015042125523e-06,
|
| 1065 |
+
"loss": 0.4363,
|
| 1066 |
+
"step": 151
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 2.9184,
|
| 1070 |
+
"grad_norm": 0.3280129025176348,
|
| 1071 |
+
"learning_rate": 8.79463319744677e-06,
|
| 1072 |
+
"loss": 0.4568,
|
| 1073 |
+
"step": 152
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 2.9375999999999998,
|
| 1077 |
+
"grad_norm": 0.29211906601620385,
|
| 1078 |
+
"learning_rate": 8.661468613247387e-06,
|
| 1079 |
+
"loss": 0.4029,
|
| 1080 |
+
"step": 153
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 2.9568,
|
| 1084 |
+
"grad_norm": 0.35086790360961523,
|
| 1085 |
+
"learning_rate": 8.528545291682839e-06,
|
| 1086 |
+
"loss": 0.4382,
|
| 1087 |
+
"step": 154
|
| 1088 |
+
},
|
| 1089 |
+
{
|
| 1090 |
+
"epoch": 2.976,
|
| 1091 |
+
"grad_norm": 0.32617832015335346,
|
| 1092 |
+
"learning_rate": 8.395887191422397e-06,
|
| 1093 |
+
"loss": 0.4048,
|
| 1094 |
+
"step": 155
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 2.9952,
|
| 1098 |
+
"grad_norm": 0.33846500818394326,
|
| 1099 |
+
"learning_rate": 8.263518223330698e-06,
|
| 1100 |
+
"loss": 0.4288,
|
| 1101 |
+
"step": 156
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 3.0144,
|
| 1105 |
+
"grad_norm": 1.020138635814637,
|
| 1106 |
+
"learning_rate": 8.131462246157953e-06,
|
| 1107 |
+
"loss": 0.793,
|
| 1108 |
+
"step": 157
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"epoch": 3.0336,
|
| 1112 |
+
"grad_norm": 0.3774264603857472,
|
| 1113 |
+
"learning_rate": 7.999743062239557e-06,
|
| 1114 |
+
"loss": 0.397,
|
| 1115 |
+
"step": 158
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 3.0528,
|
| 1119 |
+
"grad_norm": 0.4092901137443242,
|
| 1120 |
+
"learning_rate": 7.868384413205842e-06,
|
| 1121 |
+
"loss": 0.3791,
|
| 1122 |
+
"step": 159
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 3.072,
|
| 1126 |
+
"grad_norm": 0.3785101046478068,
|
| 1127 |
+
"learning_rate": 7.73740997570278e-06,
|
| 1128 |
+
"loss": 0.3934,
|
| 1129 |
+
"step": 160
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 3.0912,
|
| 1133 |
+
"grad_norm": 0.372898202964631,
|
| 1134 |
+
"learning_rate": 7.606843357124426e-06,
|
| 1135 |
+
"loss": 0.3634,
|
| 1136 |
+
"step": 161
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 3.1104,
|
| 1140 |
+
"grad_norm": 0.39133158000875656,
|
| 1141 |
+
"learning_rate": 7.476708091357783e-06,
|
| 1142 |
+
"loss": 0.4181,
|
| 1143 |
+
"step": 162
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 3.1296,
|
| 1147 |
+
"grad_norm": 0.3750444567797731,
|
| 1148 |
+
"learning_rate": 7.347027634540993e-06,
|
| 1149 |
+
"loss": 0.3691,
|
| 1150 |
+
"step": 163
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 3.1488,
|
| 1154 |
+
"grad_norm": 0.3389613665318882,
|
| 1155 |
+
"learning_rate": 7.217825360835475e-06,
|
| 1156 |
+
"loss": 0.3586,
|
| 1157 |
+
"step": 164
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 3.168,
|
| 1161 |
+
"grad_norm": 0.39580175514163224,
|
| 1162 |
+
"learning_rate": 7.089124558212872e-06,
|
| 1163 |
+
"loss": 0.3947,
|
| 1164 |
+
"step": 165
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 3.1872,
|
| 1168 |
+
"grad_norm": 0.33616894434615413,
|
| 1169 |
+
"learning_rate": 6.960948424257532e-06,
|
| 1170 |
+
"loss": 0.387,
|
| 1171 |
+
"step": 166
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 3.2064,
|
| 1175 |
+
"grad_norm": 0.31890561182423444,
|
| 1176 |
+
"learning_rate": 6.833320061985278e-06,
|
| 1177 |
+
"loss": 0.3561,
|
| 1178 |
+
"step": 167
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 3.2256,
|
| 1182 |
+
"grad_norm": 0.33546862868652977,
|
| 1183 |
+
"learning_rate": 6.706262475679205e-06,
|
| 1184 |
+
"loss": 0.3905,
|
| 1185 |
+
"step": 168
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 3.2448,
|
| 1189 |
+
"grad_norm": 0.3157109353160834,
|
| 1190 |
+
"learning_rate": 6.579798566743314e-06,
|
| 1191 |
+
"loss": 0.3684,
|
| 1192 |
+
"step": 169
|
| 1193 |
+
},
|
| 1194 |
+
{
|
| 1195 |
+
"epoch": 3.2640000000000002,
|
| 1196 |
+
"grad_norm": 0.3596941947223706,
|
| 1197 |
+
"learning_rate": 6.453951129574644e-06,
|
| 1198 |
+
"loss": 0.3504,
|
| 1199 |
+
"step": 170
|
| 1200 |
+
},
|
| 1201 |
+
{
|
| 1202 |
+
"epoch": 3.2832,
|
| 1203 |
+
"grad_norm": 0.32969200970175466,
|
| 1204 |
+
"learning_rate": 6.3287428474547256e-06,
|
| 1205 |
+
"loss": 0.392,
|
| 1206 |
+
"step": 171
|
| 1207 |
+
},
|
| 1208 |
+
{
|
| 1209 |
+
"epoch": 3.3024,
|
| 1210 |
+
"grad_norm": 0.3129030963474378,
|
| 1211 |
+
"learning_rate": 6.204196288461037e-06,
|
| 1212 |
+
"loss": 0.3803,
|
| 1213 |
+
"step": 172
|
| 1214 |
+
},
|
| 1215 |
+
{
|
| 1216 |
+
"epoch": 3.3216,
|
| 1217 |
+
"grad_norm": 0.31106697280766377,
|
| 1218 |
+
"learning_rate": 6.080333901399252e-06,
|
| 1219 |
+
"loss": 0.3753,
|
| 1220 |
+
"step": 173
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 3.3407999999999998,
|
| 1224 |
+
"grad_norm": 0.35335180316405873,
|
| 1225 |
+
"learning_rate": 5.957178011756952e-06,
|
| 1226 |
+
"loss": 0.3872,
|
| 1227 |
+
"step": 174
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"epoch": 3.36,
|
| 1231 |
+
"grad_norm": 0.338760849297763,
|
| 1232 |
+
"learning_rate": 5.834750817679606e-06,
|
| 1233 |
+
"loss": 0.3609,
|
| 1234 |
+
"step": 175
|
| 1235 |
+
},
|
| 1236 |
+
{
|
| 1237 |
+
"epoch": 3.3792,
|
| 1238 |
+
"grad_norm": 0.28013515196906613,
|
| 1239 |
+
"learning_rate": 5.713074385969457e-06,
|
| 1240 |
+
"loss": 0.3661,
|
| 1241 |
+
"step": 176
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 3.3984,
|
| 1245 |
+
"grad_norm": 0.28450874000664983,
|
| 1246 |
+
"learning_rate": 5.5921706481081405e-06,
|
| 1247 |
+
"loss": 0.3811,
|
| 1248 |
+
"step": 177
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 3.4176,
|
| 1252 |
+
"grad_norm": 0.35410125064952747,
|
| 1253 |
+
"learning_rate": 5.47206139630363e-06,
|
| 1254 |
+
"loss": 0.4106,
|
| 1255 |
+
"step": 178
|
| 1256 |
+
},
|
| 1257 |
+
{
|
| 1258 |
+
"epoch": 3.4368,
|
| 1259 |
+
"grad_norm": 0.31158016364544316,
|
| 1260 |
+
"learning_rate": 5.352768279562315e-06,
|
| 1261 |
+
"loss": 0.3524,
|
| 1262 |
+
"step": 179
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 3.456,
|
| 1266 |
+
"grad_norm": 0.269602389355477,
|
| 1267 |
+
"learning_rate": 5.234312799786921e-06,
|
| 1268 |
+
"loss": 0.4018,
|
| 1269 |
+
"step": 180
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 3.4752,
|
| 1273 |
+
"grad_norm": 0.27689334795794307,
|
| 1274 |
+
"learning_rate": 5.116716307900893e-06,
|
| 1275 |
+
"loss": 0.3559,
|
| 1276 |
+
"step": 181
|
| 1277 |
+
},
|
| 1278 |
+
{
|
| 1279 |
+
"epoch": 3.4944,
|
| 1280 |
+
"grad_norm": 0.29639536780921816,
|
| 1281 |
+
"learning_rate": 5.000000000000003e-06,
|
| 1282 |
+
"loss": 0.4214,
|
| 1283 |
+
"step": 182
|
| 1284 |
+
},
|
| 1285 |
+
{
|
| 1286 |
+
"epoch": 3.5136,
|
| 1287 |
+
"grad_norm": 0.2592670877724285,
|
| 1288 |
+
"learning_rate": 4.8841849135319015e-06,
|
| 1289 |
+
"loss": 0.3791,
|
| 1290 |
+
"step": 183
|
| 1291 |
+
},
|
| 1292 |
+
{
|
| 1293 |
+
"epoch": 3.5328,
|
| 1294 |
+
"grad_norm": 0.2636697825348893,
|
| 1295 |
+
"learning_rate": 4.769291923504226e-06,
|
| 1296 |
+
"loss": 0.3616,
|
| 1297 |
+
"step": 184
|
| 1298 |
+
},
|
| 1299 |
+
{
|
| 1300 |
+
"epoch": 3.552,
|
| 1301 |
+
"grad_norm": 0.2818080306054233,
|
| 1302 |
+
"learning_rate": 4.655341738721989e-06,
|
| 1303 |
+
"loss": 0.3834,
|
| 1304 |
+
"step": 185
|
| 1305 |
+
},
|
| 1306 |
+
{
|
| 1307 |
+
"epoch": 3.5712,
|
| 1308 |
+
"grad_norm": 0.28039661320923065,
|
| 1309 |
+
"learning_rate": 4.542354898054953e-06,
|
| 1310 |
+
"loss": 0.3775,
|
| 1311 |
+
"step": 186
|
| 1312 |
+
},
|
| 1313 |
+
{
|
| 1314 |
+
"epoch": 3.5904,
|
| 1315 |
+
"grad_norm": 0.2800741052248581,
|
| 1316 |
+
"learning_rate": 4.430351766735609e-06,
|
| 1317 |
+
"loss": 0.3735,
|
| 1318 |
+
"step": 187
|
| 1319 |
+
},
|
| 1320 |
+
{
|
| 1321 |
+
"epoch": 3.6096,
|
| 1322 |
+
"grad_norm": 0.27173177997314635,
|
| 1323 |
+
"learning_rate": 4.319352532688444e-06,
|
| 1324 |
+
"loss": 0.3794,
|
| 1325 |
+
"step": 188
|
| 1326 |
+
},
|
| 1327 |
+
{
|
| 1328 |
+
"epoch": 3.6288,
|
| 1329 |
+
"grad_norm": 0.29558057627429624,
|
| 1330 |
+
"learning_rate": 4.209377202891212e-06,
|
| 1331 |
+
"loss": 0.365,
|
| 1332 |
+
"step": 189
|
| 1333 |
+
},
|
| 1334 |
+
{
|
| 1335 |
+
"epoch": 3.648,
|
| 1336 |
+
"grad_norm": 0.264313692535776,
|
| 1337 |
+
"learning_rate": 4.100445599768774e-06,
|
| 1338 |
+
"loss": 0.3787,
|
| 1339 |
+
"step": 190
|
| 1340 |
+
},
|
| 1341 |
+
{
|
| 1342 |
+
"epoch": 3.6672000000000002,
|
| 1343 |
+
"grad_norm": 0.2633179503354047,
|
| 1344 |
+
"learning_rate": 3.99257735762021e-06,
|
| 1345 |
+
"loss": 0.3744,
|
| 1346 |
+
"step": 191
|
| 1347 |
+
},
|
| 1348 |
+
{
|
| 1349 |
+
"epoch": 3.6864,
|
| 1350 |
+
"grad_norm": 0.294359704663315,
|
| 1351 |
+
"learning_rate": 3.885791919079878e-06,
|
| 1352 |
+
"loss": 0.376,
|
| 1353 |
+
"step": 192
|
| 1354 |
+
},
|
| 1355 |
+
{
|
| 1356 |
+
"epoch": 3.7056,
|
| 1357 |
+
"grad_norm": 0.2587993420681155,
|
| 1358 |
+
"learning_rate": 3.7801085316129615e-06,
|
| 1359 |
+
"loss": 0.3787,
|
| 1360 |
+
"step": 193
|
| 1361 |
+
},
|
| 1362 |
+
{
|
| 1363 |
+
"epoch": 3.7248,
|
| 1364 |
+
"grad_norm": 0.24965892468077336,
|
| 1365 |
+
"learning_rate": 3.6755462440462288e-06,
|
| 1366 |
+
"loss": 0.381,
|
| 1367 |
+
"step": 194
|
| 1368 |
+
},
|
| 1369 |
+
{
|
| 1370 |
+
"epoch": 3.7439999999999998,
|
| 1371 |
+
"grad_norm": 0.28072547228339184,
|
| 1372 |
+
"learning_rate": 3.5721239031346067e-06,
|
| 1373 |
+
"loss": 0.4075,
|
| 1374 |
+
"step": 195
|
| 1375 |
+
},
|
| 1376 |
+
{
|
| 1377 |
+
"epoch": 3.7632,
|
| 1378 |
+
"grad_norm": 0.26049633926357213,
|
| 1379 |
+
"learning_rate": 3.4698601501641517e-06,
|
| 1380 |
+
"loss": 0.3905,
|
| 1381 |
+
"step": 196
|
| 1382 |
+
},
|
| 1383 |
+
{
|
| 1384 |
+
"epoch": 3.7824,
|
| 1385 |
+
"grad_norm": 0.254888248810916,
|
| 1386 |
+
"learning_rate": 3.3687734175920505e-06,
|
| 1387 |
+
"loss": 0.3743,
|
| 1388 |
+
"step": 197
|
| 1389 |
+
},
|
| 1390 |
+
{
|
| 1391 |
+
"epoch": 3.8016,
|
| 1392 |
+
"grad_norm": 0.24864771735101535,
|
| 1393 |
+
"learning_rate": 3.2688819257242963e-06,
|
| 1394 |
+
"loss": 0.3787,
|
| 1395 |
+
"step": 198
|
| 1396 |
+
},
|
| 1397 |
+
{
|
| 1398 |
+
"epoch": 3.8208,
|
| 1399 |
+
"grad_norm": 0.295784625351548,
|
| 1400 |
+
"learning_rate": 3.1702036794315837e-06,
|
| 1401 |
+
"loss": 0.4126,
|
| 1402 |
+
"step": 199
|
| 1403 |
+
},
|
| 1404 |
+
{
|
| 1405 |
+
"epoch": 3.84,
|
| 1406 |
+
"grad_norm": 0.2562016714282648,
|
| 1407 |
+
"learning_rate": 3.0727564649040066e-06,
|
| 1408 |
+
"loss": 0.3685,
|
| 1409 |
+
"step": 200
|
| 1410 |
+
},
|
| 1411 |
+
{
|
| 1412 |
+
"epoch": 3.8592,
|
| 1413 |
+
"grad_norm": 0.23184737210656833,
|
| 1414 |
+
"learning_rate": 2.976557846445225e-06,
|
| 1415 |
+
"loss": 0.3626,
|
| 1416 |
+
"step": 201
|
| 1417 |
+
},
|
| 1418 |
+
{
|
| 1419 |
+
"epoch": 3.8784,
|
| 1420 |
+
"grad_norm": 0.24835787721696984,
|
| 1421 |
+
"learning_rate": 2.8816251633065963e-06,
|
| 1422 |
+
"loss": 0.3603,
|
| 1423 |
+
"step": 202
|
| 1424 |
+
},
|
| 1425 |
+
{
|
| 1426 |
+
"epoch": 3.8975999999999997,
|
| 1427 |
+
"grad_norm": 0.25214551277209546,
|
| 1428 |
+
"learning_rate": 2.7879755265618558e-06,
|
| 1429 |
+
"loss": 0.39,
|
| 1430 |
+
"step": 203
|
| 1431 |
+
},
|
| 1432 |
+
{
|
| 1433 |
+
"epoch": 3.9168,
|
| 1434 |
+
"grad_norm": 0.2298876403265907,
|
| 1435 |
+
"learning_rate": 2.69562581602297e-06,
|
| 1436 |
+
"loss": 0.381,
|
| 1437 |
+
"step": 204
|
| 1438 |
+
},
|
| 1439 |
+
{
|
| 1440 |
+
"epoch": 3.936,
|
| 1441 |
+
"grad_norm": 0.22563401478833262,
|
| 1442 |
+
"learning_rate": 2.6045926771976306e-06,
|
| 1443 |
+
"loss": 0.367,
|
| 1444 |
+
"step": 205
|
| 1445 |
+
},
|
| 1446 |
+
{
|
| 1447 |
+
"epoch": 3.9552,
|
| 1448 |
+
"grad_norm": 0.23685702617344914,
|
| 1449 |
+
"learning_rate": 2.514892518288988e-06,
|
| 1450 |
+
"loss": 0.3959,
|
| 1451 |
+
"step": 206
|
| 1452 |
+
},
|
| 1453 |
+
{
|
| 1454 |
+
"epoch": 3.9744,
|
| 1455 |
+
"grad_norm": 0.23045683281281373,
|
| 1456 |
+
"learning_rate": 2.4265415072382016e-06,
|
| 1457 |
+
"loss": 0.3817,
|
| 1458 |
+
"step": 207
|
| 1459 |
+
},
|
| 1460 |
+
{
|
| 1461 |
+
"epoch": 3.9936,
|
| 1462 |
+
"grad_norm": 0.2557068378242125,
|
| 1463 |
+
"learning_rate": 2.339555568810221e-06,
|
| 1464 |
+
"loss": 0.3928,
|
| 1465 |
+
"step": 208
|
| 1466 |
+
},
|
| 1467 |
+
{
|
| 1468 |
+
"epoch": 4.0128,
|
| 1469 |
+
"grad_norm": 0.7830880693184162,
|
| 1470 |
+
"learning_rate": 2.2539503817234553e-06,
|
| 1471 |
+
"loss": 0.6178,
|
| 1472 |
+
"step": 209
|
| 1473 |
+
},
|
| 1474 |
+
{
|
| 1475 |
+
"epoch": 4.032,
|
| 1476 |
+
"grad_norm": 0.2994912397112769,
|
| 1477 |
+
"learning_rate": 2.1697413758237785e-06,
|
| 1478 |
+
"loss": 0.3751,
|
| 1479 |
+
"step": 210
|
| 1480 |
+
},
|
| 1481 |
+
{
|
| 1482 |
+
"epoch": 4.0512,
|
| 1483 |
+
"grad_norm": 0.27180004108299344,
|
| 1484 |
+
"learning_rate": 2.0869437293033835e-06,
|
| 1485 |
+
"loss": 0.3907,
|
| 1486 |
+
"step": 211
|
| 1487 |
+
},
|
| 1488 |
+
{
|
| 1489 |
+
"epoch": 4.0704,
|
| 1490 |
+
"grad_norm": 0.25228675824788255,
|
| 1491 |
+
"learning_rate": 2.0055723659649907e-06,
|
| 1492 |
+
"loss": 0.3491,
|
| 1493 |
+
"step": 212
|
| 1494 |
+
},
|
| 1495 |
+
{
|
| 1496 |
+
"epoch": 4.0896,
|
| 1497 |
+
"grad_norm": 0.25426315254601745,
|
| 1498 |
+
"learning_rate": 1.9256419525319316e-06,
|
| 1499 |
+
"loss": 0.3702,
|
| 1500 |
+
"step": 213
|
| 1501 |
+
},
|
| 1502 |
+
{
|
| 1503 |
+
"epoch": 4.1088,
|
| 1504 |
+
"grad_norm": 0.24502228586672384,
|
| 1505 |
+
"learning_rate": 1.8471668960045575e-06,
|
| 1506 |
+
"loss": 0.3579,
|
| 1507 |
+
"step": 214
|
| 1508 |
+
},
|
| 1509 |
+
{
|
| 1510 |
+
"epoch": 4.128,
|
| 1511 |
+
"grad_norm": 0.25266536909693693,
|
| 1512 |
+
"learning_rate": 1.7701613410634367e-06,
|
| 1513 |
+
"loss": 0.3637,
|
| 1514 |
+
"step": 215
|
| 1515 |
+
},
|
| 1516 |
+
{
|
| 1517 |
+
"epoch": 4.1472,
|
| 1518 |
+
"grad_norm": 0.2575736682974073,
|
| 1519 |
+
"learning_rate": 1.6946391675198838e-06,
|
| 1520 |
+
"loss": 0.3452,
|
| 1521 |
+
"step": 216
|
| 1522 |
+
},
|
| 1523 |
+
{
|
| 1524 |
+
"epoch": 4.1664,
|
| 1525 |
+
"grad_norm": 0.24916709225522912,
|
| 1526 |
+
"learning_rate": 1.620613987814189e-06,
|
| 1527 |
+
"loss": 0.3683,
|
| 1528 |
+
"step": 217
|
| 1529 |
+
},
|
| 1530 |
+
{
|
| 1531 |
+
"epoch": 4.1856,
|
| 1532 |
+
"grad_norm": 0.281807519573222,
|
| 1533 |
+
"learning_rate": 1.5480991445620541e-06,
|
| 1534 |
+
"loss": 0.3799,
|
| 1535 |
+
"step": 218
|
| 1536 |
+
},
|
| 1537 |
+
{
|
| 1538 |
+
"epoch": 4.2048,
|
| 1539 |
+
"grad_norm": 0.24176624894276244,
|
| 1540 |
+
"learning_rate": 1.4771077081496654e-06,
|
| 1541 |
+
"loss": 0.371,
|
| 1542 |
+
"step": 219
|
| 1543 |
+
},
|
| 1544 |
+
{
|
| 1545 |
+
"epoch": 4.224,
|
| 1546 |
+
"grad_norm": 0.23653186267447301,
|
| 1547 |
+
"learning_rate": 1.407652474377832e-06,
|
| 1548 |
+
"loss": 0.3395,
|
| 1549 |
+
"step": 220
|
| 1550 |
+
},
|
| 1551 |
+
{
|
| 1552 |
+
"epoch": 4.2432,
|
| 1553 |
+
"grad_norm": 0.2705961403602411,
|
| 1554 |
+
"learning_rate": 1.339745962155613e-06,
|
| 1555 |
+
"loss": 0.3608,
|
| 1556 |
+
"step": 221
|
| 1557 |
+
},
|
| 1558 |
+
{
|
| 1559 |
+
"epoch": 4.2624,
|
| 1560 |
+
"grad_norm": 0.22167993784994147,
|
| 1561 |
+
"learning_rate": 1.273400411243857e-06,
|
| 1562 |
+
"loss": 0.3186,
|
| 1563 |
+
"step": 222
|
| 1564 |
+
},
|
| 1565 |
+
{
|
| 1566 |
+
"epoch": 4.2816,
|
| 1567 |
+
"grad_norm": 0.2274269644108438,
|
| 1568 |
+
"learning_rate": 1.2086277800490554e-06,
|
| 1569 |
+
"loss": 0.3514,
|
| 1570 |
+
"step": 223
|
| 1571 |
+
},
|
| 1572 |
+
{
|
| 1573 |
+
"epoch": 4.3008,
|
| 1574 |
+
"grad_norm": 0.2403819850744848,
|
| 1575 |
+
"learning_rate": 1.1454397434679022e-06,
|
| 1576 |
+
"loss": 0.3613,
|
| 1577 |
+
"step": 224
|
| 1578 |
+
},
|
| 1579 |
+
{
|
| 1580 |
+
"epoch": 4.32,
|
| 1581 |
+
"grad_norm": 0.22064088329537343,
|
| 1582 |
+
"learning_rate": 1.083847690782972e-06,
|
| 1583 |
+
"loss": 0.3675,
|
| 1584 |
+
"step": 225
|
| 1585 |
+
},
|
| 1586 |
+
{
|
| 1587 |
+
"epoch": 4.3392,
|
| 1588 |
+
"grad_norm": 0.22977983110111344,
|
| 1589 |
+
"learning_rate": 1.0238627236098619e-06,
|
| 1590 |
+
"loss": 0.3632,
|
| 1591 |
+
"step": 226
|
| 1592 |
+
},
|
| 1593 |
+
{
|
| 1594 |
+
"epoch": 4.3584,
|
| 1595 |
+
"grad_norm": 0.2632092739749564,
|
| 1596 |
+
"learning_rate": 9.65495653896179e-07,
|
| 1597 |
+
"loss": 0.3668,
|
| 1598 |
+
"step": 227
|
| 1599 |
+
},
|
| 1600 |
+
{
|
| 1601 |
+
"epoch": 4.3776,
|
| 1602 |
+
"grad_norm": 0.23500771271067225,
|
| 1603 |
+
"learning_rate": 9.08757001972762e-07,
|
| 1604 |
+
"loss": 0.3413,
|
| 1605 |
+
"step": 228
|
| 1606 |
+
},
|
| 1607 |
+
{
|
| 1608 |
+
"epoch": 4.3968,
|
| 1609 |
+
"grad_norm": 0.23739342911891992,
|
| 1610 |
+
"learning_rate": 8.536569946574546e-07,
|
| 1611 |
+
"loss": 0.41,
|
| 1612 |
+
"step": 229
|
| 1613 |
+
},
|
| 1614 |
+
{
|
| 1615 |
+
"epoch": 4.416,
|
| 1616 |
+
"grad_norm": 0.22230388202399864,
|
| 1617 |
+
"learning_rate": 8.002055634117578e-07,
|
| 1618 |
+
"loss": 0.3368,
|
| 1619 |
+
"step": 230
|
| 1620 |
+
},
|
| 1621 |
+
{
|
| 1622 |
+
"epoch": 4.4352,
|
| 1623 |
+
"grad_norm": 0.2379276108050257,
|
| 1624 |
+
"learning_rate": 7.48412342550765e-07,
|
| 1625 |
+
"loss": 0.3817,
|
| 1626 |
+
"step": 231
|
| 1627 |
+
},
|
| 1628 |
+
{
|
| 1629 |
+
"epoch": 4.4544,
|
| 1630 |
+
"grad_norm": 0.2087482581701038,
|
| 1631 |
+
"learning_rate": 6.98286667506618e-07,
|
| 1632 |
+
"loss": 0.3302,
|
| 1633 |
+
"step": 232
|
| 1634 |
+
},
|
| 1635 |
+
{
|
| 1636 |
+
"epoch": 4.4736,
|
| 1637 |
+
"grad_norm": 0.22825791960335742,
|
| 1638 |
+
"learning_rate": 6.498375731458529e-07,
|
| 1639 |
+
"loss": 0.3782,
|
| 1640 |
+
"step": 233
|
| 1641 |
+
},
|
| 1642 |
+
{
|
| 1643 |
+
"epoch": 4.4928,
|
| 1644 |
+
"grad_norm": 0.2266085430621897,
|
| 1645 |
+
"learning_rate": 6.030737921409169e-07,
|
| 1646 |
+
"loss": 0.363,
|
| 1647 |
+
"step": 234
|
| 1648 |
+
},
|
| 1649 |
+
{
|
| 1650 |
+
"epoch": 4.5120000000000005,
|
| 1651 |
+
"grad_norm": 0.22297447299681908,
|
| 1652 |
+
"learning_rate": 5.580037533961546e-07,
|
| 1653 |
+
"loss": 0.3475,
|
| 1654 |
+
"step": 235
|
| 1655 |
+
},
|
| 1656 |
+
{
|
| 1657 |
+
"epoch": 4.5312,
|
| 1658 |
+
"grad_norm": 0.2190648093933851,
|
| 1659 |
+
"learning_rate": 5.146355805285452e-07,
|
| 1660 |
+
"loss": 0.3341,
|
| 1661 |
+
"step": 236
|
| 1662 |
+
},
|
| 1663 |
+
{
|
| 1664 |
+
"epoch": 4.5504,
|
| 1665 |
+
"grad_norm": 0.2280310827976436,
|
| 1666 |
+
"learning_rate": 4.7297709040346474e-07,
|
| 1667 |
+
"loss": 0.375,
|
| 1668 |
+
"step": 237
|
| 1669 |
+
},
|
| 1670 |
+
{
|
| 1671 |
+
"epoch": 4.5696,
|
| 1672 |
+
"grad_norm": 0.21439011621531331,
|
| 1673 |
+
"learning_rate": 4.3303579172574884e-07,
|
| 1674 |
+
"loss": 0.3387,
|
| 1675 |
+
"step": 238
|
| 1676 |
+
},
|
| 1677 |
+
{
|
| 1678 |
+
"epoch": 4.5888,
|
| 1679 |
+
"grad_norm": 0.20772804460962693,
|
| 1680 |
+
"learning_rate": 3.9481888368627764e-07,
|
| 1681 |
+
"loss": 0.3404,
|
| 1682 |
+
"step": 239
|
| 1683 |
+
},
|
| 1684 |
+
{
|
| 1685 |
+
"epoch": 4.608,
|
| 1686 |
+
"grad_norm": 0.21771574331525612,
|
| 1687 |
+
"learning_rate": 3.5833325466437697e-07,
|
| 1688 |
+
"loss": 0.3488,
|
| 1689 |
+
"step": 240
|
| 1690 |
+
},
|
| 1691 |
+
{
|
| 1692 |
+
"epoch": 4.6272,
|
| 1693 |
+
"grad_norm": 0.23594669324224943,
|
| 1694 |
+
"learning_rate": 3.235854809862193e-07,
|
| 1695 |
+
"loss": 0.39,
|
| 1696 |
+
"step": 241
|
| 1697 |
+
},
|
| 1698 |
+
{
|
| 1699 |
+
"epoch": 4.6464,
|
| 1700 |
+
"grad_norm": 0.22365330954778662,
|
| 1701 |
+
"learning_rate": 2.905818257394799e-07,
|
| 1702 |
+
"loss": 0.35,
|
| 1703 |
+
"step": 242
|
| 1704 |
+
},
|
| 1705 |
+
{
|
| 1706 |
+
"epoch": 4.6655999999999995,
|
| 1707 |
+
"grad_norm": 0.21399686382903388,
|
| 1708 |
+
"learning_rate": 2.593282376444539e-07,
|
| 1709 |
+
"loss": 0.3546,
|
| 1710 |
+
"step": 243
|
| 1711 |
+
},
|
| 1712 |
+
{
|
| 1713 |
+
"epoch": 4.6848,
|
| 1714 |
+
"grad_norm": 0.21775044132189242,
|
| 1715 |
+
"learning_rate": 2.2983034998182997e-07,
|
| 1716 |
+
"loss": 0.3426,
|
| 1717 |
+
"step": 244
|
| 1718 |
+
},
|
| 1719 |
+
{
|
| 1720 |
+
"epoch": 4.704,
|
| 1721 |
+
"grad_norm": 0.20302583983593797,
|
| 1722 |
+
"learning_rate": 2.0209347957732328e-07,
|
| 1723 |
+
"loss": 0.3154,
|
| 1724 |
+
"step": 245
|
| 1725 |
+
},
|
| 1726 |
+
{
|
| 1727 |
+
"epoch": 4.7232,
|
| 1728 |
+
"grad_norm": 0.21473249200698508,
|
| 1729 |
+
"learning_rate": 1.761226258433524e-07,
|
| 1730 |
+
"loss": 0.3724,
|
| 1731 |
+
"step": 246
|
| 1732 |
+
},
|
| 1733 |
+
{
|
| 1734 |
+
"epoch": 4.7424,
|
| 1735 |
+
"grad_norm": 0.20817447960741947,
|
| 1736 |
+
"learning_rate": 1.519224698779198e-07,
|
| 1737 |
+
"loss": 0.3682,
|
| 1738 |
+
"step": 247
|
| 1739 |
+
},
|
| 1740 |
+
{
|
| 1741 |
+
"epoch": 4.7616,
|
| 1742 |
+
"grad_norm": 0.221523557416081,
|
| 1743 |
+
"learning_rate": 1.2949737362087156e-07,
|
| 1744 |
+
"loss": 0.3644,
|
| 1745 |
+
"step": 248
|
| 1746 |
+
},
|
| 1747 |
+
{
|
| 1748 |
+
"epoch": 4.7808,
|
| 1749 |
+
"grad_norm": 0.2167063860471359,
|
| 1750 |
+
"learning_rate": 1.0885137906768373e-07,
|
| 1751 |
+
"loss": 0.3618,
|
| 1752 |
+
"step": 249
|
| 1753 |
+
},
|
| 1754 |
+
{
|
| 1755 |
+
"epoch": 4.8,
|
| 1756 |
+
"grad_norm": 0.21574694308607623,
|
| 1757 |
+
"learning_rate": 8.99882075409153e-08,
|
| 1758 |
+
"loss": 0.3652,
|
| 1759 |
+
"step": 250
|
| 1760 |
+
},
|
| 1761 |
+
{
|
| 1762 |
+
"epoch": 4.8192,
|
| 1763 |
+
"grad_norm": 0.2101434832584648,
|
| 1764 |
+
"learning_rate": 7.291125901946027e-08,
|
| 1765 |
+
"loss": 0.3227,
|
| 1766 |
+
"step": 251
|
| 1767 |
+
},
|
| 1768 |
+
{
|
| 1769 |
+
"epoch": 4.8384,
|
| 1770 |
+
"grad_norm": 0.2171188306835337,
|
| 1771 |
+
"learning_rate": 5.7623611525721155e-08,
|
| 1772 |
+
"loss": 0.3545,
|
| 1773 |
+
"step": 252
|
| 1774 |
+
},
|
| 1775 |
+
{
|
| 1776 |
+
"epoch": 4.8576,
|
| 1777 |
+
"grad_norm": 0.21423065183560364,
|
| 1778 |
+
"learning_rate": 4.412802057081278e-08,
|
| 1779 |
+
"loss": 0.3597,
|
| 1780 |
+
"step": 253
|
| 1781 |
+
},
|
| 1782 |
+
{
|
| 1783 |
+
"epoch": 4.8768,
|
| 1784 |
+
"grad_norm": 0.21016873942874703,
|
| 1785 |
+
"learning_rate": 3.242691865790071e-08,
|
| 1786 |
+
"loss": 0.3303,
|
| 1787 |
+
"step": 254
|
| 1788 |
+
},
|
| 1789 |
+
{
|
| 1790 |
+
"epoch": 4.896,
|
| 1791 |
+
"grad_norm": 0.21418559799752182,
|
| 1792 |
+
"learning_rate": 2.2522414843748618e-08,
|
| 1793 |
+
"loss": 0.3541,
|
| 1794 |
+
"step": 255
|
| 1795 |
+
},
|
| 1796 |
+
{
|
| 1797 |
+
"epoch": 4.9152000000000005,
|
| 1798 |
+
"grad_norm": 0.22532962149148705,
|
| 1799 |
+
"learning_rate": 1.4416294358582383e-08,
|
| 1800 |
+
"loss": 0.3535,
|
| 1801 |
+
"step": 256
|
| 1802 |
+
},
|
| 1803 |
+
{
|
| 1804 |
+
"epoch": 4.9344,
|
| 1805 |
+
"grad_norm": 0.20814547439921482,
|
| 1806 |
+
"learning_rate": 8.110018284304132e-09,
|
| 1807 |
+
"loss": 0.3458,
|
| 1808 |
+
"step": 257
|
| 1809 |
+
},
|
| 1810 |
+
{
|
| 1811 |
+
"epoch": 4.9536,
|
| 1812 |
+
"grad_norm": 0.2203758122033324,
|
| 1813 |
+
"learning_rate": 3.6047232911462506e-09,
|
| 1814 |
+
"loss": 0.3679,
|
| 1815 |
+
"step": 258
|
| 1816 |
+
},
|
| 1817 |
+
{
|
| 1818 |
+
"epoch": 4.9728,
|
| 1819 |
+
"grad_norm": 0.21762961639454312,
|
| 1820 |
+
"learning_rate": 9.012214327897006e-10,
|
| 1821 |
+
"loss": 0.3687,
|
| 1822 |
+
"step": 259
|
| 1823 |
+
},
|
| 1824 |
+
{
|
| 1825 |
+
"epoch": 4.992,
|
| 1826 |
+
"grad_norm": 0.22054334314840165,
|
| 1827 |
+
"learning_rate": 0.0,
|
| 1828 |
+
"loss": 0.3756,
|
| 1829 |
+
"step": 260
|
| 1830 |
+
},
|
| 1831 |
+
{
|
| 1832 |
+
"epoch": 4.992,
|
| 1833 |
+
"step": 260,
|
| 1834 |
+
"total_flos": 1.013794005747499e+18,
|
| 1835 |
+
"train_loss": 0.45938570224321806,
|
| 1836 |
+
"train_runtime": 28676.658,
|
| 1837 |
+
"train_samples_per_second": 1.743,
|
| 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
|