End of training
Browse files- README.md +2 -1
- all_results.json +8 -0
- train_results.json +8 -0
- trainer_state.json +2226 -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_globalbatchsize96_lr4e5_epochs3
|
|
@@ -15,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 15 |
|
| 16 |
# 10k_globalbatchsize96_lr4e5_epochs3
|
| 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_globalbatchsize96_lr4e5_epochs3
|
|
|
|
| 16 |
|
| 17 |
# 10k_globalbatchsize96_lr4e5_epochs3
|
| 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": 2.9904153354632586,
|
| 3 |
+
"total_flos": 4.4115474684667494e+17,
|
| 4 |
+
"train_loss": 0.43267787887881964,
|
| 5 |
+
"train_runtime": 10181.5896,
|
| 6 |
+
"train_samples_per_second": 2.946,
|
| 7 |
+
"train_steps_per_second": 0.031
|
| 8 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 2.9904153354632586,
|
| 3 |
+
"total_flos": 4.4115474684667494e+17,
|
| 4 |
+
"train_loss": 0.43267787887881964,
|
| 5 |
+
"train_runtime": 10181.5896,
|
| 6 |
+
"train_samples_per_second": 2.946,
|
| 7 |
+
"train_steps_per_second": 0.031
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,2226 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 2.9904153354632586,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 312,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.009584664536741214,
|
| 13 |
+
"grad_norm": 5.816953152720709,
|
| 14 |
+
"learning_rate": 1.25e-06,
|
| 15 |
+
"loss": 0.8809,
|
| 16 |
+
"step": 1
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.019169329073482427,
|
| 20 |
+
"grad_norm": 5.964463857967548,
|
| 21 |
+
"learning_rate": 2.5e-06,
|
| 22 |
+
"loss": 0.8476,
|
| 23 |
+
"step": 2
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.02875399361022364,
|
| 27 |
+
"grad_norm": 5.85622862207096,
|
| 28 |
+
"learning_rate": 3.7500000000000005e-06,
|
| 29 |
+
"loss": 0.8749,
|
| 30 |
+
"step": 3
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.038338658146964855,
|
| 34 |
+
"grad_norm": 5.365349431954843,
|
| 35 |
+
"learning_rate": 5e-06,
|
| 36 |
+
"loss": 0.8422,
|
| 37 |
+
"step": 4
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.04792332268370607,
|
| 41 |
+
"grad_norm": 4.006201698306361,
|
| 42 |
+
"learning_rate": 6.25e-06,
|
| 43 |
+
"loss": 0.8005,
|
| 44 |
+
"step": 5
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.05750798722044728,
|
| 48 |
+
"grad_norm": 2.122270635740227,
|
| 49 |
+
"learning_rate": 7.500000000000001e-06,
|
| 50 |
+
"loss": 0.7687,
|
| 51 |
+
"step": 6
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.0670926517571885,
|
| 55 |
+
"grad_norm": 3.975962985256971,
|
| 56 |
+
"learning_rate": 8.750000000000001e-06,
|
| 57 |
+
"loss": 0.7686,
|
| 58 |
+
"step": 7
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.07667731629392971,
|
| 62 |
+
"grad_norm": 4.3633197716567595,
|
| 63 |
+
"learning_rate": 1e-05,
|
| 64 |
+
"loss": 0.761,
|
| 65 |
+
"step": 8
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.08626198083067092,
|
| 69 |
+
"grad_norm": 4.812484653337378,
|
| 70 |
+
"learning_rate": 1.125e-05,
|
| 71 |
+
"loss": 0.7451,
|
| 72 |
+
"step": 9
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.09584664536741214,
|
| 76 |
+
"grad_norm": 4.996167212561038,
|
| 77 |
+
"learning_rate": 1.25e-05,
|
| 78 |
+
"loss": 0.7657,
|
| 79 |
+
"step": 10
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.10543130990415335,
|
| 83 |
+
"grad_norm": 3.391593689539008,
|
| 84 |
+
"learning_rate": 1.375e-05,
|
| 85 |
+
"loss": 0.7293,
|
| 86 |
+
"step": 11
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.11501597444089456,
|
| 90 |
+
"grad_norm": 2.024258637124186,
|
| 91 |
+
"learning_rate": 1.5000000000000002e-05,
|
| 92 |
+
"loss": 0.6753,
|
| 93 |
+
"step": 12
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.12460063897763578,
|
| 97 |
+
"grad_norm": 2.368340767997839,
|
| 98 |
+
"learning_rate": 1.6250000000000002e-05,
|
| 99 |
+
"loss": 0.675,
|
| 100 |
+
"step": 13
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.134185303514377,
|
| 104 |
+
"grad_norm": 1.9034533687196296,
|
| 105 |
+
"learning_rate": 1.7500000000000002e-05,
|
| 106 |
+
"loss": 0.6346,
|
| 107 |
+
"step": 14
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.14376996805111822,
|
| 111 |
+
"grad_norm": 1.511274358606332,
|
| 112 |
+
"learning_rate": 1.8750000000000002e-05,
|
| 113 |
+
"loss": 0.6395,
|
| 114 |
+
"step": 15
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.15335463258785942,
|
| 118 |
+
"grad_norm": 1.5258008797446443,
|
| 119 |
+
"learning_rate": 2e-05,
|
| 120 |
+
"loss": 0.6605,
|
| 121 |
+
"step": 16
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.16293929712460065,
|
| 125 |
+
"grad_norm": 1.4701453015485093,
|
| 126 |
+
"learning_rate": 2.125e-05,
|
| 127 |
+
"loss": 0.6345,
|
| 128 |
+
"step": 17
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.17252396166134185,
|
| 132 |
+
"grad_norm": 1.139297657192997,
|
| 133 |
+
"learning_rate": 2.25e-05,
|
| 134 |
+
"loss": 0.6313,
|
| 135 |
+
"step": 18
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.18210862619808307,
|
| 139 |
+
"grad_norm": 1.0628460283930738,
|
| 140 |
+
"learning_rate": 2.375e-05,
|
| 141 |
+
"loss": 0.6093,
|
| 142 |
+
"step": 19
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.19169329073482427,
|
| 146 |
+
"grad_norm": 1.0582580403121424,
|
| 147 |
+
"learning_rate": 2.5e-05,
|
| 148 |
+
"loss": 0.5867,
|
| 149 |
+
"step": 20
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.2012779552715655,
|
| 153 |
+
"grad_norm": 0.8977556913012904,
|
| 154 |
+
"learning_rate": 2.625e-05,
|
| 155 |
+
"loss": 0.5876,
|
| 156 |
+
"step": 21
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.2108626198083067,
|
| 160 |
+
"grad_norm": 1.3235677545982394,
|
| 161 |
+
"learning_rate": 2.75e-05,
|
| 162 |
+
"loss": 0.5909,
|
| 163 |
+
"step": 22
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.22044728434504793,
|
| 167 |
+
"grad_norm": 0.8869847698482135,
|
| 168 |
+
"learning_rate": 2.875e-05,
|
| 169 |
+
"loss": 0.5739,
|
| 170 |
+
"step": 23
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.23003194888178913,
|
| 174 |
+
"grad_norm": 1.1759129421868764,
|
| 175 |
+
"learning_rate": 3.0000000000000004e-05,
|
| 176 |
+
"loss": 0.588,
|
| 177 |
+
"step": 24
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.23961661341853036,
|
| 181 |
+
"grad_norm": 0.9909103854124853,
|
| 182 |
+
"learning_rate": 3.125e-05,
|
| 183 |
+
"loss": 0.5675,
|
| 184 |
+
"step": 25
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.24920127795527156,
|
| 188 |
+
"grad_norm": 1.0713203550156574,
|
| 189 |
+
"learning_rate": 3.2500000000000004e-05,
|
| 190 |
+
"loss": 0.5492,
|
| 191 |
+
"step": 26
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.25878594249201275,
|
| 195 |
+
"grad_norm": 0.9586606498055252,
|
| 196 |
+
"learning_rate": 3.375e-05,
|
| 197 |
+
"loss": 0.5539,
|
| 198 |
+
"step": 27
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.268370607028754,
|
| 202 |
+
"grad_norm": 0.9213875490205139,
|
| 203 |
+
"learning_rate": 3.5000000000000004e-05,
|
| 204 |
+
"loss": 0.5319,
|
| 205 |
+
"step": 28
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.2779552715654952,
|
| 209 |
+
"grad_norm": 1.0198534952792746,
|
| 210 |
+
"learning_rate": 3.625e-05,
|
| 211 |
+
"loss": 0.5722,
|
| 212 |
+
"step": 29
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.28753993610223644,
|
| 216 |
+
"grad_norm": 0.9853545050808614,
|
| 217 |
+
"learning_rate": 3.7500000000000003e-05,
|
| 218 |
+
"loss": 0.5514,
|
| 219 |
+
"step": 30
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.2971246006389776,
|
| 223 |
+
"grad_norm": 0.9771316358011265,
|
| 224 |
+
"learning_rate": 3.875e-05,
|
| 225 |
+
"loss": 0.548,
|
| 226 |
+
"step": 31
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.30670926517571884,
|
| 230 |
+
"grad_norm": 1.070148811782629,
|
| 231 |
+
"learning_rate": 4e-05,
|
| 232 |
+
"loss": 0.5776,
|
| 233 |
+
"step": 32
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.31629392971246006,
|
| 237 |
+
"grad_norm": 0.8966401903811619,
|
| 238 |
+
"learning_rate": 3.9998741135094016e-05,
|
| 239 |
+
"loss": 0.5444,
|
| 240 |
+
"step": 33
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.3258785942492013,
|
| 244 |
+
"grad_norm": 0.893875729745141,
|
| 245 |
+
"learning_rate": 3.999496469885014e-05,
|
| 246 |
+
"loss": 0.5704,
|
| 247 |
+
"step": 34
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.3354632587859425,
|
| 251 |
+
"grad_norm": 1.072867766600089,
|
| 252 |
+
"learning_rate": 3.998867116667067e-05,
|
| 253 |
+
"loss": 0.5477,
|
| 254 |
+
"step": 35
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.3450479233226837,
|
| 258 |
+
"grad_norm": 0.9635968083722172,
|
| 259 |
+
"learning_rate": 3.9979861330826295e-05,
|
| 260 |
+
"loss": 0.5534,
|
| 261 |
+
"step": 36
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.3546325878594249,
|
| 265 |
+
"grad_norm": 0.9899242418062558,
|
| 266 |
+
"learning_rate": 3.996853630035634e-05,
|
| 267 |
+
"loss": 0.5655,
|
| 268 |
+
"step": 37
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.36421725239616615,
|
| 272 |
+
"grad_norm": 0.9130531711455715,
|
| 273 |
+
"learning_rate": 3.995469750092912e-05,
|
| 274 |
+
"loss": 0.5817,
|
| 275 |
+
"step": 38
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.3738019169329074,
|
| 279 |
+
"grad_norm": 0.9387442528138782,
|
| 280 |
+
"learning_rate": 3.9938346674662565e-05,
|
| 281 |
+
"loss": 0.5044,
|
| 282 |
+
"step": 39
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.38338658146964855,
|
| 286 |
+
"grad_norm": 1.0142174519180858,
|
| 287 |
+
"learning_rate": 3.991948587990479e-05,
|
| 288 |
+
"loss": 0.5359,
|
| 289 |
+
"step": 40
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.3929712460063898,
|
| 293 |
+
"grad_norm": 0.8959196832437484,
|
| 294 |
+
"learning_rate": 3.989811749097505e-05,
|
| 295 |
+
"loss": 0.5537,
|
| 296 |
+
"step": 41
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.402555910543131,
|
| 300 |
+
"grad_norm": 0.9987977928627828,
|
| 301 |
+
"learning_rate": 3.9874244197864856e-05,
|
| 302 |
+
"loss": 0.5782,
|
| 303 |
+
"step": 42
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.41214057507987223,
|
| 307 |
+
"grad_norm": 0.8845338538227933,
|
| 308 |
+
"learning_rate": 3.984786900589929e-05,
|
| 309 |
+
"loss": 0.5415,
|
| 310 |
+
"step": 43
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.4217252396166134,
|
| 314 |
+
"grad_norm": 1.0047800925961548,
|
| 315 |
+
"learning_rate": 3.98189952353587e-05,
|
| 316 |
+
"loss": 0.5542,
|
| 317 |
+
"step": 44
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.43130990415335463,
|
| 321 |
+
"grad_norm": 0.8022173762000696,
|
| 322 |
+
"learning_rate": 3.9787626521060736e-05,
|
| 323 |
+
"loss": 0.5483,
|
| 324 |
+
"step": 45
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.44089456869009586,
|
| 328 |
+
"grad_norm": 0.7883000207976826,
|
| 329 |
+
"learning_rate": 3.9753766811902756e-05,
|
| 330 |
+
"loss": 0.5473,
|
| 331 |
+
"step": 46
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.4504792332268371,
|
| 335 |
+
"grad_norm": 0.8434281567738428,
|
| 336 |
+
"learning_rate": 3.971742037036472e-05,
|
| 337 |
+
"loss": 0.5634,
|
| 338 |
+
"step": 47
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.46006389776357826,
|
| 342 |
+
"grad_norm": 1.0277496566918978,
|
| 343 |
+
"learning_rate": 3.96785917719726e-05,
|
| 344 |
+
"loss": 0.558,
|
| 345 |
+
"step": 48
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.4696485623003195,
|
| 349 |
+
"grad_norm": 0.99005810315587,
|
| 350 |
+
"learning_rate": 3.9637285904722376e-05,
|
| 351 |
+
"loss": 0.5508,
|
| 352 |
+
"step": 49
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.4792332268370607,
|
| 356 |
+
"grad_norm": 1.0782113201945098,
|
| 357 |
+
"learning_rate": 3.9593507968464714e-05,
|
| 358 |
+
"loss": 0.5478,
|
| 359 |
+
"step": 50
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.48881789137380194,
|
| 363 |
+
"grad_norm": 0.721872938484921,
|
| 364 |
+
"learning_rate": 3.9547263474250385e-05,
|
| 365 |
+
"loss": 0.5366,
|
| 366 |
+
"step": 51
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.4984025559105431,
|
| 370 |
+
"grad_norm": 1.2142224090633027,
|
| 371 |
+
"learning_rate": 3.949855824363647e-05,
|
| 372 |
+
"loss": 0.5544,
|
| 373 |
+
"step": 52
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.5079872204472844,
|
| 377 |
+
"grad_norm": 0.9185255274236219,
|
| 378 |
+
"learning_rate": 3.9447398407953536e-05,
|
| 379 |
+
"loss": 0.4976,
|
| 380 |
+
"step": 53
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.5175718849840255,
|
| 384 |
+
"grad_norm": 1.0876007980497202,
|
| 385 |
+
"learning_rate": 3.939379040753374e-05,
|
| 386 |
+
"loss": 0.5374,
|
| 387 |
+
"step": 54
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.5271565495207667,
|
| 391 |
+
"grad_norm": 1.2327514959394714,
|
| 392 |
+
"learning_rate": 3.933774099090013e-05,
|
| 393 |
+
"loss": 0.5517,
|
| 394 |
+
"step": 55
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.536741214057508,
|
| 398 |
+
"grad_norm": 0.6905831120274069,
|
| 399 |
+
"learning_rate": 3.927925721391707e-05,
|
| 400 |
+
"loss": 0.5114,
|
| 401 |
+
"step": 56
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.5463258785942492,
|
| 405 |
+
"grad_norm": 1.3327650152930188,
|
| 406 |
+
"learning_rate": 3.9218346438901996e-05,
|
| 407 |
+
"loss": 0.5556,
|
| 408 |
+
"step": 57
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.5559105431309904,
|
| 412 |
+
"grad_norm": 0.7267801369686653,
|
| 413 |
+
"learning_rate": 3.9155016333698615e-05,
|
| 414 |
+
"loss": 0.5446,
|
| 415 |
+
"step": 58
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.5654952076677316,
|
| 419 |
+
"grad_norm": 1.1332074133198213,
|
| 420 |
+
"learning_rate": 3.908927487071162e-05,
|
| 421 |
+
"loss": 0.5297,
|
| 422 |
+
"step": 59
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.5750798722044729,
|
| 426 |
+
"grad_norm": 1.1446700448302367,
|
| 427 |
+
"learning_rate": 3.9021130325903076e-05,
|
| 428 |
+
"loss": 0.5283,
|
| 429 |
+
"step": 60
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.5846645367412141,
|
| 433 |
+
"grad_norm": 0.9428641512869639,
|
| 434 |
+
"learning_rate": 3.895059127775058e-05,
|
| 435 |
+
"loss": 0.5276,
|
| 436 |
+
"step": 61
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.5942492012779552,
|
| 440 |
+
"grad_norm": 0.984662057057416,
|
| 441 |
+
"learning_rate": 3.8877666606167354e-05,
|
| 442 |
+
"loss": 0.5053,
|
| 443 |
+
"step": 62
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.6038338658146964,
|
| 447 |
+
"grad_norm": 0.9200152881682313,
|
| 448 |
+
"learning_rate": 3.880236549138438e-05,
|
| 449 |
+
"loss": 0.5019,
|
| 450 |
+
"step": 63
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.6134185303514377,
|
| 454 |
+
"grad_norm": 0.9735224575622512,
|
| 455 |
+
"learning_rate": 3.872469741279475e-05,
|
| 456 |
+
"loss": 0.5027,
|
| 457 |
+
"step": 64
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.6230031948881789,
|
| 461 |
+
"grad_norm": 0.8601738218870937,
|
| 462 |
+
"learning_rate": 3.8644672147760286e-05,
|
| 463 |
+
"loss": 0.5373,
|
| 464 |
+
"step": 65
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.6325878594249201,
|
| 468 |
+
"grad_norm": 1.0690694771471052,
|
| 469 |
+
"learning_rate": 3.856229977038078e-05,
|
| 470 |
+
"loss": 0.5141,
|
| 471 |
+
"step": 66
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.6421725239616614,
|
| 475 |
+
"grad_norm": 0.639345682465469,
|
| 476 |
+
"learning_rate": 3.8477590650225735e-05,
|
| 477 |
+
"loss": 0.5237,
|
| 478 |
+
"step": 67
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.6517571884984026,
|
| 482 |
+
"grad_norm": 0.9825658908653957,
|
| 483 |
+
"learning_rate": 3.839055545102902e-05,
|
| 484 |
+
"loss": 0.5218,
|
| 485 |
+
"step": 68
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.6613418530351438,
|
| 489 |
+
"grad_norm": 0.6640592278410514,
|
| 490 |
+
"learning_rate": 3.83012051293464e-05,
|
| 491 |
+
"loss": 0.5106,
|
| 492 |
+
"step": 69
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 0.670926517571885,
|
| 496 |
+
"grad_norm": 0.9942824925976919,
|
| 497 |
+
"learning_rate": 3.8209550933176324e-05,
|
| 498 |
+
"loss": 0.5371,
|
| 499 |
+
"step": 70
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.6805111821086262,
|
| 503 |
+
"grad_norm": 0.8321580496371637,
|
| 504 |
+
"learning_rate": 3.8115604400543885e-05,
|
| 505 |
+
"loss": 0.5285,
|
| 506 |
+
"step": 71
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.6900958466453674,
|
| 510 |
+
"grad_norm": 0.6720920559775677,
|
| 511 |
+
"learning_rate": 3.801937735804838e-05,
|
| 512 |
+
"loss": 0.5226,
|
| 513 |
+
"step": 72
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.6996805111821086,
|
| 517 |
+
"grad_norm": 0.869173718218821,
|
| 518 |
+
"learning_rate": 3.792088191937451e-05,
|
| 519 |
+
"loss": 0.5575,
|
| 520 |
+
"step": 73
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.7092651757188498,
|
| 524 |
+
"grad_norm": 0.9252932692964674,
|
| 525 |
+
"learning_rate": 3.782013048376736e-05,
|
| 526 |
+
"loss": 0.5331,
|
| 527 |
+
"step": 74
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.7188498402555911,
|
| 531 |
+
"grad_norm": 0.7617188628088486,
|
| 532 |
+
"learning_rate": 3.77171357344716e-05,
|
| 533 |
+
"loss": 0.4955,
|
| 534 |
+
"step": 75
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.7284345047923323,
|
| 538 |
+
"grad_norm": 0.8768762141880961,
|
| 539 |
+
"learning_rate": 3.761191063713476e-05,
|
| 540 |
+
"loss": 0.5319,
|
| 541 |
+
"step": 76
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.7380191693290735,
|
| 545 |
+
"grad_norm": 0.5856671731846674,
|
| 546 |
+
"learning_rate": 3.7504468438175076e-05,
|
| 547 |
+
"loss": 0.5286,
|
| 548 |
+
"step": 77
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.7476038338658147,
|
| 552 |
+
"grad_norm": 0.7357152016449541,
|
| 553 |
+
"learning_rate": 3.7394822663113915e-05,
|
| 554 |
+
"loss": 0.5202,
|
| 555 |
+
"step": 78
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.7571884984025559,
|
| 559 |
+
"grad_norm": 0.8421603022757149,
|
| 560 |
+
"learning_rate": 3.72829871148731e-05,
|
| 561 |
+
"loss": 0.5094,
|
| 562 |
+
"step": 79
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 0.7667731629392971,
|
| 566 |
+
"grad_norm": 0.7711330057528273,
|
| 567 |
+
"learning_rate": 3.716897587203733e-05,
|
| 568 |
+
"loss": 0.5067,
|
| 569 |
+
"step": 80
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.7763578274760383,
|
| 573 |
+
"grad_norm": 0.8061539763692872,
|
| 574 |
+
"learning_rate": 3.705280328708185e-05,
|
| 575 |
+
"loss": 0.5052,
|
| 576 |
+
"step": 81
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 0.7859424920127795,
|
| 580 |
+
"grad_norm": 0.9052464984501254,
|
| 581 |
+
"learning_rate": 3.6934483984565684e-05,
|
| 582 |
+
"loss": 0.5396,
|
| 583 |
+
"step": 82
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 0.7955271565495208,
|
| 587 |
+
"grad_norm": 0.7613495063825277,
|
| 588 |
+
"learning_rate": 3.681403285929061e-05,
|
| 589 |
+
"loss": 0.5372,
|
| 590 |
+
"step": 83
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.805111821086262,
|
| 594 |
+
"grad_norm": 0.9112267984327458,
|
| 595 |
+
"learning_rate": 3.669146507442606e-05,
|
| 596 |
+
"loss": 0.5022,
|
| 597 |
+
"step": 84
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 0.8146964856230032,
|
| 601 |
+
"grad_norm": 0.776862619807027,
|
| 602 |
+
"learning_rate": 3.6566796059600334e-05,
|
| 603 |
+
"loss": 0.4943,
|
| 604 |
+
"step": 85
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 0.8242811501597445,
|
| 608 |
+
"grad_norm": 0.6857530550183694,
|
| 609 |
+
"learning_rate": 3.644004150895821e-05,
|
| 610 |
+
"loss": 0.5137,
|
| 611 |
+
"step": 86
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 0.8338658146964856,
|
| 615 |
+
"grad_norm": 1.0001214073388247,
|
| 616 |
+
"learning_rate": 3.631121737918521e-05,
|
| 617 |
+
"loss": 0.5211,
|
| 618 |
+
"step": 87
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 0.8434504792332268,
|
| 622 |
+
"grad_norm": 0.614100200634229,
|
| 623 |
+
"learning_rate": 3.6180339887498953e-05,
|
| 624 |
+
"loss": 0.5426,
|
| 625 |
+
"step": 88
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.853035143769968,
|
| 629 |
+
"grad_norm": 0.7416467757559503,
|
| 630 |
+
"learning_rate": 3.6047425509607566e-05,
|
| 631 |
+
"loss": 0.4952,
|
| 632 |
+
"step": 89
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.8626198083067093,
|
| 636 |
+
"grad_norm": 0.8216759125475441,
|
| 637 |
+
"learning_rate": 3.591249097763562e-05,
|
| 638 |
+
"loss": 0.5101,
|
| 639 |
+
"step": 90
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.8722044728434505,
|
| 643 |
+
"grad_norm": 0.8783521934817947,
|
| 644 |
+
"learning_rate": 3.5775553278017824e-05,
|
| 645 |
+
"loss": 0.5298,
|
| 646 |
+
"step": 91
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 0.8817891373801917,
|
| 650 |
+
"grad_norm": 1.0305280265606676,
|
| 651 |
+
"learning_rate": 3.56366296493606e-05,
|
| 652 |
+
"loss": 0.509,
|
| 653 |
+
"step": 92
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 0.8913738019169329,
|
| 657 |
+
"grad_norm": 0.9394240012360305,
|
| 658 |
+
"learning_rate": 3.5495737580272024e-05,
|
| 659 |
+
"loss": 0.5078,
|
| 660 |
+
"step": 93
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 0.9009584664536742,
|
| 664 |
+
"grad_norm": 0.9309348675119059,
|
| 665 |
+
"learning_rate": 3.535289480716023e-05,
|
| 666 |
+
"loss": 0.5207,
|
| 667 |
+
"step": 94
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 0.9105431309904153,
|
| 671 |
+
"grad_norm": 0.9046584824077331,
|
| 672 |
+
"learning_rate": 3.520811931200063e-05,
|
| 673 |
+
"loss": 0.5141,
|
| 674 |
+
"step": 95
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 0.9201277955271565,
|
| 678 |
+
"grad_norm": 0.9752162620950977,
|
| 679 |
+
"learning_rate": 3.5061429320072225e-05,
|
| 680 |
+
"loss": 0.509,
|
| 681 |
+
"step": 96
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 0.9297124600638977,
|
| 685 |
+
"grad_norm": 0.6999864028990369,
|
| 686 |
+
"learning_rate": 3.4912843297663315e-05,
|
| 687 |
+
"loss": 0.5009,
|
| 688 |
+
"step": 97
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 0.939297124600639,
|
| 692 |
+
"grad_norm": 0.8545396872964545,
|
| 693 |
+
"learning_rate": 3.476237994974682e-05,
|
| 694 |
+
"loss": 0.5094,
|
| 695 |
+
"step": 98
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 0.9488817891373802,
|
| 699 |
+
"grad_norm": 0.7634132561961192,
|
| 700 |
+
"learning_rate": 3.4610058217625554e-05,
|
| 701 |
+
"loss": 0.5254,
|
| 702 |
+
"step": 99
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 0.9584664536741214,
|
| 706 |
+
"grad_norm": 0.7968935260475134,
|
| 707 |
+
"learning_rate": 3.4455897276547836e-05,
|
| 708 |
+
"loss": 0.5103,
|
| 709 |
+
"step": 100
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 0.9680511182108626,
|
| 713 |
+
"grad_norm": 0.8779383622683993,
|
| 714 |
+
"learning_rate": 3.429991653329351e-05,
|
| 715 |
+
"loss": 0.4962,
|
| 716 |
+
"step": 101
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 0.9776357827476039,
|
| 720 |
+
"grad_norm": 0.8873245389104287,
|
| 721 |
+
"learning_rate": 3.4142135623730954e-05,
|
| 722 |
+
"loss": 0.4886,
|
| 723 |
+
"step": 102
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 0.987220447284345,
|
| 727 |
+
"grad_norm": 0.6369979342430059,
|
| 728 |
+
"learning_rate": 3.398257441034515e-05,
|
| 729 |
+
"loss": 0.5153,
|
| 730 |
+
"step": 103
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 0.9968051118210862,
|
| 734 |
+
"grad_norm": 0.7723810447745028,
|
| 735 |
+
"learning_rate": 3.38212529797373e-05,
|
| 736 |
+
"loss": 0.5153,
|
| 737 |
+
"step": 104
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 1.0063897763578276,
|
| 741 |
+
"grad_norm": 1.3732822336657626,
|
| 742 |
+
"learning_rate": 3.365819164009614e-05,
|
| 743 |
+
"loss": 0.8678,
|
| 744 |
+
"step": 105
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 1.0159744408945688,
|
| 748 |
+
"grad_norm": 0.6491143685994553,
|
| 749 |
+
"learning_rate": 3.349341091864149e-05,
|
| 750 |
+
"loss": 0.3831,
|
| 751 |
+
"step": 106
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 1.0255591054313098,
|
| 755 |
+
"grad_norm": 0.7068223705318214,
|
| 756 |
+
"learning_rate": 3.3326931559040084e-05,
|
| 757 |
+
"loss": 0.4093,
|
| 758 |
+
"step": 107
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 1.035143769968051,
|
| 762 |
+
"grad_norm": 0.7448924108254381,
|
| 763 |
+
"learning_rate": 3.315877451879426e-05,
|
| 764 |
+
"loss": 0.4151,
|
| 765 |
+
"step": 108
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 1.0447284345047922,
|
| 769 |
+
"grad_norm": 0.6408577512276367,
|
| 770 |
+
"learning_rate": 3.298896096660367e-05,
|
| 771 |
+
"loss": 0.4591,
|
| 772 |
+
"step": 109
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 1.0543130990415335,
|
| 776 |
+
"grad_norm": 0.7245321395554075,
|
| 777 |
+
"learning_rate": 3.2817512279700486e-05,
|
| 778 |
+
"loss": 0.3995,
|
| 779 |
+
"step": 110
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 1.0638977635782747,
|
| 783 |
+
"grad_norm": 0.8050047056671462,
|
| 784 |
+
"learning_rate": 3.26444500411582e-05,
|
| 785 |
+
"loss": 0.4517,
|
| 786 |
+
"step": 111
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 1.073482428115016,
|
| 790 |
+
"grad_norm": 0.5692172276488292,
|
| 791 |
+
"learning_rate": 3.246979603717467e-05,
|
| 792 |
+
"loss": 0.3451,
|
| 793 |
+
"step": 112
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 1.0830670926517572,
|
| 797 |
+
"grad_norm": 0.7143343812753972,
|
| 798 |
+
"learning_rate": 3.2293572254329546e-05,
|
| 799 |
+
"loss": 0.4343,
|
| 800 |
+
"step": 113
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 1.0926517571884984,
|
| 804 |
+
"grad_norm": 0.621683888292599,
|
| 805 |
+
"learning_rate": 3.21158008768164e-05,
|
| 806 |
+
"loss": 0.4192,
|
| 807 |
+
"step": 114
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 1.1022364217252396,
|
| 811 |
+
"grad_norm": 0.7045370668340967,
|
| 812 |
+
"learning_rate": 3.1936504283650076e-05,
|
| 813 |
+
"loss": 0.4369,
|
| 814 |
+
"step": 115
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 1.1118210862619808,
|
| 818 |
+
"grad_norm": 0.689049088523779,
|
| 819 |
+
"learning_rate": 3.1755705045849465e-05,
|
| 820 |
+
"loss": 0.3995,
|
| 821 |
+
"step": 116
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 1.121405750798722,
|
| 825 |
+
"grad_norm": 0.6154467041297189,
|
| 826 |
+
"learning_rate": 3.157342592359612e-05,
|
| 827 |
+
"loss": 0.4309,
|
| 828 |
+
"step": 117
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 1.1309904153354633,
|
| 832 |
+
"grad_norm": 0.6721872707368349,
|
| 833 |
+
"learning_rate": 3.138968986336904e-05,
|
| 834 |
+
"loss": 0.3839,
|
| 835 |
+
"step": 118
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 1.1405750798722045,
|
| 839 |
+
"grad_norm": 0.6919770496482566,
|
| 840 |
+
"learning_rate": 3.1204519995056056e-05,
|
| 841 |
+
"loss": 0.4536,
|
| 842 |
+
"step": 119
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 1.1501597444089458,
|
| 846 |
+
"grad_norm": 0.548860031455983,
|
| 847 |
+
"learning_rate": 3.101793962904205e-05,
|
| 848 |
+
"loss": 0.3669,
|
| 849 |
+
"step": 120
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 1.159744408945687,
|
| 853 |
+
"grad_norm": 0.662490642467861,
|
| 854 |
+
"learning_rate": 3.082997225327452e-05,
|
| 855 |
+
"loss": 0.4413,
|
| 856 |
+
"step": 121
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 1.1693290734824282,
|
| 860 |
+
"grad_norm": 0.6083191080476268,
|
| 861 |
+
"learning_rate": 3.064064153030673e-05,
|
| 862 |
+
"loss": 0.3873,
|
| 863 |
+
"step": 122
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 1.1789137380191694,
|
| 867 |
+
"grad_norm": 0.6537310951440493,
|
| 868 |
+
"learning_rate": 3.0449971294318977e-05,
|
| 869 |
+
"loss": 0.4173,
|
| 870 |
+
"step": 123
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 1.1884984025559104,
|
| 874 |
+
"grad_norm": 0.6521406167992244,
|
| 875 |
+
"learning_rate": 3.0257985548118127e-05,
|
| 876 |
+
"loss": 0.4252,
|
| 877 |
+
"step": 124
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 1.1980830670926517,
|
| 881 |
+
"grad_norm": 0.6149682224088401,
|
| 882 |
+
"learning_rate": 3.0064708460116007e-05,
|
| 883 |
+
"loss": 0.4375,
|
| 884 |
+
"step": 125
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 1.207667731629393,
|
| 888 |
+
"grad_norm": 0.5660735007468374,
|
| 889 |
+
"learning_rate": 2.987016436128694e-05,
|
| 890 |
+
"loss": 0.4002,
|
| 891 |
+
"step": 126
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 1.2172523961661341,
|
| 895 |
+
"grad_norm": 0.6860797663501914,
|
| 896 |
+
"learning_rate": 2.9674377742104798e-05,
|
| 897 |
+
"loss": 0.4501,
|
| 898 |
+
"step": 127
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 1.2268370607028753,
|
| 902 |
+
"grad_norm": 0.5579864217961026,
|
| 903 |
+
"learning_rate": 2.9477373249459974e-05,
|
| 904 |
+
"loss": 0.3881,
|
| 905 |
+
"step": 128
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 1.2364217252396166,
|
| 909 |
+
"grad_norm": 0.7520213328572914,
|
| 910 |
+
"learning_rate": 2.9279175683556684e-05,
|
| 911 |
+
"loss": 0.4309,
|
| 912 |
+
"step": 129
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 1.2460063897763578,
|
| 916 |
+
"grad_norm": 0.5836183639678406,
|
| 917 |
+
"learning_rate": 2.9079809994790937e-05,
|
| 918 |
+
"loss": 0.4493,
|
| 919 |
+
"step": 130
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 1.255591054313099,
|
| 923 |
+
"grad_norm": 0.5454961635732396,
|
| 924 |
+
"learning_rate": 2.8879301280609645e-05,
|
| 925 |
+
"loss": 0.4248,
|
| 926 |
+
"step": 131
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 1.2651757188498403,
|
| 930 |
+
"grad_norm": 0.5814578805359331,
|
| 931 |
+
"learning_rate": 2.8677674782351164e-05,
|
| 932 |
+
"loss": 0.4133,
|
| 933 |
+
"step": 132
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 1.2747603833865815,
|
| 937 |
+
"grad_norm": 0.529954247172676,
|
| 938 |
+
"learning_rate": 2.8474955882067776e-05,
|
| 939 |
+
"loss": 0.4165,
|
| 940 |
+
"step": 133
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 1.2843450479233227,
|
| 944 |
+
"grad_norm": 0.5428271229133367,
|
| 945 |
+
"learning_rate": 2.8271170099330415e-05,
|
| 946 |
+
"loss": 0.4439,
|
| 947 |
+
"step": 134
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 1.293929712460064,
|
| 951 |
+
"grad_norm": 0.6847831612932209,
|
| 952 |
+
"learning_rate": 2.8066343088016105e-05,
|
| 953 |
+
"loss": 0.441,
|
| 954 |
+
"step": 135
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 1.3035143769968052,
|
| 958 |
+
"grad_norm": 0.49210190834226275,
|
| 959 |
+
"learning_rate": 2.7860500633078475e-05,
|
| 960 |
+
"loss": 0.4165,
|
| 961 |
+
"step": 136
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 1.3130990415335464,
|
| 965 |
+
"grad_norm": 0.6253827417860478,
|
| 966 |
+
"learning_rate": 2.7653668647301797e-05,
|
| 967 |
+
"loss": 0.4177,
|
| 968 |
+
"step": 137
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 1.3226837060702876,
|
| 972 |
+
"grad_norm": 0.5902351365403176,
|
| 973 |
+
"learning_rate": 2.7445873168038906e-05,
|
| 974 |
+
"loss": 0.418,
|
| 975 |
+
"step": 138
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 1.3322683706070286,
|
| 979 |
+
"grad_norm": 0.5490945288774098,
|
| 980 |
+
"learning_rate": 2.7237140353933445e-05,
|
| 981 |
+
"loss": 0.4352,
|
| 982 |
+
"step": 139
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 1.34185303514377,
|
| 986 |
+
"grad_norm": 0.6497645956557252,
|
| 987 |
+
"learning_rate": 2.7027496481626858e-05,
|
| 988 |
+
"loss": 0.442,
|
| 989 |
+
"step": 140
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 1.351437699680511,
|
| 993 |
+
"grad_norm": 0.542052646410243,
|
| 994 |
+
"learning_rate": 2.68169679424505e-05,
|
| 995 |
+
"loss": 0.3954,
|
| 996 |
+
"step": 141
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 1.3610223642172525,
|
| 1000 |
+
"grad_norm": 0.6469294947936336,
|
| 1001 |
+
"learning_rate": 2.6605581239103347e-05,
|
| 1002 |
+
"loss": 0.4324,
|
| 1003 |
+
"step": 142
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 1.3706070287539935,
|
| 1007 |
+
"grad_norm": 0.6114843722995497,
|
| 1008 |
+
"learning_rate": 2.6393362982315632e-05,
|
| 1009 |
+
"loss": 0.4086,
|
| 1010 |
+
"step": 143
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 1.3801916932907348,
|
| 1014 |
+
"grad_norm": 0.5302481010581994,
|
| 1015 |
+
"learning_rate": 2.618033988749895e-05,
|
| 1016 |
+
"loss": 0.3797,
|
| 1017 |
+
"step": 144
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 1.389776357827476,
|
| 1021 |
+
"grad_norm": 0.6290020988684742,
|
| 1022 |
+
"learning_rate": 2.5966538771383124e-05,
|
| 1023 |
+
"loss": 0.4616,
|
| 1024 |
+
"step": 145
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 1.3993610223642172,
|
| 1028 |
+
"grad_norm": 0.4520856565609043,
|
| 1029 |
+
"learning_rate": 2.5751986548640345e-05,
|
| 1030 |
+
"loss": 0.3691,
|
| 1031 |
+
"step": 146
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 1.4089456869009584,
|
| 1035 |
+
"grad_norm": 0.6229237189143199,
|
| 1036 |
+
"learning_rate": 2.5536710228496986e-05,
|
| 1037 |
+
"loss": 0.4428,
|
| 1038 |
+
"step": 147
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 1.4185303514376997,
|
| 1042 |
+
"grad_norm": 0.5985663011286853,
|
| 1043 |
+
"learning_rate": 2.5320736911333503e-05,
|
| 1044 |
+
"loss": 0.4101,
|
| 1045 |
+
"step": 148
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 1.428115015974441,
|
| 1049 |
+
"grad_norm": 0.5602348014424229,
|
| 1050 |
+
"learning_rate": 2.5104093785272854e-05,
|
| 1051 |
+
"loss": 0.3799,
|
| 1052 |
+
"step": 149
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 1.4376996805111821,
|
| 1056 |
+
"grad_norm": 0.5489525751898331,
|
| 1057 |
+
"learning_rate": 2.4886808122757882e-05,
|
| 1058 |
+
"loss": 0.4414,
|
| 1059 |
+
"step": 150
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 1.4472843450479234,
|
| 1063 |
+
"grad_norm": 0.624773105379871,
|
| 1064 |
+
"learning_rate": 2.4668907277118114e-05,
|
| 1065 |
+
"loss": 0.381,
|
| 1066 |
+
"step": 151
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 1.4568690095846646,
|
| 1070 |
+
"grad_norm": 0.5426555057346395,
|
| 1071 |
+
"learning_rate": 2.445041867912629e-05,
|
| 1072 |
+
"loss": 0.4539,
|
| 1073 |
+
"step": 152
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 1.4664536741214058,
|
| 1077 |
+
"grad_norm": 0.55536633578182,
|
| 1078 |
+
"learning_rate": 2.423136983354526e-05,
|
| 1079 |
+
"loss": 0.3802,
|
| 1080 |
+
"step": 153
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 1.476038338658147,
|
| 1084 |
+
"grad_norm": 0.4416725668124843,
|
| 1085 |
+
"learning_rate": 2.401178831566546e-05,
|
| 1086 |
+
"loss": 0.3748,
|
| 1087 |
+
"step": 154
|
| 1088 |
+
},
|
| 1089 |
+
{
|
| 1090 |
+
"epoch": 1.4856230031948883,
|
| 1091 |
+
"grad_norm": 0.6230029207177754,
|
| 1092 |
+
"learning_rate": 2.379170176783357e-05,
|
| 1093 |
+
"loss": 0.4455,
|
| 1094 |
+
"step": 155
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 1.4952076677316293,
|
| 1098 |
+
"grad_norm": 0.42530667418850676,
|
| 1099 |
+
"learning_rate": 2.3571137895972735e-05,
|
| 1100 |
+
"loss": 0.3917,
|
| 1101 |
+
"step": 156
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 1.5047923322683707,
|
| 1105 |
+
"grad_norm": 0.5190777897196971,
|
| 1106 |
+
"learning_rate": 2.335012446609473e-05,
|
| 1107 |
+
"loss": 0.3944,
|
| 1108 |
+
"step": 157
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"epoch": 1.5143769968051117,
|
| 1112 |
+
"grad_norm": 0.46663489738662295,
|
| 1113 |
+
"learning_rate": 2.312868930080462e-05,
|
| 1114 |
+
"loss": 0.4337,
|
| 1115 |
+
"step": 158
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 1.5239616613418532,
|
| 1119 |
+
"grad_norm": 0.4797855409160743,
|
| 1120 |
+
"learning_rate": 2.2906860275798257e-05,
|
| 1121 |
+
"loss": 0.3447,
|
| 1122 |
+
"step": 159
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 1.5335463258785942,
|
| 1126 |
+
"grad_norm": 0.48961810463783534,
|
| 1127 |
+
"learning_rate": 2.2684665316353112e-05,
|
| 1128 |
+
"loss": 0.4036,
|
| 1129 |
+
"step": 160
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 1.5431309904153354,
|
| 1133 |
+
"grad_norm": 0.5015403228340246,
|
| 1134 |
+
"learning_rate": 2.246213239381286e-05,
|
| 1135 |
+
"loss": 0.4494,
|
| 1136 |
+
"step": 161
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 1.5527156549520766,
|
| 1140 |
+
"grad_norm": 0.49826419920046305,
|
| 1141 |
+
"learning_rate": 2.2239289522066157e-05,
|
| 1142 |
+
"loss": 0.3662,
|
| 1143 |
+
"step": 162
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 1.5623003194888179,
|
| 1147 |
+
"grad_norm": 0.5982999217769431,
|
| 1148 |
+
"learning_rate": 2.201616475402009e-05,
|
| 1149 |
+
"loss": 0.4716,
|
| 1150 |
+
"step": 163
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 1.571884984025559,
|
| 1154 |
+
"grad_norm": 0.4282708816156415,
|
| 1155 |
+
"learning_rate": 2.179278617806867e-05,
|
| 1156 |
+
"loss": 0.3435,
|
| 1157 |
+
"step": 164
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 1.5814696485623003,
|
| 1161 |
+
"grad_norm": 0.4734658108408669,
|
| 1162 |
+
"learning_rate": 2.1569181914556904e-05,
|
| 1163 |
+
"loss": 0.4212,
|
| 1164 |
+
"step": 165
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 1.5910543130990416,
|
| 1168 |
+
"grad_norm": 0.49859752482578384,
|
| 1169 |
+
"learning_rate": 2.1345380112240796e-05,
|
| 1170 |
+
"loss": 0.425,
|
| 1171 |
+
"step": 166
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 1.6006389776357828,
|
| 1175 |
+
"grad_norm": 0.47157462374026493,
|
| 1176 |
+
"learning_rate": 2.1121408944743838e-05,
|
| 1177 |
+
"loss": 0.4507,
|
| 1178 |
+
"step": 167
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 1.610223642172524,
|
| 1182 |
+
"grad_norm": 0.4248405831748351,
|
| 1183 |
+
"learning_rate": 2.08972966070103e-05,
|
| 1184 |
+
"loss": 0.3602,
|
| 1185 |
+
"step": 168
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 1.619808306709265,
|
| 1189 |
+
"grad_norm": 0.5006569891631065,
|
| 1190 |
+
"learning_rate": 2.0673071311755885e-05,
|
| 1191 |
+
"loss": 0.4462,
|
| 1192 |
+
"step": 169
|
| 1193 |
+
},
|
| 1194 |
+
{
|
| 1195 |
+
"epoch": 1.6293929712460065,
|
| 1196 |
+
"grad_norm": 0.43452414054342864,
|
| 1197 |
+
"learning_rate": 2.0448761285916103e-05,
|
| 1198 |
+
"loss": 0.3398,
|
| 1199 |
+
"step": 170
|
| 1200 |
+
},
|
| 1201 |
+
{
|
| 1202 |
+
"epoch": 1.6389776357827475,
|
| 1203 |
+
"grad_norm": 0.5593870865885154,
|
| 1204 |
+
"learning_rate": 2.022439476709292e-05,
|
| 1205 |
+
"loss": 0.4578,
|
| 1206 |
+
"step": 171
|
| 1207 |
+
},
|
| 1208 |
+
{
|
| 1209 |
+
"epoch": 1.648562300319489,
|
| 1210 |
+
"grad_norm": 0.48660923787373483,
|
| 1211 |
+
"learning_rate": 2e-05,
|
| 1212 |
+
"loss": 0.402,
|
| 1213 |
+
"step": 172
|
| 1214 |
+
},
|
| 1215 |
+
{
|
| 1216 |
+
"epoch": 1.65814696485623,
|
| 1217 |
+
"grad_norm": 0.47042892474320597,
|
| 1218 |
+
"learning_rate": 1.9775605232907085e-05,
|
| 1219 |
+
"loss": 0.3732,
|
| 1220 |
+
"step": 173
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 1.6677316293929714,
|
| 1224 |
+
"grad_norm": 0.4637788148192955,
|
| 1225 |
+
"learning_rate": 1.9551238714083903e-05,
|
| 1226 |
+
"loss": 0.4164,
|
| 1227 |
+
"step": 174
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"epoch": 1.6773162939297124,
|
| 1231 |
+
"grad_norm": 0.5061846620195124,
|
| 1232 |
+
"learning_rate": 1.932692868824413e-05,
|
| 1233 |
+
"loss": 0.375,
|
| 1234 |
+
"step": 175
|
| 1235 |
+
},
|
| 1236 |
+
{
|
| 1237 |
+
"epoch": 1.6869009584664538,
|
| 1238 |
+
"grad_norm": 0.4389361045425349,
|
| 1239 |
+
"learning_rate": 1.910270339298971e-05,
|
| 1240 |
+
"loss": 0.4262,
|
| 1241 |
+
"step": 176
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 1.6964856230031948,
|
| 1245 |
+
"grad_norm": 0.4781379704434354,
|
| 1246 |
+
"learning_rate": 1.8878591055256165e-05,
|
| 1247 |
+
"loss": 0.4094,
|
| 1248 |
+
"step": 177
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 1.706070287539936,
|
| 1252 |
+
"grad_norm": 0.4719818469093975,
|
| 1253 |
+
"learning_rate": 1.8654619887759207e-05,
|
| 1254 |
+
"loss": 0.4398,
|
| 1255 |
+
"step": 178
|
| 1256 |
+
},
|
| 1257 |
+
{
|
| 1258 |
+
"epoch": 1.7156549520766773,
|
| 1259 |
+
"grad_norm": 0.4401976778891567,
|
| 1260 |
+
"learning_rate": 1.8430818085443106e-05,
|
| 1261 |
+
"loss": 0.3973,
|
| 1262 |
+
"step": 179
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 1.7252396166134185,
|
| 1266 |
+
"grad_norm": 0.5070123943848236,
|
| 1267 |
+
"learning_rate": 1.8207213821931332e-05,
|
| 1268 |
+
"loss": 0.4068,
|
| 1269 |
+
"step": 180
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 1.7348242811501597,
|
| 1273 |
+
"grad_norm": 0.4165356789504617,
|
| 1274 |
+
"learning_rate": 1.7983835245979914e-05,
|
| 1275 |
+
"loss": 0.4368,
|
| 1276 |
+
"step": 181
|
| 1277 |
+
},
|
| 1278 |
+
{
|
| 1279 |
+
"epoch": 1.744408945686901,
|
| 1280 |
+
"grad_norm": 0.5166587940461813,
|
| 1281 |
+
"learning_rate": 1.7760710477933846e-05,
|
| 1282 |
+
"loss": 0.4043,
|
| 1283 |
+
"step": 182
|
| 1284 |
+
},
|
| 1285 |
+
{
|
| 1286 |
+
"epoch": 1.7539936102236422,
|
| 1287 |
+
"grad_norm": 0.48269705466364915,
|
| 1288 |
+
"learning_rate": 1.7537867606187145e-05,
|
| 1289 |
+
"loss": 0.4118,
|
| 1290 |
+
"step": 183
|
| 1291 |
+
},
|
| 1292 |
+
{
|
| 1293 |
+
"epoch": 1.7635782747603834,
|
| 1294 |
+
"grad_norm": 0.47756180477333937,
|
| 1295 |
+
"learning_rate": 1.7315334683646898e-05,
|
| 1296 |
+
"loss": 0.42,
|
| 1297 |
+
"step": 184
|
| 1298 |
+
},
|
| 1299 |
+
{
|
| 1300 |
+
"epoch": 1.7731629392971247,
|
| 1301 |
+
"grad_norm": 0.4765695567575237,
|
| 1302 |
+
"learning_rate": 1.7093139724201753e-05,
|
| 1303 |
+
"loss": 0.4012,
|
| 1304 |
+
"step": 185
|
| 1305 |
+
},
|
| 1306 |
+
{
|
| 1307 |
+
"epoch": 1.7827476038338657,
|
| 1308 |
+
"grad_norm": 0.4681108878215689,
|
| 1309 |
+
"learning_rate": 1.687131069919538e-05,
|
| 1310 |
+
"loss": 0.41,
|
| 1311 |
+
"step": 186
|
| 1312 |
+
},
|
| 1313 |
+
{
|
| 1314 |
+
"epoch": 1.792332268370607,
|
| 1315 |
+
"grad_norm": 0.46037561395697485,
|
| 1316 |
+
"learning_rate": 1.6649875533905276e-05,
|
| 1317 |
+
"loss": 0.4043,
|
| 1318 |
+
"step": 187
|
| 1319 |
+
},
|
| 1320 |
+
{
|
| 1321 |
+
"epoch": 1.8019169329073481,
|
| 1322 |
+
"grad_norm": 0.419707165268871,
|
| 1323 |
+
"learning_rate": 1.642886210402727e-05,
|
| 1324 |
+
"loss": 0.3914,
|
| 1325 |
+
"step": 188
|
| 1326 |
+
},
|
| 1327 |
+
{
|
| 1328 |
+
"epoch": 1.8115015974440896,
|
| 1329 |
+
"grad_norm": 0.4619831244311854,
|
| 1330 |
+
"learning_rate": 1.620829823216643e-05,
|
| 1331 |
+
"loss": 0.4228,
|
| 1332 |
+
"step": 189
|
| 1333 |
+
},
|
| 1334 |
+
{
|
| 1335 |
+
"epoch": 1.8210862619808306,
|
| 1336 |
+
"grad_norm": 0.4361667434307363,
|
| 1337 |
+
"learning_rate": 1.5988211684334548e-05,
|
| 1338 |
+
"loss": 0.4047,
|
| 1339 |
+
"step": 190
|
| 1340 |
+
},
|
| 1341 |
+
{
|
| 1342 |
+
"epoch": 1.830670926517572,
|
| 1343 |
+
"grad_norm": 0.46079446208057,
|
| 1344 |
+
"learning_rate": 1.5768630166454746e-05,
|
| 1345 |
+
"loss": 0.4362,
|
| 1346 |
+
"step": 191
|
| 1347 |
+
},
|
| 1348 |
+
{
|
| 1349 |
+
"epoch": 1.840255591054313,
|
| 1350 |
+
"grad_norm": 0.4550050932615957,
|
| 1351 |
+
"learning_rate": 1.5549581320873715e-05,
|
| 1352 |
+
"loss": 0.4118,
|
| 1353 |
+
"step": 192
|
| 1354 |
+
},
|
| 1355 |
+
{
|
| 1356 |
+
"epoch": 1.8498402555910545,
|
| 1357 |
+
"grad_norm": 0.39293704953570213,
|
| 1358 |
+
"learning_rate": 1.53310927228819e-05,
|
| 1359 |
+
"loss": 0.407,
|
| 1360 |
+
"step": 193
|
| 1361 |
+
},
|
| 1362 |
+
{
|
| 1363 |
+
"epoch": 1.8594249201277955,
|
| 1364 |
+
"grad_norm": 0.4352517300933002,
|
| 1365 |
+
"learning_rate": 1.5113191877242116e-05,
|
| 1366 |
+
"loss": 0.3869,
|
| 1367 |
+
"step": 194
|
| 1368 |
+
},
|
| 1369 |
+
{
|
| 1370 |
+
"epoch": 1.8690095846645367,
|
| 1371 |
+
"grad_norm": 0.41513949179810855,
|
| 1372 |
+
"learning_rate": 1.4895906214727149e-05,
|
| 1373 |
+
"loss": 0.3755,
|
| 1374 |
+
"step": 195
|
| 1375 |
+
},
|
| 1376 |
+
{
|
| 1377 |
+
"epoch": 1.878594249201278,
|
| 1378 |
+
"grad_norm": 0.422581641804126,
|
| 1379 |
+
"learning_rate": 1.46792630886665e-05,
|
| 1380 |
+
"loss": 0.3707,
|
| 1381 |
+
"step": 196
|
| 1382 |
+
},
|
| 1383 |
+
{
|
| 1384 |
+
"epoch": 1.8881789137380192,
|
| 1385 |
+
"grad_norm": 0.4315158978443658,
|
| 1386 |
+
"learning_rate": 1.4463289771503015e-05,
|
| 1387 |
+
"loss": 0.4135,
|
| 1388 |
+
"step": 197
|
| 1389 |
+
},
|
| 1390 |
+
{
|
| 1391 |
+
"epoch": 1.8977635782747604,
|
| 1392 |
+
"grad_norm": 0.4439035469190941,
|
| 1393 |
+
"learning_rate": 1.4248013451359657e-05,
|
| 1394 |
+
"loss": 0.4299,
|
| 1395 |
+
"step": 198
|
| 1396 |
+
},
|
| 1397 |
+
{
|
| 1398 |
+
"epoch": 1.9073482428115016,
|
| 1399 |
+
"grad_norm": 0.38114332649556365,
|
| 1400 |
+
"learning_rate": 1.403346122861688e-05,
|
| 1401 |
+
"loss": 0.3986,
|
| 1402 |
+
"step": 199
|
| 1403 |
+
},
|
| 1404 |
+
{
|
| 1405 |
+
"epoch": 1.9169329073482428,
|
| 1406 |
+
"grad_norm": 0.4562893320122979,
|
| 1407 |
+
"learning_rate": 1.3819660112501054e-05,
|
| 1408 |
+
"loss": 0.4447,
|
| 1409 |
+
"step": 200
|
| 1410 |
+
},
|
| 1411 |
+
{
|
| 1412 |
+
"epoch": 1.926517571884984,
|
| 1413 |
+
"grad_norm": 0.41109636995050963,
|
| 1414 |
+
"learning_rate": 1.3606637017684375e-05,
|
| 1415 |
+
"loss": 0.4231,
|
| 1416 |
+
"step": 201
|
| 1417 |
+
},
|
| 1418 |
+
{
|
| 1419 |
+
"epoch": 1.9361022364217253,
|
| 1420 |
+
"grad_norm": 0.4108514256627659,
|
| 1421 |
+
"learning_rate": 1.3394418760896665e-05,
|
| 1422 |
+
"loss": 0.3919,
|
| 1423 |
+
"step": 202
|
| 1424 |
+
},
|
| 1425 |
+
{
|
| 1426 |
+
"epoch": 1.9456869009584663,
|
| 1427 |
+
"grad_norm": 0.4403482836797348,
|
| 1428 |
+
"learning_rate": 1.31830320575495e-05,
|
| 1429 |
+
"loss": 0.4004,
|
| 1430 |
+
"step": 203
|
| 1431 |
+
},
|
| 1432 |
+
{
|
| 1433 |
+
"epoch": 1.9552715654952078,
|
| 1434 |
+
"grad_norm": 0.46058376777126453,
|
| 1435 |
+
"learning_rate": 1.2972503518373145e-05,
|
| 1436 |
+
"loss": 0.4271,
|
| 1437 |
+
"step": 204
|
| 1438 |
+
},
|
| 1439 |
+
{
|
| 1440 |
+
"epoch": 1.9648562300319488,
|
| 1441 |
+
"grad_norm": 0.4549563896458377,
|
| 1442 |
+
"learning_rate": 1.2762859646066561e-05,
|
| 1443 |
+
"loss": 0.4155,
|
| 1444 |
+
"step": 205
|
| 1445 |
+
},
|
| 1446 |
+
{
|
| 1447 |
+
"epoch": 1.9744408945686902,
|
| 1448 |
+
"grad_norm": 0.47102947639065273,
|
| 1449 |
+
"learning_rate": 1.2554126831961097e-05,
|
| 1450 |
+
"loss": 0.3947,
|
| 1451 |
+
"step": 206
|
| 1452 |
+
},
|
| 1453 |
+
{
|
| 1454 |
+
"epoch": 1.9840255591054312,
|
| 1455 |
+
"grad_norm": 0.43255866772480006,
|
| 1456 |
+
"learning_rate": 1.2346331352698206e-05,
|
| 1457 |
+
"loss": 0.3965,
|
| 1458 |
+
"step": 207
|
| 1459 |
+
},
|
| 1460 |
+
{
|
| 1461 |
+
"epoch": 1.9936102236421727,
|
| 1462 |
+
"grad_norm": 0.6055961125329625,
|
| 1463 |
+
"learning_rate": 1.213949936692153e-05,
|
| 1464 |
+
"loss": 0.4308,
|
| 1465 |
+
"step": 208
|
| 1466 |
+
},
|
| 1467 |
+
{
|
| 1468 |
+
"epoch": 2.0031948881789137,
|
| 1469 |
+
"grad_norm": 0.8201258360584602,
|
| 1470 |
+
"learning_rate": 1.1933656911983901e-05,
|
| 1471 |
+
"loss": 0.5536,
|
| 1472 |
+
"step": 209
|
| 1473 |
+
},
|
| 1474 |
+
{
|
| 1475 |
+
"epoch": 2.012779552715655,
|
| 1476 |
+
"grad_norm": 0.5116622937539458,
|
| 1477 |
+
"learning_rate": 1.1728829900669592e-05,
|
| 1478 |
+
"loss": 0.3256,
|
| 1479 |
+
"step": 210
|
| 1480 |
+
},
|
| 1481 |
+
{
|
| 1482 |
+
"epoch": 2.022364217252396,
|
| 1483 |
+
"grad_norm": 0.6977084341791524,
|
| 1484 |
+
"learning_rate": 1.1525044117932227e-05,
|
| 1485 |
+
"loss": 0.3275,
|
| 1486 |
+
"step": 211
|
| 1487 |
+
},
|
| 1488 |
+
{
|
| 1489 |
+
"epoch": 2.0319488817891376,
|
| 1490 |
+
"grad_norm": 0.639393679219998,
|
| 1491 |
+
"learning_rate": 1.132232521764884e-05,
|
| 1492 |
+
"loss": 0.3009,
|
| 1493 |
+
"step": 212
|
| 1494 |
+
},
|
| 1495 |
+
{
|
| 1496 |
+
"epoch": 2.0415335463258786,
|
| 1497 |
+
"grad_norm": 0.4890609363134351,
|
| 1498 |
+
"learning_rate": 1.1120698719390362e-05,
|
| 1499 |
+
"loss": 0.3387,
|
| 1500 |
+
"step": 213
|
| 1501 |
+
},
|
| 1502 |
+
{
|
| 1503 |
+
"epoch": 2.0511182108626196,
|
| 1504 |
+
"grad_norm": 0.5588000853620018,
|
| 1505 |
+
"learning_rate": 1.0920190005209066e-05,
|
| 1506 |
+
"loss": 0.3105,
|
| 1507 |
+
"step": 214
|
| 1508 |
+
},
|
| 1509 |
+
{
|
| 1510 |
+
"epoch": 2.060702875399361,
|
| 1511 |
+
"grad_norm": 0.5036006567654044,
|
| 1512 |
+
"learning_rate": 1.0720824316443321e-05,
|
| 1513 |
+
"loss": 0.2901,
|
| 1514 |
+
"step": 215
|
| 1515 |
+
},
|
| 1516 |
+
{
|
| 1517 |
+
"epoch": 2.070287539936102,
|
| 1518 |
+
"grad_norm": 0.5915822002117069,
|
| 1519 |
+
"learning_rate": 1.0522626750540029e-05,
|
| 1520 |
+
"loss": 0.3493,
|
| 1521 |
+
"step": 216
|
| 1522 |
+
},
|
| 1523 |
+
{
|
| 1524 |
+
"epoch": 2.0798722044728435,
|
| 1525 |
+
"grad_norm": 0.4214007118584772,
|
| 1526 |
+
"learning_rate": 1.0325622257895205e-05,
|
| 1527 |
+
"loss": 0.2985,
|
| 1528 |
+
"step": 217
|
| 1529 |
+
},
|
| 1530 |
+
{
|
| 1531 |
+
"epoch": 2.0894568690095845,
|
| 1532 |
+
"grad_norm": 0.5295101270711986,
|
| 1533 |
+
"learning_rate": 1.0129835638713064e-05,
|
| 1534 |
+
"loss": 0.3166,
|
| 1535 |
+
"step": 218
|
| 1536 |
+
},
|
| 1537 |
+
{
|
| 1538 |
+
"epoch": 2.099041533546326,
|
| 1539 |
+
"grad_norm": 0.4906090405908181,
|
| 1540 |
+
"learning_rate": 9.935291539884e-06,
|
| 1541 |
+
"loss": 0.2843,
|
| 1542 |
+
"step": 219
|
| 1543 |
+
},
|
| 1544 |
+
{
|
| 1545 |
+
"epoch": 2.108626198083067,
|
| 1546 |
+
"grad_norm": 0.3938053903767187,
|
| 1547 |
+
"learning_rate": 9.74201445188188e-06,
|
| 1548 |
+
"loss": 0.3163,
|
| 1549 |
+
"step": 220
|
| 1550 |
+
},
|
| 1551 |
+
{
|
| 1552 |
+
"epoch": 2.1182108626198084,
|
| 1553 |
+
"grad_norm": 0.5012038895473048,
|
| 1554 |
+
"learning_rate": 9.550028705681024e-06,
|
| 1555 |
+
"loss": 0.3222,
|
| 1556 |
+
"step": 221
|
| 1557 |
+
},
|
| 1558 |
+
{
|
| 1559 |
+
"epoch": 2.1277955271565494,
|
| 1560 |
+
"grad_norm": 0.4321669775385389,
|
| 1561 |
+
"learning_rate": 9.359358469693272e-06,
|
| 1562 |
+
"loss": 0.2909,
|
| 1563 |
+
"step": 222
|
| 1564 |
+
},
|
| 1565 |
+
{
|
| 1566 |
+
"epoch": 2.137380191693291,
|
| 1567 |
+
"grad_norm": 0.4117178329226722,
|
| 1568 |
+
"learning_rate": 9.170027746725487e-06,
|
| 1569 |
+
"loss": 0.3146,
|
| 1570 |
+
"step": 223
|
| 1571 |
+
},
|
| 1572 |
+
{
|
| 1573 |
+
"epoch": 2.146964856230032,
|
| 1574 |
+
"grad_norm": 0.3793388916643685,
|
| 1575 |
+
"learning_rate": 8.982060370957953e-06,
|
| 1576 |
+
"loss": 0.3017,
|
| 1577 |
+
"step": 224
|
| 1578 |
+
},
|
| 1579 |
+
{
|
| 1580 |
+
"epoch": 2.1565495207667733,
|
| 1581 |
+
"grad_norm": 0.4535610427107228,
|
| 1582 |
+
"learning_rate": 8.795480004943946e-06,
|
| 1583 |
+
"loss": 0.3456,
|
| 1584 |
+
"step": 225
|
| 1585 |
+
},
|
| 1586 |
+
{
|
| 1587 |
+
"epoch": 2.1661341853035143,
|
| 1588 |
+
"grad_norm": 0.37244536401873557,
|
| 1589 |
+
"learning_rate": 8.610310136630962e-06,
|
| 1590 |
+
"loss": 0.2761,
|
| 1591 |
+
"step": 226
|
| 1592 |
+
},
|
| 1593 |
+
{
|
| 1594 |
+
"epoch": 2.1757188498402558,
|
| 1595 |
+
"grad_norm": 0.3905670548801518,
|
| 1596 |
+
"learning_rate": 8.426574076403887e-06,
|
| 1597 |
+
"loss": 0.302,
|
| 1598 |
+
"step": 227
|
| 1599 |
+
},
|
| 1600 |
+
{
|
| 1601 |
+
"epoch": 2.1853035143769968,
|
| 1602 |
+
"grad_norm": 0.4359210822644894,
|
| 1603 |
+
"learning_rate": 8.24429495415054e-06,
|
| 1604 |
+
"loss": 0.338,
|
| 1605 |
+
"step": 228
|
| 1606 |
+
},
|
| 1607 |
+
{
|
| 1608 |
+
"epoch": 2.194888178913738,
|
| 1609 |
+
"grad_norm": 0.35920435170508774,
|
| 1610 |
+
"learning_rate": 8.063495716349929e-06,
|
| 1611 |
+
"loss": 0.3081,
|
| 1612 |
+
"step": 229
|
| 1613 |
+
},
|
| 1614 |
+
{
|
| 1615 |
+
"epoch": 2.2044728434504792,
|
| 1616 |
+
"grad_norm": 0.37621369347346134,
|
| 1617 |
+
"learning_rate": 7.884199123183604e-06,
|
| 1618 |
+
"loss": 0.3419,
|
| 1619 |
+
"step": 230
|
| 1620 |
+
},
|
| 1621 |
+
{
|
| 1622 |
+
"epoch": 2.2140575079872207,
|
| 1623 |
+
"grad_norm": 0.3458946393058899,
|
| 1624 |
+
"learning_rate": 7.706427745670458e-06,
|
| 1625 |
+
"loss": 0.2749,
|
| 1626 |
+
"step": 231
|
| 1627 |
+
},
|
| 1628 |
+
{
|
| 1629 |
+
"epoch": 2.2236421725239617,
|
| 1630 |
+
"grad_norm": 0.4279444858711213,
|
| 1631 |
+
"learning_rate": 7.530203962825331e-06,
|
| 1632 |
+
"loss": 0.3365,
|
| 1633 |
+
"step": 232
|
| 1634 |
+
},
|
| 1635 |
+
{
|
| 1636 |
+
"epoch": 2.2332268370607027,
|
| 1637 |
+
"grad_norm": 0.3940413928415214,
|
| 1638 |
+
"learning_rate": 7.355549958841808e-06,
|
| 1639 |
+
"loss": 0.3282,
|
| 1640 |
+
"step": 233
|
| 1641 |
+
},
|
| 1642 |
+
{
|
| 1643 |
+
"epoch": 2.242811501597444,
|
| 1644 |
+
"grad_norm": 0.39687410854343524,
|
| 1645 |
+
"learning_rate": 7.182487720299518e-06,
|
| 1646 |
+
"loss": 0.2943,
|
| 1647 |
+
"step": 234
|
| 1648 |
+
},
|
| 1649 |
+
{
|
| 1650 |
+
"epoch": 2.252396166134185,
|
| 1651 |
+
"grad_norm": 0.42285032823633234,
|
| 1652 |
+
"learning_rate": 7.01103903339633e-06,
|
| 1653 |
+
"loss": 0.3252,
|
| 1654 |
+
"step": 235
|
| 1655 |
+
},
|
| 1656 |
+
{
|
| 1657 |
+
"epoch": 2.2619808306709266,
|
| 1658 |
+
"grad_norm": 0.3730435843457758,
|
| 1659 |
+
"learning_rate": 6.841225481205749e-06,
|
| 1660 |
+
"loss": 0.2755,
|
| 1661 |
+
"step": 236
|
| 1662 |
+
},
|
| 1663 |
+
{
|
| 1664 |
+
"epoch": 2.2715654952076676,
|
| 1665 |
+
"grad_norm": 0.35400532023912695,
|
| 1666 |
+
"learning_rate": 6.6730684409599225e-06,
|
| 1667 |
+
"loss": 0.2962,
|
| 1668 |
+
"step": 237
|
| 1669 |
+
},
|
| 1670 |
+
{
|
| 1671 |
+
"epoch": 2.281150159744409,
|
| 1672 |
+
"grad_norm": 0.39975466800088416,
|
| 1673 |
+
"learning_rate": 6.5065890813585145e-06,
|
| 1674 |
+
"loss": 0.2951,
|
| 1675 |
+
"step": 238
|
| 1676 |
+
},
|
| 1677 |
+
{
|
| 1678 |
+
"epoch": 2.29073482428115,
|
| 1679 |
+
"grad_norm": 0.4291346122689877,
|
| 1680 |
+
"learning_rate": 6.3418083599038624e-06,
|
| 1681 |
+
"loss": 0.317,
|
| 1682 |
+
"step": 239
|
| 1683 |
+
},
|
| 1684 |
+
{
|
| 1685 |
+
"epoch": 2.3003194888178915,
|
| 1686 |
+
"grad_norm": 0.3209273222551871,
|
| 1687 |
+
"learning_rate": 6.178747020262708e-06,
|
| 1688 |
+
"loss": 0.2881,
|
| 1689 |
+
"step": 240
|
| 1690 |
+
},
|
| 1691 |
+
{
|
| 1692 |
+
"epoch": 2.3099041533546325,
|
| 1693 |
+
"grad_norm": 0.3305467433463152,
|
| 1694 |
+
"learning_rate": 6.017425589654853e-06,
|
| 1695 |
+
"loss": 0.2595,
|
| 1696 |
+
"step": 241
|
| 1697 |
+
},
|
| 1698 |
+
{
|
| 1699 |
+
"epoch": 2.319488817891374,
|
| 1700 |
+
"grad_norm": 0.38111354545103865,
|
| 1701 |
+
"learning_rate": 5.857864376269051e-06,
|
| 1702 |
+
"loss": 0.326,
|
| 1703 |
+
"step": 242
|
| 1704 |
+
},
|
| 1705 |
+
{
|
| 1706 |
+
"epoch": 2.329073482428115,
|
| 1707 |
+
"grad_norm": 0.30011533883033215,
|
| 1708 |
+
"learning_rate": 5.700083466706494e-06,
|
| 1709 |
+
"loss": 0.2744,
|
| 1710 |
+
"step": 243
|
| 1711 |
+
},
|
| 1712 |
+
{
|
| 1713 |
+
"epoch": 2.3386581469648564,
|
| 1714 |
+
"grad_norm": 0.3813259748607558,
|
| 1715 |
+
"learning_rate": 5.544102723452171e-06,
|
| 1716 |
+
"loss": 0.3105,
|
| 1717 |
+
"step": 244
|
| 1718 |
+
},
|
| 1719 |
+
{
|
| 1720 |
+
"epoch": 2.3482428115015974,
|
| 1721 |
+
"grad_norm": 0.35123991131982374,
|
| 1722 |
+
"learning_rate": 5.38994178237445e-06,
|
| 1723 |
+
"loss": 0.316,
|
| 1724 |
+
"step": 245
|
| 1725 |
+
},
|
| 1726 |
+
{
|
| 1727 |
+
"epoch": 2.357827476038339,
|
| 1728 |
+
"grad_norm": 0.32456708820534624,
|
| 1729 |
+
"learning_rate": 5.237620050253189e-06,
|
| 1730 |
+
"loss": 0.3226,
|
| 1731 |
+
"step": 246
|
| 1732 |
+
},
|
| 1733 |
+
{
|
| 1734 |
+
"epoch": 2.36741214057508,
|
| 1735 |
+
"grad_norm": 0.2976838087469898,
|
| 1736 |
+
"learning_rate": 5.087156702336689e-06,
|
| 1737 |
+
"loss": 0.2614,
|
| 1738 |
+
"step": 247
|
| 1739 |
+
},
|
| 1740 |
+
{
|
| 1741 |
+
"epoch": 2.376996805111821,
|
| 1742 |
+
"grad_norm": 0.36552843906148524,
|
| 1743 |
+
"learning_rate": 4.938570679927784e-06,
|
| 1744 |
+
"loss": 0.3123,
|
| 1745 |
+
"step": 248
|
| 1746 |
+
},
|
| 1747 |
+
{
|
| 1748 |
+
"epoch": 2.3865814696485623,
|
| 1749 |
+
"grad_norm": 0.35095503173458703,
|
| 1750 |
+
"learning_rate": 4.791880687999382e-06,
|
| 1751 |
+
"loss": 0.3123,
|
| 1752 |
+
"step": 249
|
| 1753 |
+
},
|
| 1754 |
+
{
|
| 1755 |
+
"epoch": 2.3961661341853033,
|
| 1756 |
+
"grad_norm": 0.30625640312787195,
|
| 1757 |
+
"learning_rate": 4.647105192839778e-06,
|
| 1758 |
+
"loss": 0.3176,
|
| 1759 |
+
"step": 250
|
| 1760 |
+
},
|
| 1761 |
+
{
|
| 1762 |
+
"epoch": 2.405750798722045,
|
| 1763 |
+
"grad_norm": 0.341761533375559,
|
| 1764 |
+
"learning_rate": 4.504262419727983e-06,
|
| 1765 |
+
"loss": 0.3144,
|
| 1766 |
+
"step": 251
|
| 1767 |
+
},
|
| 1768 |
+
{
|
| 1769 |
+
"epoch": 2.415335463258786,
|
| 1770 |
+
"grad_norm": 0.3533342249857026,
|
| 1771 |
+
"learning_rate": 4.363370350639405e-06,
|
| 1772 |
+
"loss": 0.34,
|
| 1773 |
+
"step": 252
|
| 1774 |
+
},
|
| 1775 |
+
{
|
| 1776 |
+
"epoch": 2.4249201277955272,
|
| 1777 |
+
"grad_norm": 0.34883281084308354,
|
| 1778 |
+
"learning_rate": 4.2244467219821806e-06,
|
| 1779 |
+
"loss": 0.2955,
|
| 1780 |
+
"step": 253
|
| 1781 |
+
},
|
| 1782 |
+
{
|
| 1783 |
+
"epoch": 2.4345047923322682,
|
| 1784 |
+
"grad_norm": 0.32463619347820316,
|
| 1785 |
+
"learning_rate": 4.087509022364382e-06,
|
| 1786 |
+
"loss": 0.283,
|
| 1787 |
+
"step": 254
|
| 1788 |
+
},
|
| 1789 |
+
{
|
| 1790 |
+
"epoch": 2.4440894568690097,
|
| 1791 |
+
"grad_norm": 0.33770338798840704,
|
| 1792 |
+
"learning_rate": 3.952574490392443e-06,
|
| 1793 |
+
"loss": 0.3177,
|
| 1794 |
+
"step": 255
|
| 1795 |
+
},
|
| 1796 |
+
{
|
| 1797 |
+
"epoch": 2.4536741214057507,
|
| 1798 |
+
"grad_norm": 0.3110029435092563,
|
| 1799 |
+
"learning_rate": 3.819660112501053e-06,
|
| 1800 |
+
"loss": 0.302,
|
| 1801 |
+
"step": 256
|
| 1802 |
+
},
|
| 1803 |
+
{
|
| 1804 |
+
"epoch": 2.463258785942492,
|
| 1805 |
+
"grad_norm": 0.30800590517974463,
|
| 1806 |
+
"learning_rate": 3.6887826208147968e-06,
|
| 1807 |
+
"loss": 0.302,
|
| 1808 |
+
"step": 257
|
| 1809 |
+
},
|
| 1810 |
+
{
|
| 1811 |
+
"epoch": 2.472843450479233,
|
| 1812 |
+
"grad_norm": 0.3232681352037157,
|
| 1813 |
+
"learning_rate": 3.5599584910418037e-06,
|
| 1814 |
+
"loss": 0.2646,
|
| 1815 |
+
"step": 258
|
| 1816 |
+
},
|
| 1817 |
+
{
|
| 1818 |
+
"epoch": 2.4824281150159746,
|
| 1819 |
+
"grad_norm": 0.36421296660907726,
|
| 1820 |
+
"learning_rate": 3.433203940399672e-06,
|
| 1821 |
+
"loss": 0.3084,
|
| 1822 |
+
"step": 259
|
| 1823 |
+
},
|
| 1824 |
+
{
|
| 1825 |
+
"epoch": 2.4920127795527156,
|
| 1826 |
+
"grad_norm": 0.3831030312811903,
|
| 1827 |
+
"learning_rate": 3.3085349255739475e-06,
|
| 1828 |
+
"loss": 0.3469,
|
| 1829 |
+
"step": 260
|
| 1830 |
+
},
|
| 1831 |
+
{
|
| 1832 |
+
"epoch": 2.501597444089457,
|
| 1833 |
+
"grad_norm": 0.3325867758576896,
|
| 1834 |
+
"learning_rate": 3.1859671407093984e-06,
|
| 1835 |
+
"loss": 0.3093,
|
| 1836 |
+
"step": 261
|
| 1837 |
+
},
|
| 1838 |
+
{
|
| 1839 |
+
"epoch": 2.511182108626198,
|
| 1840 |
+
"grad_norm": 0.337706654419708,
|
| 1841 |
+
"learning_rate": 3.0655160154343177e-06,
|
| 1842 |
+
"loss": 0.2979,
|
| 1843 |
+
"step": 262
|
| 1844 |
+
},
|
| 1845 |
+
{
|
| 1846 |
+
"epoch": 2.520766773162939,
|
| 1847 |
+
"grad_norm": 0.32967438884893885,
|
| 1848 |
+
"learning_rate": 2.947196712918157e-06,
|
| 1849 |
+
"loss": 0.3047,
|
| 1850 |
+
"step": 263
|
| 1851 |
+
},
|
| 1852 |
+
{
|
| 1853 |
+
"epoch": 2.5303514376996805,
|
| 1854 |
+
"grad_norm": 0.3233433151169985,
|
| 1855 |
+
"learning_rate": 2.8310241279626784e-06,
|
| 1856 |
+
"loss": 0.2969,
|
| 1857 |
+
"step": 264
|
| 1858 |
+
},
|
| 1859 |
+
{
|
| 1860 |
+
"epoch": 2.539936102236422,
|
| 1861 |
+
"grad_norm": 0.31301087906685726,
|
| 1862 |
+
"learning_rate": 2.7170128851269084e-06,
|
| 1863 |
+
"loss": 0.2981,
|
| 1864 |
+
"step": 265
|
| 1865 |
+
},
|
| 1866 |
+
{
|
| 1867 |
+
"epoch": 2.549520766773163,
|
| 1868 |
+
"grad_norm": 0.32786149056284464,
|
| 1869 |
+
"learning_rate": 2.6051773368860935e-06,
|
| 1870 |
+
"loss": 0.3308,
|
| 1871 |
+
"step": 266
|
| 1872 |
+
},
|
| 1873 |
+
{
|
| 1874 |
+
"epoch": 2.559105431309904,
|
| 1875 |
+
"grad_norm": 0.31611139290200363,
|
| 1876 |
+
"learning_rate": 2.4955315618249263e-06,
|
| 1877 |
+
"loss": 0.3134,
|
| 1878 |
+
"step": 267
|
| 1879 |
+
},
|
| 1880 |
+
{
|
| 1881 |
+
"epoch": 2.5686900958466454,
|
| 1882 |
+
"grad_norm": 0.3202278737140233,
|
| 1883 |
+
"learning_rate": 2.38808936286524e-06,
|
| 1884 |
+
"loss": 0.2988,
|
| 1885 |
+
"step": 268
|
| 1886 |
+
},
|
| 1887 |
+
{
|
| 1888 |
+
"epoch": 2.5782747603833864,
|
| 1889 |
+
"grad_norm": 0.3187169467092785,
|
| 1890 |
+
"learning_rate": 2.2828642655284038e-06,
|
| 1891 |
+
"loss": 0.275,
|
| 1892 |
+
"step": 269
|
| 1893 |
+
},
|
| 1894 |
+
{
|
| 1895 |
+
"epoch": 2.587859424920128,
|
| 1896 |
+
"grad_norm": 0.31150312167717303,
|
| 1897 |
+
"learning_rate": 2.1798695162326444e-06,
|
| 1898 |
+
"loss": 0.2913,
|
| 1899 |
+
"step": 270
|
| 1900 |
+
},
|
| 1901 |
+
{
|
| 1902 |
+
"epoch": 2.597444089456869,
|
| 1903 |
+
"grad_norm": 0.3199602869029221,
|
| 1904 |
+
"learning_rate": 2.0791180806254975e-06,
|
| 1905 |
+
"loss": 0.3142,
|
| 1906 |
+
"step": 271
|
| 1907 |
+
},
|
| 1908 |
+
{
|
| 1909 |
+
"epoch": 2.6070287539936103,
|
| 1910 |
+
"grad_norm": 0.2859711870729773,
|
| 1911 |
+
"learning_rate": 1.9806226419516195e-06,
|
| 1912 |
+
"loss": 0.2994,
|
| 1913 |
+
"step": 272
|
| 1914 |
+
},
|
| 1915 |
+
{
|
| 1916 |
+
"epoch": 2.6166134185303513,
|
| 1917 |
+
"grad_norm": 0.31034616192733694,
|
| 1918 |
+
"learning_rate": 1.8843955994561191e-06,
|
| 1919 |
+
"loss": 0.3301,
|
| 1920 |
+
"step": 273
|
| 1921 |
+
},
|
| 1922 |
+
{
|
| 1923 |
+
"epoch": 2.626198083067093,
|
| 1924 |
+
"grad_norm": 0.2911594390655468,
|
| 1925 |
+
"learning_rate": 1.790449066823683e-06,
|
| 1926 |
+
"loss": 0.3136,
|
| 1927 |
+
"step": 274
|
| 1928 |
+
},
|
| 1929 |
+
{
|
| 1930 |
+
"epoch": 2.635782747603834,
|
| 1931 |
+
"grad_norm": 0.31846174945632705,
|
| 1932 |
+
"learning_rate": 1.6987948706536038e-06,
|
| 1933 |
+
"loss": 0.3265,
|
| 1934 |
+
"step": 275
|
| 1935 |
+
},
|
| 1936 |
+
{
|
| 1937 |
+
"epoch": 2.6453674121405752,
|
| 1938 |
+
"grad_norm": 0.3091371745510634,
|
| 1939 |
+
"learning_rate": 1.6094445489709886e-06,
|
| 1940 |
+
"loss": 0.3116,
|
| 1941 |
+
"step": 276
|
| 1942 |
+
},
|
| 1943 |
+
{
|
| 1944 |
+
"epoch": 2.6549520766773163,
|
| 1945 |
+
"grad_norm": 0.28744483162457096,
|
| 1946 |
+
"learning_rate": 1.5224093497742654e-06,
|
| 1947 |
+
"loss": 0.3081,
|
| 1948 |
+
"step": 277
|
| 1949 |
+
},
|
| 1950 |
+
{
|
| 1951 |
+
"epoch": 2.6645367412140573,
|
| 1952 |
+
"grad_norm": 0.3010104381158104,
|
| 1953 |
+
"learning_rate": 1.4377002296192233e-06,
|
| 1954 |
+
"loss": 0.2914,
|
| 1955 |
+
"step": 278
|
| 1956 |
+
},
|
| 1957 |
+
{
|
| 1958 |
+
"epoch": 2.6741214057507987,
|
| 1959 |
+
"grad_norm": 0.3184703269379148,
|
| 1960 |
+
"learning_rate": 1.3553278522397162e-06,
|
| 1961 |
+
"loss": 0.3398,
|
| 1962 |
+
"step": 279
|
| 1963 |
+
},
|
| 1964 |
+
{
|
| 1965 |
+
"epoch": 2.68370607028754,
|
| 1966 |
+
"grad_norm": 0.3012745996881748,
|
| 1967 |
+
"learning_rate": 1.275302587205256e-06,
|
| 1968 |
+
"loss": 0.2895,
|
| 1969 |
+
"step": 280
|
| 1970 |
+
},
|
| 1971 |
+
{
|
| 1972 |
+
"epoch": 2.693290734824281,
|
| 1973 |
+
"grad_norm": 0.32944896697723586,
|
| 1974 |
+
"learning_rate": 1.1976345086156193e-06,
|
| 1975 |
+
"loss": 0.3205,
|
| 1976 |
+
"step": 281
|
| 1977 |
+
},
|
| 1978 |
+
{
|
| 1979 |
+
"epoch": 2.702875399361022,
|
| 1980 |
+
"grad_norm": 0.30329250730790236,
|
| 1981 |
+
"learning_rate": 1.1223333938326486e-06,
|
| 1982 |
+
"loss": 0.307,
|
| 1983 |
+
"step": 282
|
| 1984 |
+
},
|
| 1985 |
+
{
|
| 1986 |
+
"epoch": 2.7124600638977636,
|
| 1987 |
+
"grad_norm": 0.31477266554231703,
|
| 1988 |
+
"learning_rate": 1.0494087222494253e-06,
|
| 1989 |
+
"loss": 0.2977,
|
| 1990 |
+
"step": 283
|
| 1991 |
+
},
|
| 1992 |
+
{
|
| 1993 |
+
"epoch": 2.722044728434505,
|
| 1994 |
+
"grad_norm": 0.27218751575717026,
|
| 1995 |
+
"learning_rate": 9.788696740969295e-07,
|
| 1996 |
+
"loss": 0.2727,
|
| 1997 |
+
"step": 284
|
| 1998 |
+
},
|
| 1999 |
+
{
|
| 2000 |
+
"epoch": 2.731629392971246,
|
| 2001 |
+
"grad_norm": 0.2957347661302385,
|
| 2002 |
+
"learning_rate": 9.107251292883856e-07,
|
| 2003 |
+
"loss": 0.3241,
|
| 2004 |
+
"step": 285
|
| 2005 |
+
},
|
| 2006 |
+
{
|
| 2007 |
+
"epoch": 2.741214057507987,
|
| 2008 |
+
"grad_norm": 0.2754457767022288,
|
| 2009 |
+
"learning_rate": 8.44983666301391e-07,
|
| 2010 |
+
"loss": 0.2888,
|
| 2011 |
+
"step": 286
|
| 2012 |
+
},
|
| 2013 |
+
{
|
| 2014 |
+
"epoch": 2.7507987220447285,
|
| 2015 |
+
"grad_norm": 0.3009061607925657,
|
| 2016 |
+
"learning_rate": 7.816535610980103e-07,
|
| 2017 |
+
"loss": 0.3236,
|
| 2018 |
+
"step": 287
|
| 2019 |
+
},
|
| 2020 |
+
{
|
| 2021 |
+
"epoch": 2.7603833865814695,
|
| 2022 |
+
"grad_norm": 0.3042692164321764,
|
| 2023 |
+
"learning_rate": 7.207427860829352e-07,
|
| 2024 |
+
"loss": 0.3293,
|
| 2025 |
+
"step": 288
|
| 2026 |
+
},
|
| 2027 |
+
{
|
| 2028 |
+
"epoch": 2.769968051118211,
|
| 2029 |
+
"grad_norm": 0.3084699080559463,
|
| 2030 |
+
"learning_rate": 6.622590090998727e-07,
|
| 2031 |
+
"loss": 0.2896,
|
| 2032 |
+
"step": 289
|
| 2033 |
+
},
|
| 2034 |
+
{
|
| 2035 |
+
"epoch": 2.779552715654952,
|
| 2036 |
+
"grad_norm": 0.2874746995613232,
|
| 2037 |
+
"learning_rate": 6.062095924662625e-07,
|
| 2038 |
+
"loss": 0.2977,
|
| 2039 |
+
"step": 290
|
| 2040 |
+
},
|
| 2041 |
+
{
|
| 2042 |
+
"epoch": 2.7891373801916934,
|
| 2043 |
+
"grad_norm": 0.30432287110598966,
|
| 2044 |
+
"learning_rate": 5.526015920464689e-07,
|
| 2045 |
+
"loss": 0.3527,
|
| 2046 |
+
"step": 291
|
| 2047 |
+
},
|
| 2048 |
+
{
|
| 2049 |
+
"epoch": 2.7987220447284344,
|
| 2050 |
+
"grad_norm": 0.2752619163287045,
|
| 2051 |
+
"learning_rate": 5.014417563635276e-07,
|
| 2052 |
+
"loss": 0.2575,
|
| 2053 |
+
"step": 292
|
| 2054 |
+
},
|
| 2055 |
+
{
|
| 2056 |
+
"epoch": 2.8083067092651754,
|
| 2057 |
+
"grad_norm": 0.3000832642958543,
|
| 2058 |
+
"learning_rate": 4.5273652574961745e-07,
|
| 2059 |
+
"loss": 0.3026,
|
| 2060 |
+
"step": 293
|
| 2061 |
+
},
|
| 2062 |
+
{
|
| 2063 |
+
"epoch": 2.817891373801917,
|
| 2064 |
+
"grad_norm": 0.28459336557371145,
|
| 2065 |
+
"learning_rate": 4.064920315352905e-07,
|
| 2066 |
+
"loss": 0.2953,
|
| 2067 |
+
"step": 294
|
| 2068 |
+
},
|
| 2069 |
+
{
|
| 2070 |
+
"epoch": 2.8274760383386583,
|
| 2071 |
+
"grad_norm": 0.3403049604495616,
|
| 2072 |
+
"learning_rate": 3.62714095277632e-07,
|
| 2073 |
+
"loss": 0.3698,
|
| 2074 |
+
"step": 295
|
| 2075 |
+
},
|
| 2076 |
+
{
|
| 2077 |
+
"epoch": 2.8370607028753994,
|
| 2078 |
+
"grad_norm": 0.28503876136131184,
|
| 2079 |
+
"learning_rate": 3.214082280274067e-07,
|
| 2080 |
+
"loss": 0.2709,
|
| 2081 |
+
"step": 296
|
| 2082 |
+
},
|
| 2083 |
+
{
|
| 2084 |
+
"epoch": 2.8466453674121404,
|
| 2085 |
+
"grad_norm": 0.299085461383954,
|
| 2086 |
+
"learning_rate": 2.825796296352823e-07,
|
| 2087 |
+
"loss": 0.3268,
|
| 2088 |
+
"step": 297
|
| 2089 |
+
},
|
| 2090 |
+
{
|
| 2091 |
+
"epoch": 2.856230031948882,
|
| 2092 |
+
"grad_norm": 0.3016256794104278,
|
| 2093 |
+
"learning_rate": 2.462331880972468e-07,
|
| 2094 |
+
"loss": 0.3032,
|
| 2095 |
+
"step": 298
|
| 2096 |
+
},
|
| 2097 |
+
{
|
| 2098 |
+
"epoch": 2.8658146964856233,
|
| 2099 |
+
"grad_norm": 0.45115419737664014,
|
| 2100 |
+
"learning_rate": 2.123734789392673e-07,
|
| 2101 |
+
"loss": 0.3539,
|
| 2102 |
+
"step": 299
|
| 2103 |
+
},
|
| 2104 |
+
{
|
| 2105 |
+
"epoch": 2.8753993610223643,
|
| 2106 |
+
"grad_norm": 0.2661772363746791,
|
| 2107 |
+
"learning_rate": 1.81004764641306e-07,
|
| 2108 |
+
"loss": 0.2696,
|
| 2109 |
+
"step": 300
|
| 2110 |
+
},
|
| 2111 |
+
{
|
| 2112 |
+
"epoch": 2.8849840255591053,
|
| 2113 |
+
"grad_norm": 0.2901415907606709,
|
| 2114 |
+
"learning_rate": 1.5213099410071873e-07,
|
| 2115 |
+
"loss": 0.3199,
|
| 2116 |
+
"step": 301
|
| 2117 |
+
},
|
| 2118 |
+
{
|
| 2119 |
+
"epoch": 2.8945686900958467,
|
| 2120 |
+
"grad_norm": 0.29953435371037257,
|
| 2121 |
+
"learning_rate": 1.2575580213514792e-07,
|
| 2122 |
+
"loss": 0.3403,
|
| 2123 |
+
"step": 302
|
| 2124 |
+
},
|
| 2125 |
+
{
|
| 2126 |
+
"epoch": 2.9041533546325877,
|
| 2127 |
+
"grad_norm": 0.28555862760239925,
|
| 2128 |
+
"learning_rate": 1.0188250902495312e-07,
|
| 2129 |
+
"loss": 0.3211,
|
| 2130 |
+
"step": 303
|
| 2131 |
+
},
|
| 2132 |
+
{
|
| 2133 |
+
"epoch": 2.913738019169329,
|
| 2134 |
+
"grad_norm": 0.3026928327893977,
|
| 2135 |
+
"learning_rate": 8.051412009521864e-08,
|
| 2136 |
+
"loss": 0.3422,
|
| 2137 |
+
"step": 304
|
| 2138 |
+
},
|
| 2139 |
+
{
|
| 2140 |
+
"epoch": 2.92332268370607,
|
| 2141 |
+
"grad_norm": 0.27586171376383506,
|
| 2142 |
+
"learning_rate": 6.165332533744072e-08,
|
| 2143 |
+
"loss": 0.2635,
|
| 2144 |
+
"step": 305
|
| 2145 |
+
},
|
| 2146 |
+
{
|
| 2147 |
+
"epoch": 2.9329073482428116,
|
| 2148 |
+
"grad_norm": 0.2797989784483068,
|
| 2149 |
+
"learning_rate": 4.530249907087836e-08,
|
| 2150 |
+
"loss": 0.2942,
|
| 2151 |
+
"step": 306
|
| 2152 |
+
},
|
| 2153 |
+
{
|
| 2154 |
+
"epoch": 2.9424920127795526,
|
| 2155 |
+
"grad_norm": 0.32210333279055925,
|
| 2156 |
+
"learning_rate": 3.146369964366791e-08,
|
| 2157 |
+
"loss": 0.3829,
|
| 2158 |
+
"step": 307
|
| 2159 |
+
},
|
| 2160 |
+
{
|
| 2161 |
+
"epoch": 2.952076677316294,
|
| 2162 |
+
"grad_norm": 0.27055679198155164,
|
| 2163 |
+
"learning_rate": 2.0138669173708213e-08,
|
| 2164 |
+
"loss": 0.2838,
|
| 2165 |
+
"step": 308
|
| 2166 |
+
},
|
| 2167 |
+
{
|
| 2168 |
+
"epoch": 2.961661341853035,
|
| 2169 |
+
"grad_norm": 0.28816307097851795,
|
| 2170 |
+
"learning_rate": 1.1328833329333767e-08,
|
| 2171 |
+
"loss": 0.3031,
|
| 2172 |
+
"step": 309
|
| 2173 |
+
},
|
| 2174 |
+
{
|
| 2175 |
+
"epoch": 2.9712460063897765,
|
| 2176 |
+
"grad_norm": 0.28899005094993563,
|
| 2177 |
+
"learning_rate": 5.0353011498693875e-09,
|
| 2178 |
+
"loss": 0.3185,
|
| 2179 |
+
"step": 310
|
| 2180 |
+
},
|
| 2181 |
+
{
|
| 2182 |
+
"epoch": 2.9808306709265175,
|
| 2183 |
+
"grad_norm": 0.26574618706072956,
|
| 2184 |
+
"learning_rate": 1.2588649059885883e-09,
|
| 2185 |
+
"loss": 0.2968,
|
| 2186 |
+
"step": 311
|
| 2187 |
+
},
|
| 2188 |
+
{
|
| 2189 |
+
"epoch": 2.9904153354632586,
|
| 2190 |
+
"grad_norm": 0.28718668323400676,
|
| 2191 |
+
"learning_rate": 0.0,
|
| 2192 |
+
"loss": 0.3009,
|
| 2193 |
+
"step": 312
|
| 2194 |
+
},
|
| 2195 |
+
{
|
| 2196 |
+
"epoch": 2.9904153354632586,
|
| 2197 |
+
"step": 312,
|
| 2198 |
+
"total_flos": 4.4115474684667494e+17,
|
| 2199 |
+
"train_loss": 0.43267787887881964,
|
| 2200 |
+
"train_runtime": 10181.5896,
|
| 2201 |
+
"train_samples_per_second": 2.946,
|
| 2202 |
+
"train_steps_per_second": 0.031
|
| 2203 |
+
}
|
| 2204 |
+
],
|
| 2205 |
+
"logging_steps": 1.0,
|
| 2206 |
+
"max_steps": 312,
|
| 2207 |
+
"num_input_tokens_seen": 0,
|
| 2208 |
+
"num_train_epochs": 3,
|
| 2209 |
+
"save_steps": 500,
|
| 2210 |
+
"stateful_callbacks": {
|
| 2211 |
+
"TrainerControl": {
|
| 2212 |
+
"args": {
|
| 2213 |
+
"should_epoch_stop": false,
|
| 2214 |
+
"should_evaluate": false,
|
| 2215 |
+
"should_log": false,
|
| 2216 |
+
"should_save": true,
|
| 2217 |
+
"should_training_stop": true
|
| 2218 |
+
},
|
| 2219 |
+
"attributes": {}
|
| 2220 |
+
}
|
| 2221 |
+
},
|
| 2222 |
+
"total_flos": 4.4115474684667494e+17,
|
| 2223 |
+
"train_batch_size": 1,
|
| 2224 |
+
"trial_name": null,
|
| 2225 |
+
"trial_params": null
|
| 2226 |
+
}
|
training_loss.png
ADDED
|