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_lr2e5_epochs3
|
|
@@ -15,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 15 |
|
| 16 |
# 10k_globalbatchsize96_lr2e5_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_lr2e5_epochs3
|
|
|
|
| 16 |
|
| 17 |
# 10k_globalbatchsize96_lr2e5_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": 0.46701200955953354,
|
| 5 |
+
"train_runtime": 9975.6836,
|
| 6 |
+
"train_samples_per_second": 3.006,
|
| 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": 0.46701200955953354,
|
| 5 |
+
"train_runtime": 9975.6836,
|
| 6 |
+
"train_samples_per_second": 3.006,
|
| 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": 5.816951093345819,
|
| 14 |
+
"learning_rate": 6.25e-07,
|
| 15 |
+
"loss": 0.8809,
|
| 16 |
+
"step": 1
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.019169329073482427,
|
| 20 |
+
"grad_norm": 5.964814555570492,
|
| 21 |
+
"learning_rate": 1.25e-06,
|
| 22 |
+
"loss": 0.8476,
|
| 23 |
+
"step": 2
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.02875399361022364,
|
| 27 |
+
"grad_norm": 5.91437668793362,
|
| 28 |
+
"learning_rate": 1.8750000000000003e-06,
|
| 29 |
+
"loss": 0.8771,
|
| 30 |
+
"step": 3
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.038338658146964855,
|
| 34 |
+
"grad_norm": 5.7274742814752155,
|
| 35 |
+
"learning_rate": 2.5e-06,
|
| 36 |
+
"loss": 0.8549,
|
| 37 |
+
"step": 4
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.04792332268370607,
|
| 41 |
+
"grad_norm": 5.324202754304387,
|
| 42 |
+
"learning_rate": 3.125e-06,
|
| 43 |
+
"loss": 0.8394,
|
| 44 |
+
"step": 5
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.05750798722044728,
|
| 48 |
+
"grad_norm": 4.070433381694975,
|
| 49 |
+
"learning_rate": 3.7500000000000005e-06,
|
| 50 |
+
"loss": 0.8099,
|
| 51 |
+
"step": 6
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.0670926517571885,
|
| 55 |
+
"grad_norm": 2.4484114306405496,
|
| 56 |
+
"learning_rate": 4.3750000000000005e-06,
|
| 57 |
+
"loss": 0.7524,
|
| 58 |
+
"step": 7
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.07667731629392971,
|
| 62 |
+
"grad_norm": 2.2317547400553917,
|
| 63 |
+
"learning_rate": 5e-06,
|
| 64 |
+
"loss": 0.7533,
|
| 65 |
+
"step": 8
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.08626198083067092,
|
| 69 |
+
"grad_norm": 4.021700030471694,
|
| 70 |
+
"learning_rate": 5.625e-06,
|
| 71 |
+
"loss": 0.7731,
|
| 72 |
+
"step": 9
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.09584664536741214,
|
| 76 |
+
"grad_norm": 4.242475862537413,
|
| 77 |
+
"learning_rate": 6.25e-06,
|
| 78 |
+
"loss": 0.7965,
|
| 79 |
+
"step": 10
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.10543130990415335,
|
| 83 |
+
"grad_norm": 4.187322997258876,
|
| 84 |
+
"learning_rate": 6.875e-06,
|
| 85 |
+
"loss": 0.7696,
|
| 86 |
+
"step": 11
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.11501597444089456,
|
| 90 |
+
"grad_norm": 3.5880075046075364,
|
| 91 |
+
"learning_rate": 7.500000000000001e-06,
|
| 92 |
+
"loss": 0.7105,
|
| 93 |
+
"step": 12
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.12460063897763578,
|
| 97 |
+
"grad_norm": 3.5147425543441346,
|
| 98 |
+
"learning_rate": 8.125000000000001e-06,
|
| 99 |
+
"loss": 0.7182,
|
| 100 |
+
"step": 13
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.134185303514377,
|
| 104 |
+
"grad_norm": 2.698461280108951,
|
| 105 |
+
"learning_rate": 8.750000000000001e-06,
|
| 106 |
+
"loss": 0.6811,
|
| 107 |
+
"step": 14
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.14376996805111822,
|
| 111 |
+
"grad_norm": 1.890037499452564,
|
| 112 |
+
"learning_rate": 9.375000000000001e-06,
|
| 113 |
+
"loss": 0.6807,
|
| 114 |
+
"step": 15
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.15335463258785942,
|
| 118 |
+
"grad_norm": 1.831321882805547,
|
| 119 |
+
"learning_rate": 1e-05,
|
| 120 |
+
"loss": 0.696,
|
| 121 |
+
"step": 16
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.16293929712460065,
|
| 125 |
+
"grad_norm": 2.061073210204071,
|
| 126 |
+
"learning_rate": 1.0625e-05,
|
| 127 |
+
"loss": 0.6692,
|
| 128 |
+
"step": 17
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.17252396166134185,
|
| 132 |
+
"grad_norm": 1.8734300870572917,
|
| 133 |
+
"learning_rate": 1.125e-05,
|
| 134 |
+
"loss": 0.6655,
|
| 135 |
+
"step": 18
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.18210862619808307,
|
| 139 |
+
"grad_norm": 1.3757902351252964,
|
| 140 |
+
"learning_rate": 1.1875e-05,
|
| 141 |
+
"loss": 0.641,
|
| 142 |
+
"step": 19
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.19169329073482427,
|
| 146 |
+
"grad_norm": 1.3308745862219742,
|
| 147 |
+
"learning_rate": 1.25e-05,
|
| 148 |
+
"loss": 0.6173,
|
| 149 |
+
"step": 20
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.2012779552715655,
|
| 153 |
+
"grad_norm": 1.3502732169912994,
|
| 154 |
+
"learning_rate": 1.3125e-05,
|
| 155 |
+
"loss": 0.6143,
|
| 156 |
+
"step": 21
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.2108626198083067,
|
| 160 |
+
"grad_norm": 1.179130020860865,
|
| 161 |
+
"learning_rate": 1.375e-05,
|
| 162 |
+
"loss": 0.6165,
|
| 163 |
+
"step": 22
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.22044728434504793,
|
| 167 |
+
"grad_norm": 1.1153340921244757,
|
| 168 |
+
"learning_rate": 1.4375e-05,
|
| 169 |
+
"loss": 0.595,
|
| 170 |
+
"step": 23
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.23003194888178913,
|
| 174 |
+
"grad_norm": 0.93782430241953,
|
| 175 |
+
"learning_rate": 1.5000000000000002e-05,
|
| 176 |
+
"loss": 0.6081,
|
| 177 |
+
"step": 24
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.23961661341853036,
|
| 181 |
+
"grad_norm": 1.0101672937842767,
|
| 182 |
+
"learning_rate": 1.5625e-05,
|
| 183 |
+
"loss": 0.5846,
|
| 184 |
+
"step": 25
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.24920127795527156,
|
| 188 |
+
"grad_norm": 0.8994475569314913,
|
| 189 |
+
"learning_rate": 1.6250000000000002e-05,
|
| 190 |
+
"loss": 0.5652,
|
| 191 |
+
"step": 26
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.25878594249201275,
|
| 195 |
+
"grad_norm": 0.7372445592319721,
|
| 196 |
+
"learning_rate": 1.6875e-05,
|
| 197 |
+
"loss": 0.569,
|
| 198 |
+
"step": 27
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.268370607028754,
|
| 202 |
+
"grad_norm": 0.7790129120230198,
|
| 203 |
+
"learning_rate": 1.7500000000000002e-05,
|
| 204 |
+
"loss": 0.5454,
|
| 205 |
+
"step": 28
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.2779552715654952,
|
| 209 |
+
"grad_norm": 0.79836828642288,
|
| 210 |
+
"learning_rate": 1.8125e-05,
|
| 211 |
+
"loss": 0.5854,
|
| 212 |
+
"step": 29
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.28753993610223644,
|
| 216 |
+
"grad_norm": 0.797607856474264,
|
| 217 |
+
"learning_rate": 1.8750000000000002e-05,
|
| 218 |
+
"loss": 0.5628,
|
| 219 |
+
"step": 30
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.2971246006389776,
|
| 223 |
+
"grad_norm": 0.8495331973366639,
|
| 224 |
+
"learning_rate": 1.9375e-05,
|
| 225 |
+
"loss": 0.5589,
|
| 226 |
+
"step": 31
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.30670926517571884,
|
| 230 |
+
"grad_norm": 0.8939715492547688,
|
| 231 |
+
"learning_rate": 2e-05,
|
| 232 |
+
"loss": 0.588,
|
| 233 |
+
"step": 32
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.31629392971246006,
|
| 237 |
+
"grad_norm": 0.7945325544236974,
|
| 238 |
+
"learning_rate": 1.9999370567547008e-05,
|
| 239 |
+
"loss": 0.5531,
|
| 240 |
+
"step": 33
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.3258785942492013,
|
| 244 |
+
"grad_norm": 0.75081365184287,
|
| 245 |
+
"learning_rate": 1.999748234942507e-05,
|
| 246 |
+
"loss": 0.5781,
|
| 247 |
+
"step": 34
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.3354632587859425,
|
| 251 |
+
"grad_norm": 0.7667797396040631,
|
| 252 |
+
"learning_rate": 1.9994335583335336e-05,
|
| 253 |
+
"loss": 0.5532,
|
| 254 |
+
"step": 35
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.3450479233226837,
|
| 258 |
+
"grad_norm": 0.6810165168871454,
|
| 259 |
+
"learning_rate": 1.9989930665413148e-05,
|
| 260 |
+
"loss": 0.5592,
|
| 261 |
+
"step": 36
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.3546325878594249,
|
| 265 |
+
"grad_norm": 0.8020192657808985,
|
| 266 |
+
"learning_rate": 1.998426815017817e-05,
|
| 267 |
+
"loss": 0.5703,
|
| 268 |
+
"step": 37
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.36421725239616615,
|
| 272 |
+
"grad_norm": 0.765862886699684,
|
| 273 |
+
"learning_rate": 1.997734875046456e-05,
|
| 274 |
+
"loss": 0.5859,
|
| 275 |
+
"step": 38
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.3738019169329074,
|
| 279 |
+
"grad_norm": 0.6206240716622619,
|
| 280 |
+
"learning_rate": 1.9969173337331283e-05,
|
| 281 |
+
"loss": 0.5084,
|
| 282 |
+
"step": 39
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.38338658146964855,
|
| 286 |
+
"grad_norm": 0.6800428463570417,
|
| 287 |
+
"learning_rate": 1.9959742939952393e-05,
|
| 288 |
+
"loss": 0.5402,
|
| 289 |
+
"step": 40
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.3929712460063898,
|
| 293 |
+
"grad_norm": 0.6898204949273792,
|
| 294 |
+
"learning_rate": 1.9949058745487524e-05,
|
| 295 |
+
"loss": 0.5576,
|
| 296 |
+
"step": 41
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.402555910543131,
|
| 300 |
+
"grad_norm": 0.7038921388379388,
|
| 301 |
+
"learning_rate": 1.9937122098932428e-05,
|
| 302 |
+
"loss": 0.5814,
|
| 303 |
+
"step": 42
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.41214057507987223,
|
| 307 |
+
"grad_norm": 0.6386669258124261,
|
| 308 |
+
"learning_rate": 1.9923934502949645e-05,
|
| 309 |
+
"loss": 0.5433,
|
| 310 |
+
"step": 43
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.4217252396166134,
|
| 314 |
+
"grad_norm": 0.8544716563932903,
|
| 315 |
+
"learning_rate": 1.990949761767935e-05,
|
| 316 |
+
"loss": 0.5562,
|
| 317 |
+
"step": 44
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.43130990415335463,
|
| 321 |
+
"grad_norm": 0.6590987393801254,
|
| 322 |
+
"learning_rate": 1.9893813260530368e-05,
|
| 323 |
+
"loss": 0.5507,
|
| 324 |
+
"step": 45
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.44089456869009586,
|
| 328 |
+
"grad_norm": 0.7325945250857884,
|
| 329 |
+
"learning_rate": 1.9876883405951378e-05,
|
| 330 |
+
"loss": 0.5472,
|
| 331 |
+
"step": 46
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.4504792332268371,
|
| 335 |
+
"grad_norm": 0.8060439236010339,
|
| 336 |
+
"learning_rate": 1.985871018518236e-05,
|
| 337 |
+
"loss": 0.5651,
|
| 338 |
+
"step": 47
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.46006389776357826,
|
| 342 |
+
"grad_norm": 0.6140083254189492,
|
| 343 |
+
"learning_rate": 1.98392958859863e-05,
|
| 344 |
+
"loss": 0.5595,
|
| 345 |
+
"step": 48
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.4696485623003195,
|
| 349 |
+
"grad_norm": 0.7132583174991013,
|
| 350 |
+
"learning_rate": 1.9818642952361188e-05,
|
| 351 |
+
"loss": 0.5526,
|
| 352 |
+
"step": 49
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.4792332268370607,
|
| 356 |
+
"grad_norm": 0.59664072981028,
|
| 357 |
+
"learning_rate": 1.9796753984232357e-05,
|
| 358 |
+
"loss": 0.547,
|
| 359 |
+
"step": 50
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.48881789137380194,
|
| 363 |
+
"grad_norm": 0.6993452708095715,
|
| 364 |
+
"learning_rate": 1.9773631737125192e-05,
|
| 365 |
+
"loss": 0.5361,
|
| 366 |
+
"step": 51
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.4984025559105431,
|
| 370 |
+
"grad_norm": 0.6473604609615206,
|
| 371 |
+
"learning_rate": 1.9749279121818235e-05,
|
| 372 |
+
"loss": 0.5535,
|
| 373 |
+
"step": 52
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.5079872204472844,
|
| 377 |
+
"grad_norm": 0.7074480280853533,
|
| 378 |
+
"learning_rate": 1.9723699203976768e-05,
|
| 379 |
+
"loss": 0.4981,
|
| 380 |
+
"step": 53
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.5175718849840255,
|
| 384 |
+
"grad_norm": 0.7557340666761172,
|
| 385 |
+
"learning_rate": 1.969689520376687e-05,
|
| 386 |
+
"loss": 0.5379,
|
| 387 |
+
"step": 54
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.5271565495207667,
|
| 391 |
+
"grad_norm": 0.636305070807052,
|
| 392 |
+
"learning_rate": 1.9668870495450064e-05,
|
| 393 |
+
"loss": 0.5481,
|
| 394 |
+
"step": 55
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.536741214057508,
|
| 398 |
+
"grad_norm": 0.7057386270005085,
|
| 399 |
+
"learning_rate": 1.9639628606958535e-05,
|
| 400 |
+
"loss": 0.5118,
|
| 401 |
+
"step": 56
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.5463258785942492,
|
| 405 |
+
"grad_norm": 0.5999038558802152,
|
| 406 |
+
"learning_rate": 1.9609173219450998e-05,
|
| 407 |
+
"loss": 0.5532,
|
| 408 |
+
"step": 57
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.5559105431309904,
|
| 412 |
+
"grad_norm": 0.6446336459611116,
|
| 413 |
+
"learning_rate": 1.9577508166849308e-05,
|
| 414 |
+
"loss": 0.5443,
|
| 415 |
+
"step": 58
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.5654952076677316,
|
| 419 |
+
"grad_norm": 0.7779704431661995,
|
| 420 |
+
"learning_rate": 1.954463743535581e-05,
|
| 421 |
+
"loss": 0.5282,
|
| 422 |
+
"step": 59
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.5750798722044729,
|
| 426 |
+
"grad_norm": 0.6196806885678676,
|
| 427 |
+
"learning_rate": 1.9510565162951538e-05,
|
| 428 |
+
"loss": 0.528,
|
| 429 |
+
"step": 60
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.5846645367412141,
|
| 433 |
+
"grad_norm": 0.7481597405788767,
|
| 434 |
+
"learning_rate": 1.947529563887529e-05,
|
| 435 |
+
"loss": 0.5258,
|
| 436 |
+
"step": 61
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.5942492012779552,
|
| 440 |
+
"grad_norm": 0.5954399244612109,
|
| 441 |
+
"learning_rate": 1.9438833303083677e-05,
|
| 442 |
+
"loss": 0.5045,
|
| 443 |
+
"step": 62
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.6038338658146964,
|
| 447 |
+
"grad_norm": 0.773965376234353,
|
| 448 |
+
"learning_rate": 1.940118274569219e-05,
|
| 449 |
+
"loss": 0.5025,
|
| 450 |
+
"step": 63
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.6134185303514377,
|
| 454 |
+
"grad_norm": 0.5603381146719852,
|
| 455 |
+
"learning_rate": 1.9362348706397374e-05,
|
| 456 |
+
"loss": 0.5024,
|
| 457 |
+
"step": 64
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.6230031948881789,
|
| 461 |
+
"grad_norm": 0.7685723417664032,
|
| 462 |
+
"learning_rate": 1.9322336073880143e-05,
|
| 463 |
+
"loss": 0.5357,
|
| 464 |
+
"step": 65
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.6325878594249201,
|
| 468 |
+
"grad_norm": 0.6579285301369759,
|
| 469 |
+
"learning_rate": 1.928114988519039e-05,
|
| 470 |
+
"loss": 0.5126,
|
| 471 |
+
"step": 66
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.6421725239616614,
|
| 475 |
+
"grad_norm": 0.7627659065606073,
|
| 476 |
+
"learning_rate": 1.9238795325112867e-05,
|
| 477 |
+
"loss": 0.5234,
|
| 478 |
+
"step": 67
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.6517571884984026,
|
| 482 |
+
"grad_norm": 0.6307353614086453,
|
| 483 |
+
"learning_rate": 1.919527772551451e-05,
|
| 484 |
+
"loss": 0.5188,
|
| 485 |
+
"step": 68
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.6613418530351438,
|
| 489 |
+
"grad_norm": 0.6854214793729667,
|
| 490 |
+
"learning_rate": 1.91506025646732e-05,
|
| 491 |
+
"loss": 0.509,
|
| 492 |
+
"step": 69
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 0.670926517571885,
|
| 496 |
+
"grad_norm": 0.7177977977714395,
|
| 497 |
+
"learning_rate": 1.9104775466588162e-05,
|
| 498 |
+
"loss": 0.5363,
|
| 499 |
+
"step": 70
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.6805111821086262,
|
| 503 |
+
"grad_norm": 0.5740574361421374,
|
| 504 |
+
"learning_rate": 1.9057802200271943e-05,
|
| 505 |
+
"loss": 0.5269,
|
| 506 |
+
"step": 71
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.6900958466453674,
|
| 510 |
+
"grad_norm": 0.7278614746672593,
|
| 511 |
+
"learning_rate": 1.900968867902419e-05,
|
| 512 |
+
"loss": 0.5214,
|
| 513 |
+
"step": 72
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.6996805111821086,
|
| 517 |
+
"grad_norm": 0.6367723242487334,
|
| 518 |
+
"learning_rate": 1.8960440959687254e-05,
|
| 519 |
+
"loss": 0.5559,
|
| 520 |
+
"step": 73
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.7092651757188498,
|
| 524 |
+
"grad_norm": 0.6483330441964968,
|
| 525 |
+
"learning_rate": 1.891006524188368e-05,
|
| 526 |
+
"loss": 0.5321,
|
| 527 |
+
"step": 74
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.7188498402555911,
|
| 531 |
+
"grad_norm": 0.6290627842102959,
|
| 532 |
+
"learning_rate": 1.88585678672358e-05,
|
| 533 |
+
"loss": 0.4939,
|
| 534 |
+
"step": 75
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.7284345047923323,
|
| 538 |
+
"grad_norm": 0.6410854610772762,
|
| 539 |
+
"learning_rate": 1.880595531856738e-05,
|
| 540 |
+
"loss": 0.5319,
|
| 541 |
+
"step": 76
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.7380191693290735,
|
| 545 |
+
"grad_norm": 0.6731639380913599,
|
| 546 |
+
"learning_rate": 1.8752234219087538e-05,
|
| 547 |
+
"loss": 0.5273,
|
| 548 |
+
"step": 77
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.7476038338658147,
|
| 552 |
+
"grad_norm": 0.5316769493193355,
|
| 553 |
+
"learning_rate": 1.8697411331556958e-05,
|
| 554 |
+
"loss": 0.518,
|
| 555 |
+
"step": 78
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.7571884984025559,
|
| 559 |
+
"grad_norm": 0.6062668049547616,
|
| 560 |
+
"learning_rate": 1.864149355743655e-05,
|
| 561 |
+
"loss": 0.5075,
|
| 562 |
+
"step": 79
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 0.7667731629392971,
|
| 566 |
+
"grad_norm": 0.6276805556072127,
|
| 567 |
+
"learning_rate": 1.8584487936018663e-05,
|
| 568 |
+
"loss": 0.5058,
|
| 569 |
+
"step": 80
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.7763578274760383,
|
| 573 |
+
"grad_norm": 0.5471527462104901,
|
| 574 |
+
"learning_rate": 1.8526401643540924e-05,
|
| 575 |
+
"loss": 0.5031,
|
| 576 |
+
"step": 81
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 0.7859424920127795,
|
| 580 |
+
"grad_norm": 0.6123087953792351,
|
| 581 |
+
"learning_rate": 1.8467241992282842e-05,
|
| 582 |
+
"loss": 0.5374,
|
| 583 |
+
"step": 82
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 0.7955271565495208,
|
| 587 |
+
"grad_norm": 0.6180809301088696,
|
| 588 |
+
"learning_rate": 1.8407016429645305e-05,
|
| 589 |
+
"loss": 0.5352,
|
| 590 |
+
"step": 83
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.805111821086262,
|
| 594 |
+
"grad_norm": 0.5770189713402363,
|
| 595 |
+
"learning_rate": 1.834573253721303e-05,
|
| 596 |
+
"loss": 0.5001,
|
| 597 |
+
"step": 84
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 0.8146964856230032,
|
| 601 |
+
"grad_norm": 0.5611559252392707,
|
| 602 |
+
"learning_rate": 1.8283398029800167e-05,
|
| 603 |
+
"loss": 0.4936,
|
| 604 |
+
"step": 85
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 0.8242811501597445,
|
| 608 |
+
"grad_norm": 0.5376330084381403,
|
| 609 |
+
"learning_rate": 1.8220020754479104e-05,
|
| 610 |
+
"loss": 0.5128,
|
| 611 |
+
"step": 86
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 0.8338658146964856,
|
| 615 |
+
"grad_norm": 0.5930317746662469,
|
| 616 |
+
"learning_rate": 1.8155608689592604e-05,
|
| 617 |
+
"loss": 0.5195,
|
| 618 |
+
"step": 87
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 0.8434504792332268,
|
| 622 |
+
"grad_norm": 0.5582349649774669,
|
| 623 |
+
"learning_rate": 1.8090169943749477e-05,
|
| 624 |
+
"loss": 0.5412,
|
| 625 |
+
"step": 88
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.853035143769968,
|
| 629 |
+
"grad_norm": 0.5840770892840872,
|
| 630 |
+
"learning_rate": 1.8023712754803783e-05,
|
| 631 |
+
"loss": 0.4931,
|
| 632 |
+
"step": 89
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.8626198083067093,
|
| 636 |
+
"grad_norm": 0.5770400921032719,
|
| 637 |
+
"learning_rate": 1.795624548881781e-05,
|
| 638 |
+
"loss": 0.5077,
|
| 639 |
+
"step": 90
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.8722044728434505,
|
| 643 |
+
"grad_norm": 0.47321773400110334,
|
| 644 |
+
"learning_rate": 1.7887776639008912e-05,
|
| 645 |
+
"loss": 0.5271,
|
| 646 |
+
"step": 91
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 0.8817891373801917,
|
| 650 |
+
"grad_norm": 0.6014894804474964,
|
| 651 |
+
"learning_rate": 1.78183148246803e-05,
|
| 652 |
+
"loss": 0.5068,
|
| 653 |
+
"step": 92
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 0.8913738019169329,
|
| 657 |
+
"grad_norm": 0.5268826943097094,
|
| 658 |
+
"learning_rate": 1.7747868790136012e-05,
|
| 659 |
+
"loss": 0.5062,
|
| 660 |
+
"step": 93
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 0.9009584664536742,
|
| 664 |
+
"grad_norm": 0.538258599224071,
|
| 665 |
+
"learning_rate": 1.7676447403580114e-05,
|
| 666 |
+
"loss": 0.5192,
|
| 667 |
+
"step": 94
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 0.9105431309904153,
|
| 671 |
+
"grad_norm": 0.5235902995089281,
|
| 672 |
+
"learning_rate": 1.7604059656000313e-05,
|
| 673 |
+
"loss": 0.512,
|
| 674 |
+
"step": 95
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 0.9201277955271565,
|
| 678 |
+
"grad_norm": 0.5922248527214893,
|
| 679 |
+
"learning_rate": 1.7530714660036112e-05,
|
| 680 |
+
"loss": 0.5055,
|
| 681 |
+
"step": 96
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 0.9297124600638977,
|
| 685 |
+
"grad_norm": 0.5855685386711252,
|
| 686 |
+
"learning_rate": 1.7456421648831658e-05,
|
| 687 |
+
"loss": 0.4984,
|
| 688 |
+
"step": 97
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 0.939297124600639,
|
| 692 |
+
"grad_norm": 0.5458918187075826,
|
| 693 |
+
"learning_rate": 1.738118997487341e-05,
|
| 694 |
+
"loss": 0.5076,
|
| 695 |
+
"step": 98
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 0.9488817891373802,
|
| 699 |
+
"grad_norm": 0.6515459002851856,
|
| 700 |
+
"learning_rate": 1.7305029108812777e-05,
|
| 701 |
+
"loss": 0.5229,
|
| 702 |
+
"step": 99
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 0.9584664536741214,
|
| 706 |
+
"grad_norm": 0.5315082766100536,
|
| 707 |
+
"learning_rate": 1.7227948638273918e-05,
|
| 708 |
+
"loss": 0.5093,
|
| 709 |
+
"step": 100
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 0.9680511182108626,
|
| 713 |
+
"grad_norm": 0.7875828584853896,
|
| 714 |
+
"learning_rate": 1.7149958266646756e-05,
|
| 715 |
+
"loss": 0.494,
|
| 716 |
+
"step": 101
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 0.9776357827476039,
|
| 720 |
+
"grad_norm": 0.5868831184446074,
|
| 721 |
+
"learning_rate": 1.7071067811865477e-05,
|
| 722 |
+
"loss": 0.4865,
|
| 723 |
+
"step": 102
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 0.987220447284345,
|
| 727 |
+
"grad_norm": 0.5730857899752412,
|
| 728 |
+
"learning_rate": 1.6991287205172575e-05,
|
| 729 |
+
"loss": 0.513,
|
| 730 |
+
"step": 103
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 0.9968051118210862,
|
| 734 |
+
"grad_norm": 0.7358514287452019,
|
| 735 |
+
"learning_rate": 1.691062648986865e-05,
|
| 736 |
+
"loss": 0.5143,
|
| 737 |
+
"step": 104
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 1.0063897763578276,
|
| 741 |
+
"grad_norm": 1.1699812558564358,
|
| 742 |
+
"learning_rate": 1.682909582004807e-05,
|
| 743 |
+
"loss": 0.8838,
|
| 744 |
+
"step": 105
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 1.0159744408945688,
|
| 748 |
+
"grad_norm": 0.6902201583829356,
|
| 749 |
+
"learning_rate": 1.6746705459320746e-05,
|
| 750 |
+
"loss": 0.4084,
|
| 751 |
+
"step": 106
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 1.0255591054313098,
|
| 755 |
+
"grad_norm": 0.6792794510906274,
|
| 756 |
+
"learning_rate": 1.6663465779520042e-05,
|
| 757 |
+
"loss": 0.436,
|
| 758 |
+
"step": 107
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 1.035143769968051,
|
| 762 |
+
"grad_norm": 0.7421226478591257,
|
| 763 |
+
"learning_rate": 1.657938725939713e-05,
|
| 764 |
+
"loss": 0.4414,
|
| 765 |
+
"step": 108
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 1.0447284345047922,
|
| 769 |
+
"grad_norm": 0.6881793520210084,
|
| 770 |
+
"learning_rate": 1.6494480483301836e-05,
|
| 771 |
+
"loss": 0.4859,
|
| 772 |
+
"step": 109
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 1.0543130990415335,
|
| 776 |
+
"grad_norm": 0.821089177012622,
|
| 777 |
+
"learning_rate": 1.6408756139850243e-05,
|
| 778 |
+
"loss": 0.4273,
|
| 779 |
+
"step": 110
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 1.0638977635782747,
|
| 783 |
+
"grad_norm": 0.6928979324587471,
|
| 784 |
+
"learning_rate": 1.63222250205791e-05,
|
| 785 |
+
"loss": 0.4819,
|
| 786 |
+
"step": 111
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 1.073482428115016,
|
| 790 |
+
"grad_norm": 0.7022553739631447,
|
| 791 |
+
"learning_rate": 1.6234898018587336e-05,
|
| 792 |
+
"loss": 0.3695,
|
| 793 |
+
"step": 112
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 1.0830670926517572,
|
| 797 |
+
"grad_norm": 0.725323982592671,
|
| 798 |
+
"learning_rate": 1.6146786127164773e-05,
|
| 799 |
+
"loss": 0.461,
|
| 800 |
+
"step": 113
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 1.0926517571884984,
|
| 804 |
+
"grad_norm": 0.7048555706325064,
|
| 805 |
+
"learning_rate": 1.60579004384082e-05,
|
| 806 |
+
"loss": 0.4494,
|
| 807 |
+
"step": 114
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 1.1022364217252396,
|
| 811 |
+
"grad_norm": 0.6640188248580515,
|
| 812 |
+
"learning_rate": 1.5968252141825038e-05,
|
| 813 |
+
"loss": 0.4685,
|
| 814 |
+
"step": 115
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 1.1118210862619808,
|
| 818 |
+
"grad_norm": 0.7030304823361989,
|
| 819 |
+
"learning_rate": 1.5877852522924733e-05,
|
| 820 |
+
"loss": 0.4277,
|
| 821 |
+
"step": 116
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 1.121405750798722,
|
| 825 |
+
"grad_norm": 0.7123317956342814,
|
| 826 |
+
"learning_rate": 1.578671296179806e-05,
|
| 827 |
+
"loss": 0.4592,
|
| 828 |
+
"step": 117
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 1.1309904153354633,
|
| 832 |
+
"grad_norm": 0.6086700980301015,
|
| 833 |
+
"learning_rate": 1.569484493168452e-05,
|
| 834 |
+
"loss": 0.412,
|
| 835 |
+
"step": 118
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 1.1405750798722045,
|
| 839 |
+
"grad_norm": 0.6390103576527882,
|
| 840 |
+
"learning_rate": 1.5602259997528028e-05,
|
| 841 |
+
"loss": 0.4797,
|
| 842 |
+
"step": 119
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 1.1501597444089458,
|
| 846 |
+
"grad_norm": 0.5469865905883947,
|
| 847 |
+
"learning_rate": 1.5508969814521026e-05,
|
| 848 |
+
"loss": 0.3935,
|
| 849 |
+
"step": 120
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 1.159744408945687,
|
| 853 |
+
"grad_norm": 0.5758273790847533,
|
| 854 |
+
"learning_rate": 1.541498612663726e-05,
|
| 855 |
+
"loss": 0.4702,
|
| 856 |
+
"step": 121
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 1.1693290734824282,
|
| 860 |
+
"grad_norm": 0.4971600554124392,
|
| 861 |
+
"learning_rate": 1.5320320765153367e-05,
|
| 862 |
+
"loss": 0.416,
|
| 863 |
+
"step": 122
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 1.1789137380191694,
|
| 867 |
+
"grad_norm": 0.6214619772407622,
|
| 868 |
+
"learning_rate": 1.5224985647159489e-05,
|
| 869 |
+
"loss": 0.4474,
|
| 870 |
+
"step": 123
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 1.1884984025559104,
|
| 874 |
+
"grad_norm": 0.6547899224858453,
|
| 875 |
+
"learning_rate": 1.5128992774059063e-05,
|
| 876 |
+
"loss": 0.4558,
|
| 877 |
+
"step": 124
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 1.1980830670926517,
|
| 881 |
+
"grad_norm": 0.5890574227637935,
|
| 882 |
+
"learning_rate": 1.5032354230058004e-05,
|
| 883 |
+
"loss": 0.47,
|
| 884 |
+
"step": 125
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 1.207667731629393,
|
| 888 |
+
"grad_norm": 0.7361985894763436,
|
| 889 |
+
"learning_rate": 1.493508218064347e-05,
|
| 890 |
+
"loss": 0.4309,
|
| 891 |
+
"step": 126
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 1.2172523961661341,
|
| 895 |
+
"grad_norm": 0.5173156701874413,
|
| 896 |
+
"learning_rate": 1.4837188871052399e-05,
|
| 897 |
+
"loss": 0.4814,
|
| 898 |
+
"step": 127
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 1.2268370607028753,
|
| 902 |
+
"grad_norm": 0.6237781339666125,
|
| 903 |
+
"learning_rate": 1.4738686624729987e-05,
|
| 904 |
+
"loss": 0.4185,
|
| 905 |
+
"step": 128
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 1.2364217252396166,
|
| 909 |
+
"grad_norm": 0.6273705352740975,
|
| 910 |
+
"learning_rate": 1.4639587841778342e-05,
|
| 911 |
+
"loss": 0.4637,
|
| 912 |
+
"step": 129
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 1.2460063897763578,
|
| 916 |
+
"grad_norm": 0.5605628666508281,
|
| 917 |
+
"learning_rate": 1.4539904997395468e-05,
|
| 918 |
+
"loss": 0.4803,
|
| 919 |
+
"step": 130
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 1.255591054313099,
|
| 923 |
+
"grad_norm": 0.597522913497888,
|
| 924 |
+
"learning_rate": 1.4439650640304822e-05,
|
| 925 |
+
"loss": 0.454,
|
| 926 |
+
"step": 131
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 1.2651757188498403,
|
| 930 |
+
"grad_norm": 0.5715006951258484,
|
| 931 |
+
"learning_rate": 1.4338837391175582e-05,
|
| 932 |
+
"loss": 0.4441,
|
| 933 |
+
"step": 132
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 1.2747603833865815,
|
| 937 |
+
"grad_norm": 0.5270300631154146,
|
| 938 |
+
"learning_rate": 1.4237477941033888e-05,
|
| 939 |
+
"loss": 0.4424,
|
| 940 |
+
"step": 133
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 1.2843450479233227,
|
| 944 |
+
"grad_norm": 0.6111693532495731,
|
| 945 |
+
"learning_rate": 1.4135585049665207e-05,
|
| 946 |
+
"loss": 0.4688,
|
| 947 |
+
"step": 134
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 1.293929712460064,
|
| 951 |
+
"grad_norm": 0.5720421473056696,
|
| 952 |
+
"learning_rate": 1.4033171544008053e-05,
|
| 953 |
+
"loss": 0.47,
|
| 954 |
+
"step": 135
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 1.3035143769968052,
|
| 958 |
+
"grad_norm": 0.6142039193499456,
|
| 959 |
+
"learning_rate": 1.3930250316539237e-05,
|
| 960 |
+
"loss": 0.444,
|
| 961 |
+
"step": 136
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 1.3130990415335464,
|
| 965 |
+
"grad_norm": 0.5632554602858334,
|
| 966 |
+
"learning_rate": 1.3826834323650899e-05,
|
| 967 |
+
"loss": 0.4447,
|
| 968 |
+
"step": 137
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 1.3226837060702876,
|
| 972 |
+
"grad_norm": 0.5306960211119153,
|
| 973 |
+
"learning_rate": 1.3722936584019453e-05,
|
| 974 |
+
"loss": 0.4464,
|
| 975 |
+
"step": 138
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 1.3322683706070286,
|
| 979 |
+
"grad_norm": 0.5026216587572911,
|
| 980 |
+
"learning_rate": 1.3618570176966723e-05,
|
| 981 |
+
"loss": 0.4635,
|
| 982 |
+
"step": 139
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 1.34185303514377,
|
| 986 |
+
"grad_norm": 0.5048409273626058,
|
| 987 |
+
"learning_rate": 1.3513748240813429e-05,
|
| 988 |
+
"loss": 0.4719,
|
| 989 |
+
"step": 140
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 1.351437699680511,
|
| 993 |
+
"grad_norm": 0.5143189141152029,
|
| 994 |
+
"learning_rate": 1.340848397122525e-05,
|
| 995 |
+
"loss": 0.423,
|
| 996 |
+
"step": 141
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 1.3610223642172525,
|
| 1000 |
+
"grad_norm": 0.4906781147146303,
|
| 1001 |
+
"learning_rate": 1.3302790619551673e-05,
|
| 1002 |
+
"loss": 0.4658,
|
| 1003 |
+
"step": 142
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 1.3706070287539935,
|
| 1007 |
+
"grad_norm": 0.45394412594825295,
|
| 1008 |
+
"learning_rate": 1.3196681491157816e-05,
|
| 1009 |
+
"loss": 0.4383,
|
| 1010 |
+
"step": 143
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 1.3801916932907348,
|
| 1014 |
+
"grad_norm": 0.4565249809642115,
|
| 1015 |
+
"learning_rate": 1.3090169943749475e-05,
|
| 1016 |
+
"loss": 0.4071,
|
| 1017 |
+
"step": 144
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 1.389776357827476,
|
| 1021 |
+
"grad_norm": 0.5436018324080563,
|
| 1022 |
+
"learning_rate": 1.2983269385691562e-05,
|
| 1023 |
+
"loss": 0.4942,
|
| 1024 |
+
"step": 145
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 1.3993610223642172,
|
| 1028 |
+
"grad_norm": 0.40894998786022985,
|
| 1029 |
+
"learning_rate": 1.2875993274320173e-05,
|
| 1030 |
+
"loss": 0.3947,
|
| 1031 |
+
"step": 146
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 1.4089456869009584,
|
| 1035 |
+
"grad_norm": 0.5288187343789331,
|
| 1036 |
+
"learning_rate": 1.2768355114248493e-05,
|
| 1037 |
+
"loss": 0.4712,
|
| 1038 |
+
"step": 147
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 1.4185303514376997,
|
| 1042 |
+
"grad_norm": 0.4768621544922896,
|
| 1043 |
+
"learning_rate": 1.2660368455666752e-05,
|
| 1044 |
+
"loss": 0.4382,
|
| 1045 |
+
"step": 148
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 1.428115015974441,
|
| 1049 |
+
"grad_norm": 0.431738859820575,
|
| 1050 |
+
"learning_rate": 1.2552046892636427e-05,
|
| 1051 |
+
"loss": 0.4071,
|
| 1052 |
+
"step": 149
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 1.4376996805111821,
|
| 1056 |
+
"grad_norm": 0.498599086228044,
|
| 1057 |
+
"learning_rate": 1.2443404061378941e-05,
|
| 1058 |
+
"loss": 0.4755,
|
| 1059 |
+
"step": 150
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 1.4472843450479234,
|
| 1063 |
+
"grad_norm": 0.4988206100753027,
|
| 1064 |
+
"learning_rate": 1.2334453638559057e-05,
|
| 1065 |
+
"loss": 0.4083,
|
| 1066 |
+
"step": 151
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 1.4568690095846646,
|
| 1070 |
+
"grad_norm": 0.45822495711811395,
|
| 1071 |
+
"learning_rate": 1.2225209339563144e-05,
|
| 1072 |
+
"loss": 0.485,
|
| 1073 |
+
"step": 152
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 1.4664536741214058,
|
| 1077 |
+
"grad_norm": 0.5023477915233621,
|
| 1078 |
+
"learning_rate": 1.211568491677263e-05,
|
| 1079 |
+
"loss": 0.4076,
|
| 1080 |
+
"step": 153
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 1.476038338658147,
|
| 1084 |
+
"grad_norm": 0.45113635635588584,
|
| 1085 |
+
"learning_rate": 1.200589415783273e-05,
|
| 1086 |
+
"loss": 0.4015,
|
| 1087 |
+
"step": 154
|
| 1088 |
+
},
|
| 1089 |
+
{
|
| 1090 |
+
"epoch": 1.4856230031948883,
|
| 1091 |
+
"grad_norm": 0.5334811995129992,
|
| 1092 |
+
"learning_rate": 1.1895850883916786e-05,
|
| 1093 |
+
"loss": 0.4759,
|
| 1094 |
+
"step": 155
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 1.4952076677316293,
|
| 1098 |
+
"grad_norm": 0.40436266318440944,
|
| 1099 |
+
"learning_rate": 1.1785568947986368e-05,
|
| 1100 |
+
"loss": 0.419,
|
| 1101 |
+
"step": 156
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 1.5047923322683707,
|
| 1105 |
+
"grad_norm": 0.4755461357936275,
|
| 1106 |
+
"learning_rate": 1.1675062233047365e-05,
|
| 1107 |
+
"loss": 0.4221,
|
| 1108 |
+
"step": 157
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"epoch": 1.5143769968051117,
|
| 1112 |
+
"grad_norm": 0.5300609622813354,
|
| 1113 |
+
"learning_rate": 1.156434465040231e-05,
|
| 1114 |
+
"loss": 0.4645,
|
| 1115 |
+
"step": 158
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 1.5239616613418532,
|
| 1119 |
+
"grad_norm": 0.40635331262756796,
|
| 1120 |
+
"learning_rate": 1.1453430137899129e-05,
|
| 1121 |
+
"loss": 0.3713,
|
| 1122 |
+
"step": 159
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 1.5335463258785942,
|
| 1126 |
+
"grad_norm": 0.4735547448374356,
|
| 1127 |
+
"learning_rate": 1.1342332658176556e-05,
|
| 1128 |
+
"loss": 0.4324,
|
| 1129 |
+
"step": 160
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 1.5431309904153354,
|
| 1133 |
+
"grad_norm": 0.4495839729557387,
|
| 1134 |
+
"learning_rate": 1.123106619690643e-05,
|
| 1135 |
+
"loss": 0.4791,
|
| 1136 |
+
"step": 161
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 1.5527156549520766,
|
| 1140 |
+
"grad_norm": 0.4173934080478346,
|
| 1141 |
+
"learning_rate": 1.1119644761033079e-05,
|
| 1142 |
+
"loss": 0.3956,
|
| 1143 |
+
"step": 162
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 1.5623003194888179,
|
| 1147 |
+
"grad_norm": 0.5099574697612899,
|
| 1148 |
+
"learning_rate": 1.1008082377010045e-05,
|
| 1149 |
+
"loss": 0.502,
|
| 1150 |
+
"step": 163
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 1.571884984025559,
|
| 1154 |
+
"grad_norm": 0.4022180449613417,
|
| 1155 |
+
"learning_rate": 1.0896393089034336e-05,
|
| 1156 |
+
"loss": 0.3678,
|
| 1157 |
+
"step": 164
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 1.5814696485623003,
|
| 1161 |
+
"grad_norm": 0.46434170894732985,
|
| 1162 |
+
"learning_rate": 1.0784590957278452e-05,
|
| 1163 |
+
"loss": 0.4521,
|
| 1164 |
+
"step": 165
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 1.5910543130990416,
|
| 1168 |
+
"grad_norm": 0.5222429942261165,
|
| 1169 |
+
"learning_rate": 1.0672690056120398e-05,
|
| 1170 |
+
"loss": 0.4578,
|
| 1171 |
+
"step": 166
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 1.6006389776357828,
|
| 1175 |
+
"grad_norm": 0.4714562379867376,
|
| 1176 |
+
"learning_rate": 1.0560704472371919e-05,
|
| 1177 |
+
"loss": 0.4804,
|
| 1178 |
+
"step": 167
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 1.610223642172524,
|
| 1182 |
+
"grad_norm": 0.4257910311693338,
|
| 1183 |
+
"learning_rate": 1.044864830350515e-05,
|
| 1184 |
+
"loss": 0.3879,
|
| 1185 |
+
"step": 168
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 1.619808306709265,
|
| 1189 |
+
"grad_norm": 0.4931956302496282,
|
| 1190 |
+
"learning_rate": 1.0336535655877942e-05,
|
| 1191 |
+
"loss": 0.4756,
|
| 1192 |
+
"step": 169
|
| 1193 |
+
},
|
| 1194 |
+
{
|
| 1195 |
+
"epoch": 1.6293929712460065,
|
| 1196 |
+
"grad_norm": 0.4307049907622935,
|
| 1197 |
+
"learning_rate": 1.0224380642958052e-05,
|
| 1198 |
+
"loss": 0.3645,
|
| 1199 |
+
"step": 170
|
| 1200 |
+
},
|
| 1201 |
+
{
|
| 1202 |
+
"epoch": 1.6389776357827475,
|
| 1203 |
+
"grad_norm": 0.5837734683656293,
|
| 1204 |
+
"learning_rate": 1.011219738354646e-05,
|
| 1205 |
+
"loss": 0.4896,
|
| 1206 |
+
"step": 171
|
| 1207 |
+
},
|
| 1208 |
+
{
|
| 1209 |
+
"epoch": 1.648562300319489,
|
| 1210 |
+
"grad_norm": 0.4274446462676884,
|
| 1211 |
+
"learning_rate": 1e-05,
|
| 1212 |
+
"loss": 0.4297,
|
| 1213 |
+
"step": 172
|
| 1214 |
+
},
|
| 1215 |
+
{
|
| 1216 |
+
"epoch": 1.65814696485623,
|
| 1217 |
+
"grad_norm": 0.47050426719612987,
|
| 1218 |
+
"learning_rate": 9.887802616453543e-06,
|
| 1219 |
+
"loss": 0.4016,
|
| 1220 |
+
"step": 173
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 1.6677316293929714,
|
| 1224 |
+
"grad_norm": 0.46798902119678215,
|
| 1225 |
+
"learning_rate": 9.775619357041952e-06,
|
| 1226 |
+
"loss": 0.4468,
|
| 1227 |
+
"step": 174
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"epoch": 1.6773162939297124,
|
| 1231 |
+
"grad_norm": 0.4231517848150463,
|
| 1232 |
+
"learning_rate": 9.663464344122064e-06,
|
| 1233 |
+
"loss": 0.4054,
|
| 1234 |
+
"step": 175
|
| 1235 |
+
},
|
| 1236 |
+
{
|
| 1237 |
+
"epoch": 1.6869009584664538,
|
| 1238 |
+
"grad_norm": 0.5047554859497767,
|
| 1239 |
+
"learning_rate": 9.551351696494854e-06,
|
| 1240 |
+
"loss": 0.4564,
|
| 1241 |
+
"step": 176
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 1.6964856230031948,
|
| 1245 |
+
"grad_norm": 0.37022389130163336,
|
| 1246 |
+
"learning_rate": 9.439295527628083e-06,
|
| 1247 |
+
"loss": 0.4373,
|
| 1248 |
+
"step": 177
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 1.706070287539936,
|
| 1252 |
+
"grad_norm": 0.5285912537102634,
|
| 1253 |
+
"learning_rate": 9.327309943879604e-06,
|
| 1254 |
+
"loss": 0.4722,
|
| 1255 |
+
"step": 178
|
| 1256 |
+
},
|
| 1257 |
+
{
|
| 1258 |
+
"epoch": 1.7156549520766773,
|
| 1259 |
+
"grad_norm": 0.4785755524774783,
|
| 1260 |
+
"learning_rate": 9.215409042721553e-06,
|
| 1261 |
+
"loss": 0.426,
|
| 1262 |
+
"step": 179
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 1.7252396166134185,
|
| 1266 |
+
"grad_norm": 0.4506434737215472,
|
| 1267 |
+
"learning_rate": 9.103606910965666e-06,
|
| 1268 |
+
"loss": 0.4384,
|
| 1269 |
+
"step": 180
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 1.7348242811501597,
|
| 1273 |
+
"grad_norm": 0.45206163197522226,
|
| 1274 |
+
"learning_rate": 8.991917622989957e-06,
|
| 1275 |
+
"loss": 0.465,
|
| 1276 |
+
"step": 181
|
| 1277 |
+
},
|
| 1278 |
+
{
|
| 1279 |
+
"epoch": 1.744408945686901,
|
| 1280 |
+
"grad_norm": 0.4264275243649636,
|
| 1281 |
+
"learning_rate": 8.880355238966923e-06,
|
| 1282 |
+
"loss": 0.4294,
|
| 1283 |
+
"step": 182
|
| 1284 |
+
},
|
| 1285 |
+
{
|
| 1286 |
+
"epoch": 1.7539936102236422,
|
| 1287 |
+
"grad_norm": 0.4305322367018383,
|
| 1288 |
+
"learning_rate": 8.768933803093573e-06,
|
| 1289 |
+
"loss": 0.439,
|
| 1290 |
+
"step": 183
|
| 1291 |
+
},
|
| 1292 |
+
{
|
| 1293 |
+
"epoch": 1.7635782747603834,
|
| 1294 |
+
"grad_norm": 0.444413629694292,
|
| 1295 |
+
"learning_rate": 8.657667341823449e-06,
|
| 1296 |
+
"loss": 0.4471,
|
| 1297 |
+
"step": 184
|
| 1298 |
+
},
|
| 1299 |
+
{
|
| 1300 |
+
"epoch": 1.7731629392971247,
|
| 1301 |
+
"grad_norm": 0.44726871944510105,
|
| 1302 |
+
"learning_rate": 8.546569862100876e-06,
|
| 1303 |
+
"loss": 0.4291,
|
| 1304 |
+
"step": 185
|
| 1305 |
+
},
|
| 1306 |
+
{
|
| 1307 |
+
"epoch": 1.7827476038338657,
|
| 1308 |
+
"grad_norm": 0.5141706810151837,
|
| 1309 |
+
"learning_rate": 8.43565534959769e-06,
|
| 1310 |
+
"loss": 0.4411,
|
| 1311 |
+
"step": 186
|
| 1312 |
+
},
|
| 1313 |
+
{
|
| 1314 |
+
"epoch": 1.792332268370607,
|
| 1315 |
+
"grad_norm": 0.4135951238897178,
|
| 1316 |
+
"learning_rate": 8.324937766952638e-06,
|
| 1317 |
+
"loss": 0.4325,
|
| 1318 |
+
"step": 187
|
| 1319 |
+
},
|
| 1320 |
+
{
|
| 1321 |
+
"epoch": 1.8019169329073481,
|
| 1322 |
+
"grad_norm": 0.45368873581779895,
|
| 1323 |
+
"learning_rate": 8.214431052013636e-06,
|
| 1324 |
+
"loss": 0.4228,
|
| 1325 |
+
"step": 188
|
| 1326 |
+
},
|
| 1327 |
+
{
|
| 1328 |
+
"epoch": 1.8115015974440896,
|
| 1329 |
+
"grad_norm": 0.491585413229281,
|
| 1330 |
+
"learning_rate": 8.104149116083216e-06,
|
| 1331 |
+
"loss": 0.4516,
|
| 1332 |
+
"step": 189
|
| 1333 |
+
},
|
| 1334 |
+
{
|
| 1335 |
+
"epoch": 1.8210862619808306,
|
| 1336 |
+
"grad_norm": 0.40058123705120063,
|
| 1337 |
+
"learning_rate": 7.994105842167274e-06,
|
| 1338 |
+
"loss": 0.4331,
|
| 1339 |
+
"step": 190
|
| 1340 |
+
},
|
| 1341 |
+
{
|
| 1342 |
+
"epoch": 1.830670926517572,
|
| 1343 |
+
"grad_norm": 0.4537379960494452,
|
| 1344 |
+
"learning_rate": 7.884315083227373e-06,
|
| 1345 |
+
"loss": 0.4685,
|
| 1346 |
+
"step": 191
|
| 1347 |
+
},
|
| 1348 |
+
{
|
| 1349 |
+
"epoch": 1.840255591054313,
|
| 1350 |
+
"grad_norm": 0.4212878067511151,
|
| 1351 |
+
"learning_rate": 7.774790660436857e-06,
|
| 1352 |
+
"loss": 0.4405,
|
| 1353 |
+
"step": 192
|
| 1354 |
+
},
|
| 1355 |
+
{
|
| 1356 |
+
"epoch": 1.8498402555910545,
|
| 1357 |
+
"grad_norm": 0.4051053145614115,
|
| 1358 |
+
"learning_rate": 7.66554636144095e-06,
|
| 1359 |
+
"loss": 0.438,
|
| 1360 |
+
"step": 193
|
| 1361 |
+
},
|
| 1362 |
+
{
|
| 1363 |
+
"epoch": 1.8594249201277955,
|
| 1364 |
+
"grad_norm": 0.40403658818972926,
|
| 1365 |
+
"learning_rate": 7.556595938621058e-06,
|
| 1366 |
+
"loss": 0.4153,
|
| 1367 |
+
"step": 194
|
| 1368 |
+
},
|
| 1369 |
+
{
|
| 1370 |
+
"epoch": 1.8690095846645367,
|
| 1371 |
+
"grad_norm": 0.39919476552418626,
|
| 1372 |
+
"learning_rate": 7.447953107363574e-06,
|
| 1373 |
+
"loss": 0.4042,
|
| 1374 |
+
"step": 195
|
| 1375 |
+
},
|
| 1376 |
+
{
|
| 1377 |
+
"epoch": 1.878594249201278,
|
| 1378 |
+
"grad_norm": 0.40390408443700876,
|
| 1379 |
+
"learning_rate": 7.33963154433325e-06,
|
| 1380 |
+
"loss": 0.4027,
|
| 1381 |
+
"step": 196
|
| 1382 |
+
},
|
| 1383 |
+
{
|
| 1384 |
+
"epoch": 1.8881789137380192,
|
| 1385 |
+
"grad_norm": 0.39441465040012735,
|
| 1386 |
+
"learning_rate": 7.2316448857515076e-06,
|
| 1387 |
+
"loss": 0.4435,
|
| 1388 |
+
"step": 197
|
| 1389 |
+
},
|
| 1390 |
+
{
|
| 1391 |
+
"epoch": 1.8977635782747604,
|
| 1392 |
+
"grad_norm": 0.43435324742018255,
|
| 1393 |
+
"learning_rate": 7.124006725679828e-06,
|
| 1394 |
+
"loss": 0.4618,
|
| 1395 |
+
"step": 198
|
| 1396 |
+
},
|
| 1397 |
+
{
|
| 1398 |
+
"epoch": 1.9073482428115016,
|
| 1399 |
+
"grad_norm": 0.4117241007254349,
|
| 1400 |
+
"learning_rate": 7.01673061430844e-06,
|
| 1401 |
+
"loss": 0.4302,
|
| 1402 |
+
"step": 199
|
| 1403 |
+
},
|
| 1404 |
+
{
|
| 1405 |
+
"epoch": 1.9169329073482428,
|
| 1406 |
+
"grad_norm": 0.42716318836747486,
|
| 1407 |
+
"learning_rate": 6.909830056250527e-06,
|
| 1408 |
+
"loss": 0.4766,
|
| 1409 |
+
"step": 200
|
| 1410 |
+
},
|
| 1411 |
+
{
|
| 1412 |
+
"epoch": 1.926517571884984,
|
| 1413 |
+
"grad_norm": 0.4284737496521204,
|
| 1414 |
+
"learning_rate": 6.8033185088421874e-06,
|
| 1415 |
+
"loss": 0.455,
|
| 1416 |
+
"step": 201
|
| 1417 |
+
},
|
| 1418 |
+
{
|
| 1419 |
+
"epoch": 1.9361022364217253,
|
| 1420 |
+
"grad_norm": 0.4709549975113482,
|
| 1421 |
+
"learning_rate": 6.697209380448333e-06,
|
| 1422 |
+
"loss": 0.4249,
|
| 1423 |
+
"step": 202
|
| 1424 |
+
},
|
| 1425 |
+
{
|
| 1426 |
+
"epoch": 1.9456869009584663,
|
| 1427 |
+
"grad_norm": 0.38457354298640695,
|
| 1428 |
+
"learning_rate": 6.59151602877475e-06,
|
| 1429 |
+
"loss": 0.4298,
|
| 1430 |
+
"step": 203
|
| 1431 |
+
},
|
| 1432 |
+
{
|
| 1433 |
+
"epoch": 1.9552715654952078,
|
| 1434 |
+
"grad_norm": 0.40219036129356933,
|
| 1435 |
+
"learning_rate": 6.486251759186573e-06,
|
| 1436 |
+
"loss": 0.46,
|
| 1437 |
+
"step": 204
|
| 1438 |
+
},
|
| 1439 |
+
{
|
| 1440 |
+
"epoch": 1.9648562300319488,
|
| 1441 |
+
"grad_norm": 0.4394898406507064,
|
| 1442 |
+
"learning_rate": 6.381429823033281e-06,
|
| 1443 |
+
"loss": 0.4507,
|
| 1444 |
+
"step": 205
|
| 1445 |
+
},
|
| 1446 |
+
{
|
| 1447 |
+
"epoch": 1.9744408945686902,
|
| 1448 |
+
"grad_norm": 0.428010655667797,
|
| 1449 |
+
"learning_rate": 6.277063415980549e-06,
|
| 1450 |
+
"loss": 0.4289,
|
| 1451 |
+
"step": 206
|
| 1452 |
+
},
|
| 1453 |
+
{
|
| 1454 |
+
"epoch": 1.9840255591054312,
|
| 1455 |
+
"grad_norm": 0.41752005507808904,
|
| 1456 |
+
"learning_rate": 6.173165676349103e-06,
|
| 1457 |
+
"loss": 0.4253,
|
| 1458 |
+
"step": 207
|
| 1459 |
+
},
|
| 1460 |
+
{
|
| 1461 |
+
"epoch": 1.9936102236421727,
|
| 1462 |
+
"grad_norm": 0.4446826344625602,
|
| 1463 |
+
"learning_rate": 6.069749683460765e-06,
|
| 1464 |
+
"loss": 0.4653,
|
| 1465 |
+
"step": 208
|
| 1466 |
+
},
|
| 1467 |
+
{
|
| 1468 |
+
"epoch": 2.0031948881789137,
|
| 1469 |
+
"grad_norm": 0.7825903799604366,
|
| 1470 |
+
"learning_rate": 5.966828455991951e-06,
|
| 1471 |
+
"loss": 0.6218,
|
| 1472 |
+
"step": 209
|
| 1473 |
+
},
|
| 1474 |
+
{
|
| 1475 |
+
"epoch": 2.012779552715655,
|
| 1476 |
+
"grad_norm": 0.4401481205860557,
|
| 1477 |
+
"learning_rate": 5.864414950334796e-06,
|
| 1478 |
+
"loss": 0.3884,
|
| 1479 |
+
"step": 210
|
| 1480 |
+
},
|
| 1481 |
+
{
|
| 1482 |
+
"epoch": 2.022364217252396,
|
| 1483 |
+
"grad_norm": 0.45142190780046537,
|
| 1484 |
+
"learning_rate": 5.7625220589661136e-06,
|
| 1485 |
+
"loss": 0.392,
|
| 1486 |
+
"step": 211
|
| 1487 |
+
},
|
| 1488 |
+
{
|
| 1489 |
+
"epoch": 2.0319488817891376,
|
| 1490 |
+
"grad_norm": 0.48929701776811646,
|
| 1491 |
+
"learning_rate": 5.66116260882442e-06,
|
| 1492 |
+
"loss": 0.3631,
|
| 1493 |
+
"step": 212
|
| 1494 |
+
},
|
| 1495 |
+
{
|
| 1496 |
+
"epoch": 2.0415335463258786,
|
| 1497 |
+
"grad_norm": 0.6210973844231107,
|
| 1498 |
+
"learning_rate": 5.560349359695181e-06,
|
| 1499 |
+
"loss": 0.4092,
|
| 1500 |
+
"step": 213
|
| 1501 |
+
},
|
| 1502 |
+
{
|
| 1503 |
+
"epoch": 2.0511182108626196,
|
| 1504 |
+
"grad_norm": 0.44860747421683334,
|
| 1505 |
+
"learning_rate": 5.460095002604533e-06,
|
| 1506 |
+
"loss": 0.3753,
|
| 1507 |
+
"step": 214
|
| 1508 |
+
},
|
| 1509 |
+
{
|
| 1510 |
+
"epoch": 2.060702875399361,
|
| 1511 |
+
"grad_norm": 0.43389253461155286,
|
| 1512 |
+
"learning_rate": 5.360412158221661e-06,
|
| 1513 |
+
"loss": 0.356,
|
| 1514 |
+
"step": 215
|
| 1515 |
+
},
|
| 1516 |
+
{
|
| 1517 |
+
"epoch": 2.070287539936102,
|
| 1518 |
+
"grad_norm": 0.5565953093774033,
|
| 1519 |
+
"learning_rate": 5.2613133752700145e-06,
|
| 1520 |
+
"loss": 0.4177,
|
| 1521 |
+
"step": 216
|
| 1522 |
+
},
|
| 1523 |
+
{
|
| 1524 |
+
"epoch": 2.0798722044728435,
|
| 1525 |
+
"grad_norm": 0.5439364657270599,
|
| 1526 |
+
"learning_rate": 5.1628111289476025e-06,
|
| 1527 |
+
"loss": 0.3691,
|
| 1528 |
+
"step": 217
|
| 1529 |
+
},
|
| 1530 |
+
{
|
| 1531 |
+
"epoch": 2.0894568690095845,
|
| 1532 |
+
"grad_norm": 0.4250244085573423,
|
| 1533 |
+
"learning_rate": 5.064917819356532e-06,
|
| 1534 |
+
"loss": 0.3823,
|
| 1535 |
+
"step": 218
|
| 1536 |
+
},
|
| 1537 |
+
{
|
| 1538 |
+
"epoch": 2.099041533546326,
|
| 1539 |
+
"grad_norm": 0.4633828235328361,
|
| 1540 |
+
"learning_rate": 4.967645769942e-06,
|
| 1541 |
+
"loss": 0.3456,
|
| 1542 |
+
"step": 219
|
| 1543 |
+
},
|
| 1544 |
+
{
|
| 1545 |
+
"epoch": 2.108626198083067,
|
| 1546 |
+
"grad_norm": 0.4872033142578477,
|
| 1547 |
+
"learning_rate": 4.87100722594094e-06,
|
| 1548 |
+
"loss": 0.3792,
|
| 1549 |
+
"step": 220
|
| 1550 |
+
},
|
| 1551 |
+
{
|
| 1552 |
+
"epoch": 2.1182108626198084,
|
| 1553 |
+
"grad_norm": 0.434441186947985,
|
| 1554 |
+
"learning_rate": 4.775014352840512e-06,
|
| 1555 |
+
"loss": 0.3891,
|
| 1556 |
+
"step": 221
|
| 1557 |
+
},
|
| 1558 |
+
{
|
| 1559 |
+
"epoch": 2.1277955271565494,
|
| 1560 |
+
"grad_norm": 0.4065479253193355,
|
| 1561 |
+
"learning_rate": 4.679679234846636e-06,
|
| 1562 |
+
"loss": 0.35,
|
| 1563 |
+
"step": 222
|
| 1564 |
+
},
|
| 1565 |
+
{
|
| 1566 |
+
"epoch": 2.137380191693291,
|
| 1567 |
+
"grad_norm": 0.4617361261302904,
|
| 1568 |
+
"learning_rate": 4.5850138733627435e-06,
|
| 1569 |
+
"loss": 0.3809,
|
| 1570 |
+
"step": 223
|
| 1571 |
+
},
|
| 1572 |
+
{
|
| 1573 |
+
"epoch": 2.146964856230032,
|
| 1574 |
+
"grad_norm": 0.4239909798768424,
|
| 1575 |
+
"learning_rate": 4.491030185478976e-06,
|
| 1576 |
+
"loss": 0.3707,
|
| 1577 |
+
"step": 224
|
| 1578 |
+
},
|
| 1579 |
+
{
|
| 1580 |
+
"epoch": 2.1565495207667733,
|
| 1581 |
+
"grad_norm": 0.4556545694731096,
|
| 1582 |
+
"learning_rate": 4.397740002471973e-06,
|
| 1583 |
+
"loss": 0.4152,
|
| 1584 |
+
"step": 225
|
| 1585 |
+
},
|
| 1586 |
+
{
|
| 1587 |
+
"epoch": 2.1661341853035143,
|
| 1588 |
+
"grad_norm": 0.39176510868793746,
|
| 1589 |
+
"learning_rate": 4.305155068315481e-06,
|
| 1590 |
+
"loss": 0.3372,
|
| 1591 |
+
"step": 226
|
| 1592 |
+
},
|
| 1593 |
+
{
|
| 1594 |
+
"epoch": 2.1757188498402558,
|
| 1595 |
+
"grad_norm": 0.4030244459032741,
|
| 1596 |
+
"learning_rate": 4.213287038201943e-06,
|
| 1597 |
+
"loss": 0.3655,
|
| 1598 |
+
"step": 227
|
| 1599 |
+
},
|
| 1600 |
+
{
|
| 1601 |
+
"epoch": 2.1853035143769968,
|
| 1602 |
+
"grad_norm": 0.44687274849513065,
|
| 1603 |
+
"learning_rate": 4.12214747707527e-06,
|
| 1604 |
+
"loss": 0.406,
|
| 1605 |
+
"step": 228
|
| 1606 |
+
},
|
| 1607 |
+
{
|
| 1608 |
+
"epoch": 2.194888178913738,
|
| 1609 |
+
"grad_norm": 0.3944935531176264,
|
| 1610 |
+
"learning_rate": 4.0317478581749644e-06,
|
| 1611 |
+
"loss": 0.372,
|
| 1612 |
+
"step": 229
|
| 1613 |
+
},
|
| 1614 |
+
{
|
| 1615 |
+
"epoch": 2.2044728434504792,
|
| 1616 |
+
"grad_norm": 0.40549466106843335,
|
| 1617 |
+
"learning_rate": 3.942099561591802e-06,
|
| 1618 |
+
"loss": 0.4101,
|
| 1619 |
+
"step": 230
|
| 1620 |
+
},
|
| 1621 |
+
{
|
| 1622 |
+
"epoch": 2.2140575079872207,
|
| 1623 |
+
"grad_norm": 0.33462197396443605,
|
| 1624 |
+
"learning_rate": 3.853213872835229e-06,
|
| 1625 |
+
"loss": 0.3337,
|
| 1626 |
+
"step": 231
|
| 1627 |
+
},
|
| 1628 |
+
{
|
| 1629 |
+
"epoch": 2.2236421725239617,
|
| 1630 |
+
"grad_norm": 0.4099961550027212,
|
| 1631 |
+
"learning_rate": 3.7651019814126656e-06,
|
| 1632 |
+
"loss": 0.4058,
|
| 1633 |
+
"step": 232
|
| 1634 |
+
},
|
| 1635 |
+
{
|
| 1636 |
+
"epoch": 2.2332268370607027,
|
| 1637 |
+
"grad_norm": 0.4303748197397045,
|
| 1638 |
+
"learning_rate": 3.677774979420904e-06,
|
| 1639 |
+
"loss": 0.3918,
|
| 1640 |
+
"step": 233
|
| 1641 |
+
},
|
| 1642 |
+
{
|
| 1643 |
+
"epoch": 2.242811501597444,
|
| 1644 |
+
"grad_norm": 0.35212450130725803,
|
| 1645 |
+
"learning_rate": 3.591243860149759e-06,
|
| 1646 |
+
"loss": 0.3621,
|
| 1647 |
+
"step": 234
|
| 1648 |
+
},
|
| 1649 |
+
{
|
| 1650 |
+
"epoch": 2.252396166134185,
|
| 1651 |
+
"grad_norm": 0.35000139197525687,
|
| 1652 |
+
"learning_rate": 3.505519516698165e-06,
|
| 1653 |
+
"loss": 0.3925,
|
| 1654 |
+
"step": 235
|
| 1655 |
+
},
|
| 1656 |
+
{
|
| 1657 |
+
"epoch": 2.2619808306709266,
|
| 1658 |
+
"grad_norm": 0.34903321724384345,
|
| 1659 |
+
"learning_rate": 3.4206127406028744e-06,
|
| 1660 |
+
"loss": 0.3365,
|
| 1661 |
+
"step": 236
|
| 1662 |
+
},
|
| 1663 |
+
{
|
| 1664 |
+
"epoch": 2.2715654952076676,
|
| 1665 |
+
"grad_norm": 0.3495761955307319,
|
| 1666 |
+
"learning_rate": 3.3365342204799613e-06,
|
| 1667 |
+
"loss": 0.3639,
|
| 1668 |
+
"step": 237
|
| 1669 |
+
},
|
| 1670 |
+
{
|
| 1671 |
+
"epoch": 2.281150159744409,
|
| 1672 |
+
"grad_norm": 0.3559973478644341,
|
| 1673 |
+
"learning_rate": 3.2532945406792573e-06,
|
| 1674 |
+
"loss": 0.3617,
|
| 1675 |
+
"step": 238
|
| 1676 |
+
},
|
| 1677 |
+
{
|
| 1678 |
+
"epoch": 2.29073482428115,
|
| 1679 |
+
"grad_norm": 0.3830718690572721,
|
| 1680 |
+
"learning_rate": 3.1709041799519312e-06,
|
| 1681 |
+
"loss": 0.3911,
|
| 1682 |
+
"step": 239
|
| 1683 |
+
},
|
| 1684 |
+
{
|
| 1685 |
+
"epoch": 2.3003194888178915,
|
| 1686 |
+
"grad_norm": 0.3328630776804002,
|
| 1687 |
+
"learning_rate": 3.089373510131354e-06,
|
| 1688 |
+
"loss": 0.3543,
|
| 1689 |
+
"step": 240
|
| 1690 |
+
},
|
| 1691 |
+
{
|
| 1692 |
+
"epoch": 2.3099041533546325,
|
| 1693 |
+
"grad_norm": 0.31154242898607376,
|
| 1694 |
+
"learning_rate": 3.0087127948274264e-06,
|
| 1695 |
+
"loss": 0.3149,
|
| 1696 |
+
"step": 241
|
| 1697 |
+
},
|
| 1698 |
+
{
|
| 1699 |
+
"epoch": 2.319488817891374,
|
| 1700 |
+
"grad_norm": 0.3589233869362801,
|
| 1701 |
+
"learning_rate": 2.9289321881345257e-06,
|
| 1702 |
+
"loss": 0.3974,
|
| 1703 |
+
"step": 242
|
| 1704 |
+
},
|
| 1705 |
+
{
|
| 1706 |
+
"epoch": 2.329073482428115,
|
| 1707 |
+
"grad_norm": 0.32834847417622853,
|
| 1708 |
+
"learning_rate": 2.850041733353247e-06,
|
| 1709 |
+
"loss": 0.3367,
|
| 1710 |
+
"step": 243
|
| 1711 |
+
},
|
| 1712 |
+
{
|
| 1713 |
+
"epoch": 2.3386581469648564,
|
| 1714 |
+
"grad_norm": 0.35226107708331433,
|
| 1715 |
+
"learning_rate": 2.7720513617260857e-06,
|
| 1716 |
+
"loss": 0.3803,
|
| 1717 |
+
"step": 244
|
| 1718 |
+
},
|
| 1719 |
+
{
|
| 1720 |
+
"epoch": 2.3482428115015974,
|
| 1721 |
+
"grad_norm": 0.32975966920662375,
|
| 1722 |
+
"learning_rate": 2.694970891187225e-06,
|
| 1723 |
+
"loss": 0.3815,
|
| 1724 |
+
"step": 245
|
| 1725 |
+
},
|
| 1726 |
+
{
|
| 1727 |
+
"epoch": 2.357827476038339,
|
| 1728 |
+
"grad_norm": 0.35188866709000055,
|
| 1729 |
+
"learning_rate": 2.6188100251265947e-06,
|
| 1730 |
+
"loss": 0.3935,
|
| 1731 |
+
"step": 246
|
| 1732 |
+
},
|
| 1733 |
+
{
|
| 1734 |
+
"epoch": 2.36741214057508,
|
| 1735 |
+
"grad_norm": 0.3180683406057308,
|
| 1736 |
+
"learning_rate": 2.5435783511683444e-06,
|
| 1737 |
+
"loss": 0.3229,
|
| 1738 |
+
"step": 247
|
| 1739 |
+
},
|
| 1740 |
+
{
|
| 1741 |
+
"epoch": 2.376996805111821,
|
| 1742 |
+
"grad_norm": 0.3659434112836393,
|
| 1743 |
+
"learning_rate": 2.469285339963892e-06,
|
| 1744 |
+
"loss": 0.383,
|
| 1745 |
+
"step": 248
|
| 1746 |
+
},
|
| 1747 |
+
{
|
| 1748 |
+
"epoch": 2.3865814696485623,
|
| 1749 |
+
"grad_norm": 0.3548003669740148,
|
| 1750 |
+
"learning_rate": 2.395940343999691e-06,
|
| 1751 |
+
"loss": 0.3828,
|
| 1752 |
+
"step": 249
|
| 1753 |
+
},
|
| 1754 |
+
{
|
| 1755 |
+
"epoch": 2.3961661341853033,
|
| 1756 |
+
"grad_norm": 0.3271621525630648,
|
| 1757 |
+
"learning_rate": 2.323552596419889e-06,
|
| 1758 |
+
"loss": 0.3815,
|
| 1759 |
+
"step": 250
|
| 1760 |
+
},
|
| 1761 |
+
{
|
| 1762 |
+
"epoch": 2.405750798722045,
|
| 1763 |
+
"grad_norm": 0.33849807601480136,
|
| 1764 |
+
"learning_rate": 2.2521312098639914e-06,
|
| 1765 |
+
"loss": 0.3819,
|
| 1766 |
+
"step": 251
|
| 1767 |
+
},
|
| 1768 |
+
{
|
| 1769 |
+
"epoch": 2.415335463258786,
|
| 1770 |
+
"grad_norm": 0.3558165011110794,
|
| 1771 |
+
"learning_rate": 2.1816851753197023e-06,
|
| 1772 |
+
"loss": 0.4087,
|
| 1773 |
+
"step": 252
|
| 1774 |
+
},
|
| 1775 |
+
{
|
| 1776 |
+
"epoch": 2.4249201277955272,
|
| 1777 |
+
"grad_norm": 0.3353146638437276,
|
| 1778 |
+
"learning_rate": 2.1122233609910903e-06,
|
| 1779 |
+
"loss": 0.3657,
|
| 1780 |
+
"step": 253
|
| 1781 |
+
},
|
| 1782 |
+
{
|
| 1783 |
+
"epoch": 2.4345047923322682,
|
| 1784 |
+
"grad_norm": 0.33861735298320605,
|
| 1785 |
+
"learning_rate": 2.043754511182191e-06,
|
| 1786 |
+
"loss": 0.3457,
|
| 1787 |
+
"step": 254
|
| 1788 |
+
},
|
| 1789 |
+
{
|
| 1790 |
+
"epoch": 2.4440894568690097,
|
| 1791 |
+
"grad_norm": 0.35349259037933695,
|
| 1792 |
+
"learning_rate": 1.9762872451962214e-06,
|
| 1793 |
+
"loss": 0.3919,
|
| 1794 |
+
"step": 255
|
| 1795 |
+
},
|
| 1796 |
+
{
|
| 1797 |
+
"epoch": 2.4536741214057507,
|
| 1798 |
+
"grad_norm": 0.3027836405978437,
|
| 1799 |
+
"learning_rate": 1.9098300562505266e-06,
|
| 1800 |
+
"loss": 0.3691,
|
| 1801 |
+
"step": 256
|
| 1802 |
+
},
|
| 1803 |
+
{
|
| 1804 |
+
"epoch": 2.463258785942492,
|
| 1805 |
+
"grad_norm": 0.3156869857843323,
|
| 1806 |
+
"learning_rate": 1.8443913104073984e-06,
|
| 1807 |
+
"loss": 0.364,
|
| 1808 |
+
"step": 257
|
| 1809 |
+
},
|
| 1810 |
+
{
|
| 1811 |
+
"epoch": 2.472843450479233,
|
| 1812 |
+
"grad_norm": 0.3386183902480921,
|
| 1813 |
+
"learning_rate": 1.7799792455209019e-06,
|
| 1814 |
+
"loss": 0.3308,
|
| 1815 |
+
"step": 258
|
| 1816 |
+
},
|
| 1817 |
+
{
|
| 1818 |
+
"epoch": 2.4824281150159746,
|
| 1819 |
+
"grad_norm": 0.3665955325247484,
|
| 1820 |
+
"learning_rate": 1.716601970199836e-06,
|
| 1821 |
+
"loss": 0.3769,
|
| 1822 |
+
"step": 259
|
| 1823 |
+
},
|
| 1824 |
+
{
|
| 1825 |
+
"epoch": 2.4920127795527156,
|
| 1826 |
+
"grad_norm": 0.36101096856608017,
|
| 1827 |
+
"learning_rate": 1.6542674627869738e-06,
|
| 1828 |
+
"loss": 0.4193,
|
| 1829 |
+
"step": 260
|
| 1830 |
+
},
|
| 1831 |
+
{
|
| 1832 |
+
"epoch": 2.501597444089457,
|
| 1833 |
+
"grad_norm": 0.3385060552291423,
|
| 1834 |
+
"learning_rate": 1.5929835703546992e-06,
|
| 1835 |
+
"loss": 0.3798,
|
| 1836 |
+
"step": 261
|
| 1837 |
+
},
|
| 1838 |
+
{
|
| 1839 |
+
"epoch": 2.511182108626198,
|
| 1840 |
+
"grad_norm": 0.3429361880047483,
|
| 1841 |
+
"learning_rate": 1.5327580077171589e-06,
|
| 1842 |
+
"loss": 0.3632,
|
| 1843 |
+
"step": 262
|
| 1844 |
+
},
|
| 1845 |
+
{
|
| 1846 |
+
"epoch": 2.520766773162939,
|
| 1847 |
+
"grad_norm": 0.33218156060244974,
|
| 1848 |
+
"learning_rate": 1.4735983564590784e-06,
|
| 1849 |
+
"loss": 0.3762,
|
| 1850 |
+
"step": 263
|
| 1851 |
+
},
|
| 1852 |
+
{
|
| 1853 |
+
"epoch": 2.5303514376996805,
|
| 1854 |
+
"grad_norm": 0.33211325486143817,
|
| 1855 |
+
"learning_rate": 1.4155120639813392e-06,
|
| 1856 |
+
"loss": 0.3675,
|
| 1857 |
+
"step": 264
|
| 1858 |
+
},
|
| 1859 |
+
{
|
| 1860 |
+
"epoch": 2.539936102236422,
|
| 1861 |
+
"grad_norm": 0.33667239529882725,
|
| 1862 |
+
"learning_rate": 1.3585064425634542e-06,
|
| 1863 |
+
"loss": 0.3627,
|
| 1864 |
+
"step": 265
|
| 1865 |
+
},
|
| 1866 |
+
{
|
| 1867 |
+
"epoch": 2.549520766773163,
|
| 1868 |
+
"grad_norm": 0.3520337488712381,
|
| 1869 |
+
"learning_rate": 1.3025886684430467e-06,
|
| 1870 |
+
"loss": 0.4046,
|
| 1871 |
+
"step": 266
|
| 1872 |
+
},
|
| 1873 |
+
{
|
| 1874 |
+
"epoch": 2.559105431309904,
|
| 1875 |
+
"grad_norm": 0.3238937013404258,
|
| 1876 |
+
"learning_rate": 1.2477657809124632e-06,
|
| 1877 |
+
"loss": 0.3789,
|
| 1878 |
+
"step": 267
|
| 1879 |
+
},
|
| 1880 |
+
{
|
| 1881 |
+
"epoch": 2.5686900958466454,
|
| 1882 |
+
"grad_norm": 0.3268026171586817,
|
| 1883 |
+
"learning_rate": 1.19404468143262e-06,
|
| 1884 |
+
"loss": 0.3703,
|
| 1885 |
+
"step": 268
|
| 1886 |
+
},
|
| 1887 |
+
{
|
| 1888 |
+
"epoch": 2.5782747603833864,
|
| 1889 |
+
"grad_norm": 0.3255582279602733,
|
| 1890 |
+
"learning_rate": 1.1414321327642019e-06,
|
| 1891 |
+
"loss": 0.3394,
|
| 1892 |
+
"step": 269
|
| 1893 |
+
},
|
| 1894 |
+
{
|
| 1895 |
+
"epoch": 2.587859424920128,
|
| 1896 |
+
"grad_norm": 0.31640120333334965,
|
| 1897 |
+
"learning_rate": 1.0899347581163222e-06,
|
| 1898 |
+
"loss": 0.3589,
|
| 1899 |
+
"step": 270
|
| 1900 |
+
},
|
| 1901 |
+
{
|
| 1902 |
+
"epoch": 2.597444089456869,
|
| 1903 |
+
"grad_norm": 0.3426513122706542,
|
| 1904 |
+
"learning_rate": 1.0395590403127487e-06,
|
| 1905 |
+
"loss": 0.3832,
|
| 1906 |
+
"step": 271
|
| 1907 |
+
},
|
| 1908 |
+
{
|
| 1909 |
+
"epoch": 2.6070287539936103,
|
| 1910 |
+
"grad_norm": 0.2998344813799427,
|
| 1911 |
+
"learning_rate": 9.903113209758098e-07,
|
| 1912 |
+
"loss": 0.3646,
|
| 1913 |
+
"step": 272
|
| 1914 |
+
},
|
| 1915 |
+
{
|
| 1916 |
+
"epoch": 2.6166134185303513,
|
| 1917 |
+
"grad_norm": 0.3271156026842227,
|
| 1918 |
+
"learning_rate": 9.421977997280596e-07,
|
| 1919 |
+
"loss": 0.3973,
|
| 1920 |
+
"step": 273
|
| 1921 |
+
},
|
| 1922 |
+
{
|
| 1923 |
+
"epoch": 2.626198083067093,
|
| 1924 |
+
"grad_norm": 0.3276947619903817,
|
| 1925 |
+
"learning_rate": 8.952245334118415e-07,
|
| 1926 |
+
"loss": 0.3828,
|
| 1927 |
+
"step": 274
|
| 1928 |
+
},
|
| 1929 |
+
{
|
| 1930 |
+
"epoch": 2.635782747603834,
|
| 1931 |
+
"grad_norm": 0.3208361600065217,
|
| 1932 |
+
"learning_rate": 8.493974353268019e-07,
|
| 1933 |
+
"loss": 0.3961,
|
| 1934 |
+
"step": 275
|
| 1935 |
+
},
|
| 1936 |
+
{
|
| 1937 |
+
"epoch": 2.6453674121405752,
|
| 1938 |
+
"grad_norm": 0.3187235157801297,
|
| 1939 |
+
"learning_rate": 8.047222744854943e-07,
|
| 1940 |
+
"loss": 0.3796,
|
| 1941 |
+
"step": 276
|
| 1942 |
+
},
|
| 1943 |
+
{
|
| 1944 |
+
"epoch": 2.6549520766773163,
|
| 1945 |
+
"grad_norm": 0.31827291746375636,
|
| 1946 |
+
"learning_rate": 7.612046748871327e-07,
|
| 1947 |
+
"loss": 0.3777,
|
| 1948 |
+
"step": 277
|
| 1949 |
+
},
|
| 1950 |
+
{
|
| 1951 |
+
"epoch": 2.6645367412140573,
|
| 1952 |
+
"grad_norm": 0.314592617245762,
|
| 1953 |
+
"learning_rate": 7.188501148096117e-07,
|
| 1954 |
+
"loss": 0.354,
|
| 1955 |
+
"step": 278
|
| 1956 |
+
},
|
| 1957 |
+
{
|
| 1958 |
+
"epoch": 2.6741214057507987,
|
| 1959 |
+
"grad_norm": 0.3411145251233779,
|
| 1960 |
+
"learning_rate": 6.776639261198581e-07,
|
| 1961 |
+
"loss": 0.4123,
|
| 1962 |
+
"step": 279
|
| 1963 |
+
},
|
| 1964 |
+
{
|
| 1965 |
+
"epoch": 2.68370607028754,
|
| 1966 |
+
"grad_norm": 0.30394694643551917,
|
| 1967 |
+
"learning_rate": 6.37651293602628e-07,
|
| 1968 |
+
"loss": 0.3542,
|
| 1969 |
+
"step": 280
|
| 1970 |
+
},
|
| 1971 |
+
{
|
| 1972 |
+
"epoch": 2.693290734824281,
|
| 1973 |
+
"grad_norm": 0.3294079402554833,
|
| 1974 |
+
"learning_rate": 5.988172543078097e-07,
|
| 1975 |
+
"loss": 0.3847,
|
| 1976 |
+
"step": 281
|
| 1977 |
+
},
|
| 1978 |
+
{
|
| 1979 |
+
"epoch": 2.702875399361022,
|
| 1980 |
+
"grad_norm": 0.3286750975459561,
|
| 1981 |
+
"learning_rate": 5.611666969163243e-07,
|
| 1982 |
+
"loss": 0.374,
|
| 1983 |
+
"step": 282
|
| 1984 |
+
},
|
| 1985 |
+
{
|
| 1986 |
+
"epoch": 2.7124600638977636,
|
| 1987 |
+
"grad_norm": 0.31556283150275327,
|
| 1988 |
+
"learning_rate": 5.247043611247127e-07,
|
| 1989 |
+
"loss": 0.3634,
|
| 1990 |
+
"step": 283
|
| 1991 |
+
},
|
| 1992 |
+
{
|
| 1993 |
+
"epoch": 2.722044728434505,
|
| 1994 |
+
"grad_norm": 0.28616841430584555,
|
| 1995 |
+
"learning_rate": 4.894348370484648e-07,
|
| 1996 |
+
"loss": 0.3363,
|
| 1997 |
+
"step": 284
|
| 1998 |
+
},
|
| 1999 |
+
{
|
| 2000 |
+
"epoch": 2.731629392971246,
|
| 2001 |
+
"grad_norm": 0.3347741210841466,
|
| 2002 |
+
"learning_rate": 4.553625646441928e-07,
|
| 2003 |
+
"loss": 0.3938,
|
| 2004 |
+
"step": 285
|
| 2005 |
+
},
|
| 2006 |
+
{
|
| 2007 |
+
"epoch": 2.741214057507987,
|
| 2008 |
+
"grad_norm": 0.303400322030674,
|
| 2009 |
+
"learning_rate": 4.224918331506955e-07,
|
| 2010 |
+
"loss": 0.3544,
|
| 2011 |
+
"step": 286
|
| 2012 |
+
},
|
| 2013 |
+
{
|
| 2014 |
+
"epoch": 2.7507987220447285,
|
| 2015 |
+
"grad_norm": 0.32472397378256507,
|
| 2016 |
+
"learning_rate": 3.908267805490051e-07,
|
| 2017 |
+
"loss": 0.3973,
|
| 2018 |
+
"step": 287
|
| 2019 |
+
},
|
| 2020 |
+
{
|
| 2021 |
+
"epoch": 2.7603833865814695,
|
| 2022 |
+
"grad_norm": 0.31173649557105365,
|
| 2023 |
+
"learning_rate": 3.603713930414676e-07,
|
| 2024 |
+
"loss": 0.3991,
|
| 2025 |
+
"step": 288
|
| 2026 |
+
},
|
| 2027 |
+
{
|
| 2028 |
+
"epoch": 2.769968051118211,
|
| 2029 |
+
"grad_norm": 0.3137833568204568,
|
| 2030 |
+
"learning_rate": 3.3112950454993633e-07,
|
| 2031 |
+
"loss": 0.3604,
|
| 2032 |
+
"step": 289
|
| 2033 |
+
},
|
| 2034 |
+
{
|
| 2035 |
+
"epoch": 2.779552715654952,
|
| 2036 |
+
"grad_norm": 0.30759530236879995,
|
| 2037 |
+
"learning_rate": 3.0310479623313125e-07,
|
| 2038 |
+
"loss": 0.3654,
|
| 2039 |
+
"step": 290
|
| 2040 |
+
},
|
| 2041 |
+
{
|
| 2042 |
+
"epoch": 2.7891373801916934,
|
| 2043 |
+
"grad_norm": 0.3181021720539179,
|
| 2044 |
+
"learning_rate": 2.7630079602323447e-07,
|
| 2045 |
+
"loss": 0.427,
|
| 2046 |
+
"step": 291
|
| 2047 |
+
},
|
| 2048 |
+
{
|
| 2049 |
+
"epoch": 2.7987220447284344,
|
| 2050 |
+
"grad_norm": 0.29030962682069206,
|
| 2051 |
+
"learning_rate": 2.507208781817638e-07,
|
| 2052 |
+
"loss": 0.3213,
|
| 2053 |
+
"step": 292
|
| 2054 |
+
},
|
| 2055 |
+
{
|
| 2056 |
+
"epoch": 2.8083067092651754,
|
| 2057 |
+
"grad_norm": 0.32615431635966197,
|
| 2058 |
+
"learning_rate": 2.2636826287480872e-07,
|
| 2059 |
+
"loss": 0.372,
|
| 2060 |
+
"step": 293
|
| 2061 |
+
},
|
| 2062 |
+
{
|
| 2063 |
+
"epoch": 2.817891373801917,
|
| 2064 |
+
"grad_norm": 0.29846601491141916,
|
| 2065 |
+
"learning_rate": 2.0324601576764525e-07,
|
| 2066 |
+
"loss": 0.3641,
|
| 2067 |
+
"step": 294
|
| 2068 |
+
},
|
| 2069 |
+
{
|
| 2070 |
+
"epoch": 2.8274760383386583,
|
| 2071 |
+
"grad_norm": 0.34873022597553716,
|
| 2072 |
+
"learning_rate": 1.81357047638816e-07,
|
| 2073 |
+
"loss": 0.4412,
|
| 2074 |
+
"step": 295
|
| 2075 |
+
},
|
| 2076 |
+
{
|
| 2077 |
+
"epoch": 2.8370607028753994,
|
| 2078 |
+
"grad_norm": 0.3147497934101791,
|
| 2079 |
+
"learning_rate": 1.6070411401370335e-07,
|
| 2080 |
+
"loss": 0.3375,
|
| 2081 |
+
"step": 296
|
| 2082 |
+
},
|
| 2083 |
+
{
|
| 2084 |
+
"epoch": 2.8466453674121404,
|
| 2085 |
+
"grad_norm": 0.3104782796474926,
|
| 2086 |
+
"learning_rate": 1.4128981481764115e-07,
|
| 2087 |
+
"loss": 0.398,
|
| 2088 |
+
"step": 297
|
| 2089 |
+
},
|
| 2090 |
+
{
|
| 2091 |
+
"epoch": 2.856230031948882,
|
| 2092 |
+
"grad_norm": 0.3140238921021234,
|
| 2093 |
+
"learning_rate": 1.231165940486234e-07,
|
| 2094 |
+
"loss": 0.3686,
|
| 2095 |
+
"step": 298
|
| 2096 |
+
},
|
| 2097 |
+
{
|
| 2098 |
+
"epoch": 2.8658146964856233,
|
| 2099 |
+
"grad_norm": 0.4288142439350452,
|
| 2100 |
+
"learning_rate": 1.0618673946963365e-07,
|
| 2101 |
+
"loss": 0.4175,
|
| 2102 |
+
"step": 299
|
| 2103 |
+
},
|
| 2104 |
+
{
|
| 2105 |
+
"epoch": 2.8753993610223643,
|
| 2106 |
+
"grad_norm": 0.2842824480889967,
|
| 2107 |
+
"learning_rate": 9.0502382320653e-08,
|
| 2108 |
+
"loss": 0.3317,
|
| 2109 |
+
"step": 300
|
| 2110 |
+
},
|
| 2111 |
+
{
|
| 2112 |
+
"epoch": 2.8849840255591053,
|
| 2113 |
+
"grad_norm": 0.3129493055661085,
|
| 2114 |
+
"learning_rate": 7.606549705035937e-08,
|
| 2115 |
+
"loss": 0.3893,
|
| 2116 |
+
"step": 301
|
| 2117 |
+
},
|
| 2118 |
+
{
|
| 2119 |
+
"epoch": 2.8945686900958467,
|
| 2120 |
+
"grad_norm": 0.31085989958792076,
|
| 2121 |
+
"learning_rate": 6.287790106757396e-08,
|
| 2122 |
+
"loss": 0.411,
|
| 2123 |
+
"step": 302
|
| 2124 |
+
},
|
| 2125 |
+
{
|
| 2126 |
+
"epoch": 2.9041533546325877,
|
| 2127 |
+
"grad_norm": 0.29788052650610436,
|
| 2128 |
+
"learning_rate": 5.094125451247656e-08,
|
| 2129 |
+
"loss": 0.3882,
|
| 2130 |
+
"step": 303
|
| 2131 |
+
},
|
| 2132 |
+
{
|
| 2133 |
+
"epoch": 2.913738019169329,
|
| 2134 |
+
"grad_norm": 0.330785113441627,
|
| 2135 |
+
"learning_rate": 4.025706004760932e-08,
|
| 2136 |
+
"loss": 0.4154,
|
| 2137 |
+
"step": 304
|
| 2138 |
+
},
|
| 2139 |
+
{
|
| 2140 |
+
"epoch": 2.92332268370607,
|
| 2141 |
+
"grad_norm": 0.29464975259883275,
|
| 2142 |
+
"learning_rate": 3.082666266872036e-08,
|
| 2143 |
+
"loss": 0.3293,
|
| 2144 |
+
"step": 305
|
| 2145 |
+
},
|
| 2146 |
+
{
|
| 2147 |
+
"epoch": 2.9329073482428116,
|
| 2148 |
+
"grad_norm": 0.29689954664312995,
|
| 2149 |
+
"learning_rate": 2.265124953543918e-08,
|
| 2150 |
+
"loss": 0.3585,
|
| 2151 |
+
"step": 306
|
| 2152 |
+
},
|
| 2153 |
+
{
|
| 2154 |
+
"epoch": 2.9424920127795526,
|
| 2155 |
+
"grad_norm": 0.3533149363735046,
|
| 2156 |
+
"learning_rate": 1.5731849821833955e-08,
|
| 2157 |
+
"loss": 0.4593,
|
| 2158 |
+
"step": 307
|
| 2159 |
+
},
|
| 2160 |
+
{
|
| 2161 |
+
"epoch": 2.952076677316294,
|
| 2162 |
+
"grad_norm": 0.28841598481778036,
|
| 2163 |
+
"learning_rate": 1.0069334586854106e-08,
|
| 2164 |
+
"loss": 0.3501,
|
| 2165 |
+
"step": 308
|
| 2166 |
+
},
|
| 2167 |
+
{
|
| 2168 |
+
"epoch": 2.961661341853035,
|
| 2169 |
+
"grad_norm": 0.3018611204905917,
|
| 2170 |
+
"learning_rate": 5.664416664666883e-09,
|
| 2171 |
+
"loss": 0.3676,
|
| 2172 |
+
"step": 309
|
| 2173 |
+
},
|
| 2174 |
+
{
|
| 2175 |
+
"epoch": 2.9712460063897765,
|
| 2176 |
+
"grad_norm": 0.3072862218552703,
|
| 2177 |
+
"learning_rate": 2.5176505749346937e-09,
|
| 2178 |
+
"loss": 0.3948,
|
| 2179 |
+
"step": 310
|
| 2180 |
+
},
|
| 2181 |
+
{
|
| 2182 |
+
"epoch": 2.9808306709265175,
|
| 2183 |
+
"grad_norm": 0.29271951419638015,
|
| 2184 |
+
"learning_rate": 6.294324529942942e-10,
|
| 2185 |
+
"loss": 0.3643,
|
| 2186 |
+
"step": 311
|
| 2187 |
+
},
|
| 2188 |
+
{
|
| 2189 |
+
"epoch": 2.9904153354632586,
|
| 2190 |
+
"grad_norm": 0.30292122661734633,
|
| 2191 |
+
"learning_rate": 0.0,
|
| 2192 |
+
"loss": 0.3668,
|
| 2193 |
+
"step": 312
|
| 2194 |
+
},
|
| 2195 |
+
{
|
| 2196 |
+
"epoch": 2.9904153354632586,
|
| 2197 |
+
"step": 312,
|
| 2198 |
+
"total_flos": 4.4115474684667494e+17,
|
| 2199 |
+
"train_loss": 0.46701200955953354,
|
| 2200 |
+
"train_runtime": 9975.6836,
|
| 2201 |
+
"train_samples_per_second": 3.006,
|
| 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
|