End of training
Browse files- README.md +2 -1
- all_results.json +8 -0
- train_results.json +8 -0
- trainer_state.json +2373 -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: global_batchsize_1024_laradjusted16
|
|
@@ -15,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 15 |
|
| 16 |
# global_batchsize_1024_laradjusted16
|
| 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: global_batchsize_1024_laradjusted16
|
|
|
|
| 16 |
|
| 17 |
# global_batchsize_1024_laradjusted16
|
| 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 open-thoughts/OpenThoughts-114k dataset.
|
| 20 |
|
| 21 |
## Model description
|
| 22 |
|
all_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 2.9943851768669285,
|
| 3 |
+
"total_flos": 4607052149424128.0,
|
| 4 |
+
"train_loss": 0.4566754535869793,
|
| 5 |
+
"train_runtime": 115300.1106,
|
| 6 |
+
"train_samples_per_second": 2.965,
|
| 7 |
+
"train_steps_per_second": 0.003
|
| 8 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 2.9943851768669285,
|
| 3 |
+
"total_flos": 4607052149424128.0,
|
| 4 |
+
"train_loss": 0.4566754535869793,
|
| 5 |
+
"train_runtime": 115300.1106,
|
| 6 |
+
"train_samples_per_second": 2.965,
|
| 7 |
+
"train_steps_per_second": 0.003
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,2373 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 2.9943851768669285,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 333,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.008983717012914094,
|
| 13 |
+
"grad_norm": 5.839048375887439,
|
| 14 |
+
"learning_rate": 2.3529411764705885e-06,
|
| 15 |
+
"loss": 0.851,
|
| 16 |
+
"step": 1
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.017967434025828188,
|
| 20 |
+
"grad_norm": 5.796622055110998,
|
| 21 |
+
"learning_rate": 4.705882352941177e-06,
|
| 22 |
+
"loss": 0.8589,
|
| 23 |
+
"step": 2
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.02695115103874228,
|
| 27 |
+
"grad_norm": 5.449611767460796,
|
| 28 |
+
"learning_rate": 7.058823529411766e-06,
|
| 29 |
+
"loss": 0.851,
|
| 30 |
+
"step": 3
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.035934868051656375,
|
| 34 |
+
"grad_norm": 3.918769913446889,
|
| 35 |
+
"learning_rate": 9.411764705882354e-06,
|
| 36 |
+
"loss": 0.8013,
|
| 37 |
+
"step": 4
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.044918585064570464,
|
| 41 |
+
"grad_norm": 2.1146491352495644,
|
| 42 |
+
"learning_rate": 1.1764705882352942e-05,
|
| 43 |
+
"loss": 0.7675,
|
| 44 |
+
"step": 5
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.05390230207748456,
|
| 48 |
+
"grad_norm": 5.3866836077244535,
|
| 49 |
+
"learning_rate": 1.4117647058823532e-05,
|
| 50 |
+
"loss": 0.8021,
|
| 51 |
+
"step": 6
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.06288601909039865,
|
| 55 |
+
"grad_norm": 7.1506220356372205,
|
| 56 |
+
"learning_rate": 1.647058823529412e-05,
|
| 57 |
+
"loss": 0.7729,
|
| 58 |
+
"step": 7
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.07186973610331275,
|
| 62 |
+
"grad_norm": 7.3684198638012575,
|
| 63 |
+
"learning_rate": 1.8823529411764708e-05,
|
| 64 |
+
"loss": 0.7895,
|
| 65 |
+
"step": 8
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.08085345311622684,
|
| 69 |
+
"grad_norm": 4.2008371048182624,
|
| 70 |
+
"learning_rate": 2.1176470588235296e-05,
|
| 71 |
+
"loss": 0.7502,
|
| 72 |
+
"step": 9
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.08983717012914093,
|
| 76 |
+
"grad_norm": 3.018632704832958,
|
| 77 |
+
"learning_rate": 2.3529411764705884e-05,
|
| 78 |
+
"loss": 0.7085,
|
| 79 |
+
"step": 10
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.09882088714205503,
|
| 83 |
+
"grad_norm": 2.4103057201747564,
|
| 84 |
+
"learning_rate": 2.5882352941176475e-05,
|
| 85 |
+
"loss": 0.68,
|
| 86 |
+
"step": 11
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.10780460415496912,
|
| 90 |
+
"grad_norm": 1.6029821095889687,
|
| 91 |
+
"learning_rate": 2.8235294117647063e-05,
|
| 92 |
+
"loss": 0.6579,
|
| 93 |
+
"step": 12
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.11678832116788321,
|
| 97 |
+
"grad_norm": 1.5359955676401171,
|
| 98 |
+
"learning_rate": 3.0588235294117644e-05,
|
| 99 |
+
"loss": 0.6348,
|
| 100 |
+
"step": 13
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.1257720381807973,
|
| 104 |
+
"grad_norm": 1.2677879393997815,
|
| 105 |
+
"learning_rate": 3.294117647058824e-05,
|
| 106 |
+
"loss": 0.6259,
|
| 107 |
+
"step": 14
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.13475575519371139,
|
| 111 |
+
"grad_norm": 1.1307239717924904,
|
| 112 |
+
"learning_rate": 3.529411764705883e-05,
|
| 113 |
+
"loss": 0.6217,
|
| 114 |
+
"step": 15
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.1437394722066255,
|
| 118 |
+
"grad_norm": 1.1919595690235936,
|
| 119 |
+
"learning_rate": 3.7647058823529415e-05,
|
| 120 |
+
"loss": 0.6117,
|
| 121 |
+
"step": 16
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.1527231892195396,
|
| 125 |
+
"grad_norm": 1.5137586529647031,
|
| 126 |
+
"learning_rate": 4e-05,
|
| 127 |
+
"loss": 0.5951,
|
| 128 |
+
"step": 17
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.16170690623245368,
|
| 132 |
+
"grad_norm": 1.0839410029716163,
|
| 133 |
+
"learning_rate": 4.235294117647059e-05,
|
| 134 |
+
"loss": 0.5873,
|
| 135 |
+
"step": 18
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.17069062324536777,
|
| 139 |
+
"grad_norm": 1.411013539783664,
|
| 140 |
+
"learning_rate": 4.470588235294118e-05,
|
| 141 |
+
"loss": 0.5855,
|
| 142 |
+
"step": 19
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.17967434025828186,
|
| 146 |
+
"grad_norm": 0.9604494585660002,
|
| 147 |
+
"learning_rate": 4.705882352941177e-05,
|
| 148 |
+
"loss": 0.5855,
|
| 149 |
+
"step": 20
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.18865805727119594,
|
| 153 |
+
"grad_norm": 1.7966858442462152,
|
| 154 |
+
"learning_rate": 4.941176470588236e-05,
|
| 155 |
+
"loss": 0.5834,
|
| 156 |
+
"step": 21
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.19764177428411006,
|
| 160 |
+
"grad_norm": 1.2639231326008076,
|
| 161 |
+
"learning_rate": 5.176470588235295e-05,
|
| 162 |
+
"loss": 0.573,
|
| 163 |
+
"step": 22
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.20662549129702415,
|
| 167 |
+
"grad_norm": 1.6189242064013842,
|
| 168 |
+
"learning_rate": 5.411764705882354e-05,
|
| 169 |
+
"loss": 0.5687,
|
| 170 |
+
"step": 23
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.21560920830993824,
|
| 174 |
+
"grad_norm": 1.2918939880837212,
|
| 175 |
+
"learning_rate": 5.6470588235294126e-05,
|
| 176 |
+
"loss": 0.5629,
|
| 177 |
+
"step": 24
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.22459292532285233,
|
| 181 |
+
"grad_norm": 1.8153484123304104,
|
| 182 |
+
"learning_rate": 5.8823529411764714e-05,
|
| 183 |
+
"loss": 0.5644,
|
| 184 |
+
"step": 25
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.23357664233576642,
|
| 188 |
+
"grad_norm": 1.383080233694585,
|
| 189 |
+
"learning_rate": 6.117647058823529e-05,
|
| 190 |
+
"loss": 0.561,
|
| 191 |
+
"step": 26
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.2425603593486805,
|
| 195 |
+
"grad_norm": 1.908843640323105,
|
| 196 |
+
"learning_rate": 6.352941176470589e-05,
|
| 197 |
+
"loss": 0.5581,
|
| 198 |
+
"step": 27
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.2515440763615946,
|
| 202 |
+
"grad_norm": 1.5638595834141664,
|
| 203 |
+
"learning_rate": 6.588235294117648e-05,
|
| 204 |
+
"loss": 0.5497,
|
| 205 |
+
"step": 28
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.2605277933745087,
|
| 209 |
+
"grad_norm": 1.0799402527915973,
|
| 210 |
+
"learning_rate": 6.823529411764707e-05,
|
| 211 |
+
"loss": 0.5405,
|
| 212 |
+
"step": 29
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.26951151038742277,
|
| 216 |
+
"grad_norm": 1.2842268786606617,
|
| 217 |
+
"learning_rate": 7.058823529411765e-05,
|
| 218 |
+
"loss": 0.5489,
|
| 219 |
+
"step": 30
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.2784952274003369,
|
| 223 |
+
"grad_norm": 2.1296312845067953,
|
| 224 |
+
"learning_rate": 7.294117647058824e-05,
|
| 225 |
+
"loss": 0.5429,
|
| 226 |
+
"step": 31
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.287478944413251,
|
| 230 |
+
"grad_norm": 1.2081689918064082,
|
| 231 |
+
"learning_rate": 7.529411764705883e-05,
|
| 232 |
+
"loss": 0.5439,
|
| 233 |
+
"step": 32
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.29646266142616506,
|
| 237 |
+
"grad_norm": 2.681914408101531,
|
| 238 |
+
"learning_rate": 7.764705882352942e-05,
|
| 239 |
+
"loss": 0.5409,
|
| 240 |
+
"step": 33
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.3054463784390792,
|
| 244 |
+
"grad_norm": 2.088525203831682,
|
| 245 |
+
"learning_rate": 8e-05,
|
| 246 |
+
"loss": 0.5448,
|
| 247 |
+
"step": 34
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.31443009545199324,
|
| 251 |
+
"grad_norm": 2.429087878227431,
|
| 252 |
+
"learning_rate": 7.999779207981935e-05,
|
| 253 |
+
"loss": 0.5376,
|
| 254 |
+
"step": 35
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.32341381246490736,
|
| 258 |
+
"grad_norm": 2.2118789436187467,
|
| 259 |
+
"learning_rate": 7.999116856302298e-05,
|
| 260 |
+
"loss": 0.5429,
|
| 261 |
+
"step": 36
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.3323975294778215,
|
| 265 |
+
"grad_norm": 1.998359697047038,
|
| 266 |
+
"learning_rate": 7.998013018082072e-05,
|
| 267 |
+
"loss": 0.5372,
|
| 268 |
+
"step": 37
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.34138124649073553,
|
| 272 |
+
"grad_norm": 1.4739747352645118,
|
| 273 |
+
"learning_rate": 7.996467815180588e-05,
|
| 274 |
+
"loss": 0.5306,
|
| 275 |
+
"step": 38
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.35036496350364965,
|
| 279 |
+
"grad_norm": 1.8627456810217522,
|
| 280 |
+
"learning_rate": 7.994481418182082e-05,
|
| 281 |
+
"loss": 0.5309,
|
| 282 |
+
"step": 39
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.3593486805165637,
|
| 286 |
+
"grad_norm": 1.2832017910437872,
|
| 287 |
+
"learning_rate": 7.992054046376854e-05,
|
| 288 |
+
"loss": 0.5301,
|
| 289 |
+
"step": 40
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.36833239752947783,
|
| 293 |
+
"grad_norm": 1.9368339421908154,
|
| 294 |
+
"learning_rate": 7.989185967737066e-05,
|
| 295 |
+
"loss": 0.534,
|
| 296 |
+
"step": 41
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.3773161145423919,
|
| 300 |
+
"grad_norm": 1.4631687551659387,
|
| 301 |
+
"learning_rate": 7.985877498887149e-05,
|
| 302 |
+
"loss": 0.526,
|
| 303 |
+
"step": 42
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.386299831555306,
|
| 307 |
+
"grad_norm": 1.4992485001174218,
|
| 308 |
+
"learning_rate": 7.982129005068865e-05,
|
| 309 |
+
"loss": 0.5169,
|
| 310 |
+
"step": 43
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.3952835485682201,
|
| 314 |
+
"grad_norm": 1.2556436640866477,
|
| 315 |
+
"learning_rate": 7.977940900100967e-05,
|
| 316 |
+
"loss": 0.5086,
|
| 317 |
+
"step": 44
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.4042672655811342,
|
| 321 |
+
"grad_norm": 1.4032135906881233,
|
| 322 |
+
"learning_rate": 7.973313646333532e-05,
|
| 323 |
+
"loss": 0.5214,
|
| 324 |
+
"step": 45
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.4132509825940483,
|
| 328 |
+
"grad_norm": 1.3909591395416454,
|
| 329 |
+
"learning_rate": 7.968247754596908e-05,
|
| 330 |
+
"loss": 0.5143,
|
| 331 |
+
"step": 46
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.42223469960696236,
|
| 335 |
+
"grad_norm": 1.9540476320319087,
|
| 336 |
+
"learning_rate": 7.962743784145323e-05,
|
| 337 |
+
"loss": 0.5207,
|
| 338 |
+
"step": 47
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.4312184166198765,
|
| 342 |
+
"grad_norm": 1.0309999384978459,
|
| 343 |
+
"learning_rate": 7.956802342595152e-05,
|
| 344 |
+
"loss": 0.5069,
|
| 345 |
+
"step": 48
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.4402021336327906,
|
| 349 |
+
"grad_norm": 1.3811159742393722,
|
| 350 |
+
"learning_rate": 7.950424085857827e-05,
|
| 351 |
+
"loss": 0.5145,
|
| 352 |
+
"step": 49
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.44918585064570465,
|
| 356 |
+
"grad_norm": 1.4750899151595693,
|
| 357 |
+
"learning_rate": 7.943609718067437e-05,
|
| 358 |
+
"loss": 0.5034,
|
| 359 |
+
"step": 50
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.45816956765861877,
|
| 363 |
+
"grad_norm": 1.586345005999154,
|
| 364 |
+
"learning_rate": 7.936359991502993e-05,
|
| 365 |
+
"loss": 0.5139,
|
| 366 |
+
"step": 51
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.46715328467153283,
|
| 370 |
+
"grad_norm": 0.8294429144133623,
|
| 371 |
+
"learning_rate": 7.92867570650537e-05,
|
| 372 |
+
"loss": 0.5024,
|
| 373 |
+
"step": 52
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.47613700168444695,
|
| 377 |
+
"grad_norm": 1.4768401149936898,
|
| 378 |
+
"learning_rate": 7.920557711388967e-05,
|
| 379 |
+
"loss": 0.5024,
|
| 380 |
+
"step": 53
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.485120718697361,
|
| 384 |
+
"grad_norm": 1.4206688134227474,
|
| 385 |
+
"learning_rate": 7.912006902348045e-05,
|
| 386 |
+
"loss": 0.5087,
|
| 387 |
+
"step": 54
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.4941044357102751,
|
| 391 |
+
"grad_norm": 0.9288407004391388,
|
| 392 |
+
"learning_rate": 7.903024223357797e-05,
|
| 393 |
+
"loss": 0.5059,
|
| 394 |
+
"step": 55
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.5030881527231892,
|
| 398 |
+
"grad_norm": 1.0360922211438188,
|
| 399 |
+
"learning_rate": 7.893610666070134e-05,
|
| 400 |
+
"loss": 0.5025,
|
| 401 |
+
"step": 56
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.5120718697361033,
|
| 405 |
+
"grad_norm": 1.5478822264915786,
|
| 406 |
+
"learning_rate": 7.883767269704209e-05,
|
| 407 |
+
"loss": 0.5166,
|
| 408 |
+
"step": 57
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.5210555867490174,
|
| 412 |
+
"grad_norm": 1.2814412272627784,
|
| 413 |
+
"learning_rate": 7.873495120931697e-05,
|
| 414 |
+
"loss": 0.5053,
|
| 415 |
+
"step": 58
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.5300393037619315,
|
| 419 |
+
"grad_norm": 1.2394259373221026,
|
| 420 |
+
"learning_rate": 7.86279535375683e-05,
|
| 421 |
+
"loss": 0.506,
|
| 422 |
+
"step": 59
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.5390230207748455,
|
| 426 |
+
"grad_norm": 0.7402164979257034,
|
| 427 |
+
"learning_rate": 7.851669149391198e-05,
|
| 428 |
+
"loss": 0.4972,
|
| 429 |
+
"step": 60
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.5480067377877597,
|
| 433 |
+
"grad_norm": 1.0415746091150728,
|
| 434 |
+
"learning_rate": 7.84011773612336e-05,
|
| 435 |
+
"loss": 0.503,
|
| 436 |
+
"step": 61
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.5569904548006738,
|
| 440 |
+
"grad_norm": 1.287772964853304,
|
| 441 |
+
"learning_rate": 7.828142389183239e-05,
|
| 442 |
+
"loss": 0.5002,
|
| 443 |
+
"step": 62
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.5659741718135879,
|
| 447 |
+
"grad_norm": 1.5812766703675953,
|
| 448 |
+
"learning_rate": 7.815744430601344e-05,
|
| 449 |
+
"loss": 0.5056,
|
| 450 |
+
"step": 63
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.574957888826502,
|
| 454 |
+
"grad_norm": 0.8792970340323266,
|
| 455 |
+
"learning_rate": 7.802925229062823e-05,
|
| 456 |
+
"loss": 0.4918,
|
| 457 |
+
"step": 64
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.583941605839416,
|
| 461 |
+
"grad_norm": 1.1398022409490918,
|
| 462 |
+
"learning_rate": 7.789686199756365e-05,
|
| 463 |
+
"loss": 0.4969,
|
| 464 |
+
"step": 65
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.5929253228523301,
|
| 468 |
+
"grad_norm": 1.7704394806497858,
|
| 469 |
+
"learning_rate": 7.776028804217968e-05,
|
| 470 |
+
"loss": 0.5026,
|
| 471 |
+
"step": 66
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.6019090398652442,
|
| 475 |
+
"grad_norm": 0.9434307642741575,
|
| 476 |
+
"learning_rate": 7.761954550169593e-05,
|
| 477 |
+
"loss": 0.4932,
|
| 478 |
+
"step": 67
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.6108927568781584,
|
| 482 |
+
"grad_norm": 2.082446516632287,
|
| 483 |
+
"learning_rate": 7.74746499135272e-05,
|
| 484 |
+
"loss": 0.5172,
|
| 485 |
+
"step": 68
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.6198764738910725,
|
| 489 |
+
"grad_norm": 1.8262658550358597,
|
| 490 |
+
"learning_rate": 7.732561727356811e-05,
|
| 491 |
+
"loss": 0.5039,
|
| 492 |
+
"step": 69
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 0.6288601909039865,
|
| 496 |
+
"grad_norm": 1.2365083047972374,
|
| 497 |
+
"learning_rate": 7.717246403442735e-05,
|
| 498 |
+
"loss": 0.4973,
|
| 499 |
+
"step": 70
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.6378439079169006,
|
| 503 |
+
"grad_norm": 1.4539572435843178,
|
| 504 |
+
"learning_rate": 7.701520710361129e-05,
|
| 505 |
+
"loss": 0.4988,
|
| 506 |
+
"step": 71
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.6468276249298147,
|
| 510 |
+
"grad_norm": 1.0089172545802498,
|
| 511 |
+
"learning_rate": 7.685386384165748e-05,
|
| 512 |
+
"loss": 0.4874,
|
| 513 |
+
"step": 72
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.6558113419427288,
|
| 517 |
+
"grad_norm": 1.3604668572334533,
|
| 518 |
+
"learning_rate": 7.668845206021812e-05,
|
| 519 |
+
"loss": 0.5035,
|
| 520 |
+
"step": 73
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.664795058955643,
|
| 524 |
+
"grad_norm": 1.0804166408590006,
|
| 525 |
+
"learning_rate": 7.651899002009375e-05,
|
| 526 |
+
"loss": 0.4926,
|
| 527 |
+
"step": 74
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.673778775968557,
|
| 531 |
+
"grad_norm": 1.4514655022285154,
|
| 532 |
+
"learning_rate": 7.634549642921725e-05,
|
| 533 |
+
"loss": 0.4939,
|
| 534 |
+
"step": 75
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.6827624929814711,
|
| 538 |
+
"grad_norm": 0.991441799340063,
|
| 539 |
+
"learning_rate": 7.616799044058867e-05,
|
| 540 |
+
"loss": 0.5067,
|
| 541 |
+
"step": 76
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.6917462099943852,
|
| 545 |
+
"grad_norm": 1.117806439503576,
|
| 546 |
+
"learning_rate": 7.598649165016073e-05,
|
| 547 |
+
"loss": 0.4915,
|
| 548 |
+
"step": 77
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.7007299270072993,
|
| 552 |
+
"grad_norm": 0.8700173580337157,
|
| 553 |
+
"learning_rate": 7.58010200946755e-05,
|
| 554 |
+
"loss": 0.4934,
|
| 555 |
+
"step": 78
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.7097136440202133,
|
| 559 |
+
"grad_norm": 0.9327016278300901,
|
| 560 |
+
"learning_rate": 7.561159624945257e-05,
|
| 561 |
+
"loss": 0.4823,
|
| 562 |
+
"step": 79
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 0.7186973610331274,
|
| 566 |
+
"grad_norm": 1.357805985616199,
|
| 567 |
+
"learning_rate": 7.541824102612839e-05,
|
| 568 |
+
"loss": 0.5005,
|
| 569 |
+
"step": 80
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.7276810780460415,
|
| 573 |
+
"grad_norm": 0.8155442555029125,
|
| 574 |
+
"learning_rate": 7.5220975770348e-05,
|
| 575 |
+
"loss": 0.4846,
|
| 576 |
+
"step": 81
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 0.7366647950589557,
|
| 580 |
+
"grad_norm": 0.8879306236067503,
|
| 581 |
+
"learning_rate": 7.501982225940833e-05,
|
| 582 |
+
"loss": 0.4716,
|
| 583 |
+
"step": 82
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 0.7456485120718698,
|
| 587 |
+
"grad_norm": 1.1674726644441007,
|
| 588 |
+
"learning_rate": 7.48148026998542e-05,
|
| 589 |
+
"loss": 0.4912,
|
| 590 |
+
"step": 83
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.7546322290847838,
|
| 594 |
+
"grad_norm": 0.9150265900061878,
|
| 595 |
+
"learning_rate": 7.460593972502674e-05,
|
| 596 |
+
"loss": 0.4857,
|
| 597 |
+
"step": 84
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 0.7636159460976979,
|
| 601 |
+
"grad_norm": 1.2818859952439805,
|
| 602 |
+
"learning_rate": 7.439325639256483e-05,
|
| 603 |
+
"loss": 0.4815,
|
| 604 |
+
"step": 85
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 0.772599663110612,
|
| 608 |
+
"grad_norm": 0.7413198899597869,
|
| 609 |
+
"learning_rate": 7.417677618185955e-05,
|
| 610 |
+
"loss": 0.4837,
|
| 611 |
+
"step": 86
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 0.7815833801235261,
|
| 615 |
+
"grad_norm": 0.9478143258160291,
|
| 616 |
+
"learning_rate": 7.39565229914622e-05,
|
| 617 |
+
"loss": 0.4765,
|
| 618 |
+
"step": 87
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 0.7905670971364402,
|
| 622 |
+
"grad_norm": 0.963846436149164,
|
| 623 |
+
"learning_rate": 7.373252113644596e-05,
|
| 624 |
+
"loss": 0.4879,
|
| 625 |
+
"step": 88
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.7995508141493542,
|
| 629 |
+
"grad_norm": 0.9081468443162725,
|
| 630 |
+
"learning_rate": 7.350479534572166e-05,
|
| 631 |
+
"loss": 0.4705,
|
| 632 |
+
"step": 89
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.8085345311622684,
|
| 636 |
+
"grad_norm": 0.6474133013047098,
|
| 637 |
+
"learning_rate": 7.327337075930775e-05,
|
| 638 |
+
"loss": 0.4806,
|
| 639 |
+
"step": 90
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.8175182481751825,
|
| 643 |
+
"grad_norm": 0.6809036448679276,
|
| 644 |
+
"learning_rate": 7.303827292555495e-05,
|
| 645 |
+
"loss": 0.481,
|
| 646 |
+
"step": 91
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 0.8265019651880966,
|
| 650 |
+
"grad_norm": 1.0720011034315253,
|
| 651 |
+
"learning_rate": 7.279952779832584e-05,
|
| 652 |
+
"loss": 0.4759,
|
| 653 |
+
"step": 92
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 0.8354856822010107,
|
| 657 |
+
"grad_norm": 1.0029995670688614,
|
| 658 |
+
"learning_rate": 7.255716173412966e-05,
|
| 659 |
+
"loss": 0.4736,
|
| 660 |
+
"step": 93
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 0.8444693992139247,
|
| 664 |
+
"grad_norm": 0.8353160428003878,
|
| 665 |
+
"learning_rate": 7.23112014892126e-05,
|
| 666 |
+
"loss": 0.4754,
|
| 667 |
+
"step": 94
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 0.8534531162268388,
|
| 671 |
+
"grad_norm": 0.9066101057502527,
|
| 672 |
+
"learning_rate": 7.20616742166041e-05,
|
| 673 |
+
"loss": 0.4801,
|
| 674 |
+
"step": 95
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 0.862436833239753,
|
| 678 |
+
"grad_norm": 1.3700422899595712,
|
| 679 |
+
"learning_rate": 7.180860746311917e-05,
|
| 680 |
+
"loss": 0.4904,
|
| 681 |
+
"step": 96
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 0.8714205502526671,
|
| 685 |
+
"grad_norm": 0.8940940289737819,
|
| 686 |
+
"learning_rate": 7.155202916631743e-05,
|
| 687 |
+
"loss": 0.4824,
|
| 688 |
+
"step": 97
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 0.8804042672655812,
|
| 692 |
+
"grad_norm": 1.3226382972793307,
|
| 693 |
+
"learning_rate": 7.129196765141886e-05,
|
| 694 |
+
"loss": 0.4679,
|
| 695 |
+
"step": 98
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 0.8893879842784952,
|
| 699 |
+
"grad_norm": 0.8696732708344425,
|
| 700 |
+
"learning_rate": 7.10284516281768e-05,
|
| 701 |
+
"loss": 0.4786,
|
| 702 |
+
"step": 99
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 0.8983717012914093,
|
| 706 |
+
"grad_norm": 1.0583061403429452,
|
| 707 |
+
"learning_rate": 7.076151018770854e-05,
|
| 708 |
+
"loss": 0.4669,
|
| 709 |
+
"step": 100
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 0.9073554183043234,
|
| 713 |
+
"grad_norm": 1.0722660670077375,
|
| 714 |
+
"learning_rate": 7.049117279928374e-05,
|
| 715 |
+
"loss": 0.4801,
|
| 716 |
+
"step": 101
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 0.9163391353172375,
|
| 720 |
+
"grad_norm": 1.2855639166116766,
|
| 721 |
+
"learning_rate": 7.021746930707117e-05,
|
| 722 |
+
"loss": 0.4674,
|
| 723 |
+
"step": 102
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 0.9253228523301515,
|
| 727 |
+
"grad_norm": 0.6812836053141724,
|
| 728 |
+
"learning_rate": 6.994042992684406e-05,
|
| 729 |
+
"loss": 0.4788,
|
| 730 |
+
"step": 103
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 0.9343065693430657,
|
| 734 |
+
"grad_norm": 0.6913667167229325,
|
| 735 |
+
"learning_rate": 6.966008524264429e-05,
|
| 736 |
+
"loss": 0.4679,
|
| 737 |
+
"step": 104
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 0.9432902863559798,
|
| 741 |
+
"grad_norm": 0.9914419206668401,
|
| 742 |
+
"learning_rate": 6.937646620340618e-05,
|
| 743 |
+
"loss": 0.4792,
|
| 744 |
+
"step": 105
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 0.9522740033688939,
|
| 748 |
+
"grad_norm": 1.3797231509055095,
|
| 749 |
+
"learning_rate": 6.908960411953973e-05,
|
| 750 |
+
"loss": 0.4683,
|
| 751 |
+
"step": 106
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 0.961257720381808,
|
| 755 |
+
"grad_norm": 0.5888339311760277,
|
| 756 |
+
"learning_rate": 6.879953065947416e-05,
|
| 757 |
+
"loss": 0.4678,
|
| 758 |
+
"step": 107
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 0.970241437394722,
|
| 762 |
+
"grad_norm": 1.038793566408693,
|
| 763 |
+
"learning_rate": 6.850627784616178e-05,
|
| 764 |
+
"loss": 0.4601,
|
| 765 |
+
"step": 108
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 0.9792251544076361,
|
| 769 |
+
"grad_norm": 1.0842893632221857,
|
| 770 |
+
"learning_rate": 6.82098780535428e-05,
|
| 771 |
+
"loss": 0.4681,
|
| 772 |
+
"step": 109
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 0.9882088714205502,
|
| 776 |
+
"grad_norm": 0.8626180542597122,
|
| 777 |
+
"learning_rate": 6.791036400297142e-05,
|
| 778 |
+
"loss": 0.4761,
|
| 779 |
+
"step": 110
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 0.9971925884334644,
|
| 783 |
+
"grad_norm": 0.7965083948145554,
|
| 784 |
+
"learning_rate": 6.760776875960347e-05,
|
| 785 |
+
"loss": 0.4689,
|
| 786 |
+
"step": 111
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 1.0075800112296462,
|
| 790 |
+
"grad_norm": 1.786141860362385,
|
| 791 |
+
"learning_rate": 6.730212572874618e-05,
|
| 792 |
+
"loss": 0.8334,
|
| 793 |
+
"step": 112
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 1.0165637282425604,
|
| 797 |
+
"grad_norm": 0.8882542778931486,
|
| 798 |
+
"learning_rate": 6.699346865217031e-05,
|
| 799 |
+
"loss": 0.4299,
|
| 800 |
+
"step": 113
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 1.0255474452554745,
|
| 804 |
+
"grad_norm": 0.6042242319747863,
|
| 805 |
+
"learning_rate": 6.668183160438531e-05,
|
| 806 |
+
"loss": 0.4431,
|
| 807 |
+
"step": 114
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 1.0345311622683886,
|
| 811 |
+
"grad_norm": 0.5382725444188637,
|
| 812 |
+
"learning_rate": 6.636724898887751e-05,
|
| 813 |
+
"loss": 0.4464,
|
| 814 |
+
"step": 115
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 1.0435148792813027,
|
| 818 |
+
"grad_norm": 0.6092419399947249,
|
| 819 |
+
"learning_rate": 6.604975553431219e-05,
|
| 820 |
+
"loss": 0.45,
|
| 821 |
+
"step": 116
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 1.0524985962942168,
|
| 825 |
+
"grad_norm": 0.952732534414775,
|
| 826 |
+
"learning_rate": 6.572938629069959e-05,
|
| 827 |
+
"loss": 0.4416,
|
| 828 |
+
"step": 117
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 1.0614823133071307,
|
| 832 |
+
"grad_norm": 1.2586357901517062,
|
| 833 |
+
"learning_rate": 6.540617662552565e-05,
|
| 834 |
+
"loss": 0.4478,
|
| 835 |
+
"step": 118
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 1.0704660303200448,
|
| 839 |
+
"grad_norm": 0.5882439478167614,
|
| 840 |
+
"learning_rate": 6.508016221984747e-05,
|
| 841 |
+
"loss": 0.4346,
|
| 842 |
+
"step": 119
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 1.079449747332959,
|
| 846 |
+
"grad_norm": 0.8126561068118403,
|
| 847 |
+
"learning_rate": 6.475137906435435e-05,
|
| 848 |
+
"loss": 0.4373,
|
| 849 |
+
"step": 120
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 1.088433464345873,
|
| 853 |
+
"grad_norm": 1.5656070861661977,
|
| 854 |
+
"learning_rate": 6.441986345539446e-05,
|
| 855 |
+
"loss": 0.4461,
|
| 856 |
+
"step": 121
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 1.0974171813587872,
|
| 860 |
+
"grad_norm": 0.5066258143638921,
|
| 861 |
+
"learning_rate": 6.408565199096798e-05,
|
| 862 |
+
"loss": 0.4327,
|
| 863 |
+
"step": 122
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 1.1064008983717013,
|
| 867 |
+
"grad_norm": 1.6682432153253044,
|
| 868 |
+
"learning_rate": 6.374878156668676e-05,
|
| 869 |
+
"loss": 0.4568,
|
| 870 |
+
"step": 123
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 1.1153846153846154,
|
| 874 |
+
"grad_norm": 0.6299972385025496,
|
| 875 |
+
"learning_rate": 6.340928937170118e-05,
|
| 876 |
+
"loss": 0.4285,
|
| 877 |
+
"step": 124
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 1.1243683323975295,
|
| 881 |
+
"grad_norm": 1.9783962130226307,
|
| 882 |
+
"learning_rate": 6.30672128845947e-05,
|
| 883 |
+
"loss": 0.4643,
|
| 884 |
+
"step": 125
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 1.1333520494104437,
|
| 888 |
+
"grad_norm": 1.2699186089072336,
|
| 889 |
+
"learning_rate": 6.272258986924624e-05,
|
| 890 |
+
"loss": 0.4362,
|
| 891 |
+
"step": 126
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 1.1423357664233578,
|
| 895 |
+
"grad_norm": 1.617840953040159,
|
| 896 |
+
"learning_rate": 6.237545837066133e-05,
|
| 897 |
+
"loss": 0.4476,
|
| 898 |
+
"step": 127
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 1.1513194834362717,
|
| 902 |
+
"grad_norm": 1.5459662606407818,
|
| 903 |
+
"learning_rate": 6.202585671077204e-05,
|
| 904 |
+
"loss": 0.4452,
|
| 905 |
+
"step": 128
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 1.1603032004491858,
|
| 909 |
+
"grad_norm": 0.8520351440042236,
|
| 910 |
+
"learning_rate": 6.167382348420637e-05,
|
| 911 |
+
"loss": 0.4485,
|
| 912 |
+
"step": 129
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 1.1692869174621,
|
| 916 |
+
"grad_norm": 1.2109454920551714,
|
| 917 |
+
"learning_rate": 6.131939755402755e-05,
|
| 918 |
+
"loss": 0.4436,
|
| 919 |
+
"step": 130
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 1.178270634475014,
|
| 923 |
+
"grad_norm": 0.7621427547756707,
|
| 924 |
+
"learning_rate": 6.09626180474438e-05,
|
| 925 |
+
"loss": 0.4436,
|
| 926 |
+
"step": 131
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 1.1872543514879281,
|
| 930 |
+
"grad_norm": 0.9377335412029888,
|
| 931 |
+
"learning_rate": 6.060352435148874e-05,
|
| 932 |
+
"loss": 0.4404,
|
| 933 |
+
"step": 132
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 1.1962380685008422,
|
| 937 |
+
"grad_norm": 0.742883271426497,
|
| 938 |
+
"learning_rate": 6.024215610867327e-05,
|
| 939 |
+
"loss": 0.4431,
|
| 940 |
+
"step": 133
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 1.2052217855137564,
|
| 944 |
+
"grad_norm": 0.623519085897906,
|
| 945 |
+
"learning_rate": 5.9878553212609184e-05,
|
| 946 |
+
"loss": 0.4349,
|
| 947 |
+
"step": 134
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 1.2142055025266705,
|
| 951 |
+
"grad_norm": 0.5738395916616672,
|
| 952 |
+
"learning_rate": 5.95127558036051e-05,
|
| 953 |
+
"loss": 0.4383,
|
| 954 |
+
"step": 135
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 1.2231892195395846,
|
| 958 |
+
"grad_norm": 0.5745090863281419,
|
| 959 |
+
"learning_rate": 5.9144804264235066e-05,
|
| 960 |
+
"loss": 0.4353,
|
| 961 |
+
"step": 136
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 1.2321729365524985,
|
| 965 |
+
"grad_norm": 0.6173172910448796,
|
| 966 |
+
"learning_rate": 5.8774739214880554e-05,
|
| 967 |
+
"loss": 0.4397,
|
| 968 |
+
"step": 137
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 1.2411566535654126,
|
| 972 |
+
"grad_norm": 0.44131140572281446,
|
| 973 |
+
"learning_rate": 5.840260150924609e-05,
|
| 974 |
+
"loss": 0.4358,
|
| 975 |
+
"step": 138
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 1.2501403705783267,
|
| 979 |
+
"grad_norm": 0.48929805430187906,
|
| 980 |
+
"learning_rate": 5.802843222984919e-05,
|
| 981 |
+
"loss": 0.4348,
|
| 982 |
+
"step": 139
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 1.2591240875912408,
|
| 986 |
+
"grad_norm": 0.46891906471051287,
|
| 987 |
+
"learning_rate": 5.765227268348501e-05,
|
| 988 |
+
"loss": 0.433,
|
| 989 |
+
"step": 140
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 1.268107804604155,
|
| 993 |
+
"grad_norm": 0.4197153657865162,
|
| 994 |
+
"learning_rate": 5.727416439666622e-05,
|
| 995 |
+
"loss": 0.4373,
|
| 996 |
+
"step": 141
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 1.277091521617069,
|
| 1000 |
+
"grad_norm": 0.49836695670940545,
|
| 1001 |
+
"learning_rate": 5.689414911103867e-05,
|
| 1002 |
+
"loss": 0.4339,
|
| 1003 |
+
"step": 142
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 1.2860752386299832,
|
| 1007 |
+
"grad_norm": 0.4282436546394893,
|
| 1008 |
+
"learning_rate": 5.651226877877326e-05,
|
| 1009 |
+
"loss": 0.4371,
|
| 1010 |
+
"step": 143
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 1.2950589556428973,
|
| 1014 |
+
"grad_norm": 0.39801460777108744,
|
| 1015 |
+
"learning_rate": 5.612856555793459e-05,
|
| 1016 |
+
"loss": 0.4422,
|
| 1017 |
+
"step": 144
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 1.3040426726558114,
|
| 1021 |
+
"grad_norm": 0.384138892088657,
|
| 1022 |
+
"learning_rate": 5.574308180782693e-05,
|
| 1023 |
+
"loss": 0.4235,
|
| 1024 |
+
"step": 145
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 1.3130263896687255,
|
| 1028 |
+
"grad_norm": 0.4344092290487113,
|
| 1029 |
+
"learning_rate": 5.5355860084317787e-05,
|
| 1030 |
+
"loss": 0.4356,
|
| 1031 |
+
"step": 146
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 1.3220101066816397,
|
| 1035 |
+
"grad_norm": 0.37928071414714964,
|
| 1036 |
+
"learning_rate": 5.496694313514009e-05,
|
| 1037 |
+
"loss": 0.4362,
|
| 1038 |
+
"step": 147
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 1.3309938236945535,
|
| 1042 |
+
"grad_norm": 0.34476337733896867,
|
| 1043 |
+
"learning_rate": 5.457637389517285e-05,
|
| 1044 |
+
"loss": 0.4398,
|
| 1045 |
+
"step": 148
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 1.3399775407074677,
|
| 1049 |
+
"grad_norm": 0.4506193928055396,
|
| 1050 |
+
"learning_rate": 5.4184195481701425e-05,
|
| 1051 |
+
"loss": 0.428,
|
| 1052 |
+
"step": 149
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 1.3489612577203818,
|
| 1056 |
+
"grad_norm": 0.32358078080164415,
|
| 1057 |
+
"learning_rate": 5.3790451189657486e-05,
|
| 1058 |
+
"loss": 0.4354,
|
| 1059 |
+
"step": 150
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 1.357944974733296,
|
| 1063 |
+
"grad_norm": 0.3714258728709878,
|
| 1064 |
+
"learning_rate": 5.339518448683945e-05,
|
| 1065 |
+
"loss": 0.4336,
|
| 1066 |
+
"step": 151
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 1.36692869174621,
|
| 1070 |
+
"grad_norm": 0.3698334511165891,
|
| 1071 |
+
"learning_rate": 5.2998439009113814e-05,
|
| 1072 |
+
"loss": 0.421,
|
| 1073 |
+
"step": 152
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 1.3759124087591241,
|
| 1077 |
+
"grad_norm": 0.3039344886888513,
|
| 1078 |
+
"learning_rate": 5.260025855559792e-05,
|
| 1079 |
+
"loss": 0.4265,
|
| 1080 |
+
"step": 153
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 1.3848961257720382,
|
| 1084 |
+
"grad_norm": 0.38707783731390033,
|
| 1085 |
+
"learning_rate": 5.2200687083824706e-05,
|
| 1086 |
+
"loss": 0.4285,
|
| 1087 |
+
"step": 154
|
| 1088 |
+
},
|
| 1089 |
+
{
|
| 1090 |
+
"epoch": 1.3938798427849521,
|
| 1091 |
+
"grad_norm": 0.3708434779281009,
|
| 1092 |
+
"learning_rate": 5.179976870488999e-05,
|
| 1093 |
+
"loss": 0.4242,
|
| 1094 |
+
"step": 155
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 1.4028635597978663,
|
| 1098 |
+
"grad_norm": 0.2827560899827463,
|
| 1099 |
+
"learning_rate": 5.1397547678582745e-05,
|
| 1100 |
+
"loss": 0.4215,
|
| 1101 |
+
"step": 156
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 1.4118472768107804,
|
| 1105 |
+
"grad_norm": 0.36357591101838377,
|
| 1106 |
+
"learning_rate": 5.099406840849902e-05,
|
| 1107 |
+
"loss": 0.4344,
|
| 1108 |
+
"step": 157
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"epoch": 1.4208309938236945,
|
| 1112 |
+
"grad_norm": 0.2815376075683387,
|
| 1113 |
+
"learning_rate": 5.058937543713999e-05,
|
| 1114 |
+
"loss": 0.4234,
|
| 1115 |
+
"step": 158
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 1.4298147108366086,
|
| 1119 |
+
"grad_norm": 0.34011781572335237,
|
| 1120 |
+
"learning_rate": 5.018351344099453e-05,
|
| 1121 |
+
"loss": 0.438,
|
| 1122 |
+
"step": 159
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 1.4387984278495227,
|
| 1126 |
+
"grad_norm": 0.32400444292287883,
|
| 1127 |
+
"learning_rate": 4.9776527225607274e-05,
|
| 1128 |
+
"loss": 0.4328,
|
| 1129 |
+
"step": 160
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 1.4477821448624368,
|
| 1133 |
+
"grad_norm": 0.2529768229883554,
|
| 1134 |
+
"learning_rate": 4.93684617206321e-05,
|
| 1135 |
+
"loss": 0.4339,
|
| 1136 |
+
"step": 161
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 1.456765861875351,
|
| 1140 |
+
"grad_norm": 0.2629261907748739,
|
| 1141 |
+
"learning_rate": 4.89593619748722e-05,
|
| 1142 |
+
"loss": 0.4394,
|
| 1143 |
+
"step": 162
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 1.465749578888265,
|
| 1147 |
+
"grad_norm": 0.32833332654685193,
|
| 1148 |
+
"learning_rate": 4.8549273151306795e-05,
|
| 1149 |
+
"loss": 0.4331,
|
| 1150 |
+
"step": 163
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 1.4747332959011792,
|
| 1154 |
+
"grad_norm": 0.37621666962913386,
|
| 1155 |
+
"learning_rate": 4.8138240522105365e-05,
|
| 1156 |
+
"loss": 0.4294,
|
| 1157 |
+
"step": 164
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 1.4837170129140933,
|
| 1161 |
+
"grad_norm": 0.33680087947524656,
|
| 1162 |
+
"learning_rate": 4.7726309463629733e-05,
|
| 1163 |
+
"loss": 0.4317,
|
| 1164 |
+
"step": 165
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 1.4927007299270074,
|
| 1168 |
+
"grad_norm": 0.22831905889281706,
|
| 1169 |
+
"learning_rate": 4.731352545142478e-05,
|
| 1170 |
+
"loss": 0.4265,
|
| 1171 |
+
"step": 166
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 1.5016844469399215,
|
| 1175 |
+
"grad_norm": 0.23855204275905859,
|
| 1176 |
+
"learning_rate": 4.689993405519802e-05,
|
| 1177 |
+
"loss": 0.4343,
|
| 1178 |
+
"step": 167
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 1.5106681639528357,
|
| 1182 |
+
"grad_norm": 0.3165956064567229,
|
| 1183 |
+
"learning_rate": 4.648558093378899e-05,
|
| 1184 |
+
"loss": 0.4331,
|
| 1185 |
+
"step": 168
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 1.5196518809657495,
|
| 1189 |
+
"grad_norm": 0.3792988696086109,
|
| 1190 |
+
"learning_rate": 4.607051183012862e-05,
|
| 1191 |
+
"loss": 0.4214,
|
| 1192 |
+
"step": 169
|
| 1193 |
+
},
|
| 1194 |
+
{
|
| 1195 |
+
"epoch": 1.5286355979786637,
|
| 1196 |
+
"grad_norm": 0.29743099805783,
|
| 1197 |
+
"learning_rate": 4.5654772566189415e-05,
|
| 1198 |
+
"loss": 0.4322,
|
| 1199 |
+
"step": 170
|
| 1200 |
+
},
|
| 1201 |
+
{
|
| 1202 |
+
"epoch": 1.5376193149915778,
|
| 1203 |
+
"grad_norm": 0.22315650518088825,
|
| 1204 |
+
"learning_rate": 4.5238409037926905e-05,
|
| 1205 |
+
"loss": 0.4286,
|
| 1206 |
+
"step": 171
|
| 1207 |
+
},
|
| 1208 |
+
{
|
| 1209 |
+
"epoch": 1.546603032004492,
|
| 1210 |
+
"grad_norm": 0.3745132482990998,
|
| 1211 |
+
"learning_rate": 4.4821467210212924e-05,
|
| 1212 |
+
"loss": 0.4325,
|
| 1213 |
+
"step": 172
|
| 1214 |
+
},
|
| 1215 |
+
{
|
| 1216 |
+
"epoch": 1.5555867490174058,
|
| 1217 |
+
"grad_norm": 0.4349435850403825,
|
| 1218 |
+
"learning_rate": 4.4403993111761265e-05,
|
| 1219 |
+
"loss": 0.4278,
|
| 1220 |
+
"step": 173
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 1.56457046603032,
|
| 1224 |
+
"grad_norm": 0.34417305619739286,
|
| 1225 |
+
"learning_rate": 4.398603283004626e-05,
|
| 1226 |
+
"loss": 0.421,
|
| 1227 |
+
"step": 174
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"epoch": 1.573554183043234,
|
| 1231 |
+
"grad_norm": 0.2697125284825644,
|
| 1232 |
+
"learning_rate": 4.356763250621496e-05,
|
| 1233 |
+
"loss": 0.4355,
|
| 1234 |
+
"step": 175
|
| 1235 |
+
},
|
| 1236 |
+
{
|
| 1237 |
+
"epoch": 1.5825379000561481,
|
| 1238 |
+
"grad_norm": 0.3759486979388303,
|
| 1239 |
+
"learning_rate": 4.314883832999326e-05,
|
| 1240 |
+
"loss": 0.4336,
|
| 1241 |
+
"step": 176
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 1.5915216170690623,
|
| 1245 |
+
"grad_norm": 0.42378540392492775,
|
| 1246 |
+
"learning_rate": 4.272969653458685e-05,
|
| 1247 |
+
"loss": 0.4258,
|
| 1248 |
+
"step": 177
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 1.6005053340819764,
|
| 1252 |
+
"grad_norm": 0.36295049407047797,
|
| 1253 |
+
"learning_rate": 4.231025339157714e-05,
|
| 1254 |
+
"loss": 0.4214,
|
| 1255 |
+
"step": 178
|
| 1256 |
+
},
|
| 1257 |
+
{
|
| 1258 |
+
"epoch": 1.6094890510948905,
|
| 1259 |
+
"grad_norm": 0.22472885668491016,
|
| 1260 |
+
"learning_rate": 4.189055520581315e-05,
|
| 1261 |
+
"loss": 0.4302,
|
| 1262 |
+
"step": 179
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 1.6184727681078046,
|
| 1266 |
+
"grad_norm": 0.24842552091044895,
|
| 1267 |
+
"learning_rate": 4.147064831029959e-05,
|
| 1268 |
+
"loss": 0.4267,
|
| 1269 |
+
"step": 180
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 1.6274564851207187,
|
| 1273 |
+
"grad_norm": 0.27444720145312784,
|
| 1274 |
+
"learning_rate": 4.105057906108189e-05,
|
| 1275 |
+
"loss": 0.4306,
|
| 1276 |
+
"step": 181
|
| 1277 |
+
},
|
| 1278 |
+
{
|
| 1279 |
+
"epoch": 1.6364402021336328,
|
| 1280 |
+
"grad_norm": 0.24170155717400021,
|
| 1281 |
+
"learning_rate": 4.063039383212866e-05,
|
| 1282 |
+
"loss": 0.4226,
|
| 1283 |
+
"step": 182
|
| 1284 |
+
},
|
| 1285 |
+
{
|
| 1286 |
+
"epoch": 1.645423919146547,
|
| 1287 |
+
"grad_norm": 0.2158635971279587,
|
| 1288 |
+
"learning_rate": 4.021013901021225e-05,
|
| 1289 |
+
"loss": 0.4182,
|
| 1290 |
+
"step": 183
|
| 1291 |
+
},
|
| 1292 |
+
{
|
| 1293 |
+
"epoch": 1.654407636159461,
|
| 1294 |
+
"grad_norm": 0.2392521978049934,
|
| 1295 |
+
"learning_rate": 3.978986098978777e-05,
|
| 1296 |
+
"loss": 0.4184,
|
| 1297 |
+
"step": 184
|
| 1298 |
+
},
|
| 1299 |
+
{
|
| 1300 |
+
"epoch": 1.6633913531723752,
|
| 1301 |
+
"grad_norm": 0.2333669024637109,
|
| 1302 |
+
"learning_rate": 3.936960616787135e-05,
|
| 1303 |
+
"loss": 0.4272,
|
| 1304 |
+
"step": 185
|
| 1305 |
+
},
|
| 1306 |
+
{
|
| 1307 |
+
"epoch": 1.6723750701852893,
|
| 1308 |
+
"grad_norm": 0.303971238157218,
|
| 1309 |
+
"learning_rate": 3.8949420938918124e-05,
|
| 1310 |
+
"loss": 0.4278,
|
| 1311 |
+
"step": 186
|
| 1312 |
+
},
|
| 1313 |
+
{
|
| 1314 |
+
"epoch": 1.6813587871982034,
|
| 1315 |
+
"grad_norm": 0.20260216895624977,
|
| 1316 |
+
"learning_rate": 3.852935168970042e-05,
|
| 1317 |
+
"loss": 0.4274,
|
| 1318 |
+
"step": 187
|
| 1319 |
+
},
|
| 1320 |
+
{
|
| 1321 |
+
"epoch": 1.6903425042111173,
|
| 1322 |
+
"grad_norm": 0.27940230833327445,
|
| 1323 |
+
"learning_rate": 3.810944479418686e-05,
|
| 1324 |
+
"loss": 0.4235,
|
| 1325 |
+
"step": 188
|
| 1326 |
+
},
|
| 1327 |
+
{
|
| 1328 |
+
"epoch": 1.6993262212240314,
|
| 1329 |
+
"grad_norm": 0.24151727930977668,
|
| 1330 |
+
"learning_rate": 3.768974660842287e-05,
|
| 1331 |
+
"loss": 0.43,
|
| 1332 |
+
"step": 189
|
| 1333 |
+
},
|
| 1334 |
+
{
|
| 1335 |
+
"epoch": 1.7083099382369455,
|
| 1336 |
+
"grad_norm": 0.28305027971796287,
|
| 1337 |
+
"learning_rate": 3.727030346541317e-05,
|
| 1338 |
+
"loss": 0.4204,
|
| 1339 |
+
"step": 190
|
| 1340 |
+
},
|
| 1341 |
+
{
|
| 1342 |
+
"epoch": 1.7172936552498597,
|
| 1343 |
+
"grad_norm": 0.23944450177743382,
|
| 1344 |
+
"learning_rate": 3.685116167000675e-05,
|
| 1345 |
+
"loss": 0.4303,
|
| 1346 |
+
"step": 191
|
| 1347 |
+
},
|
| 1348 |
+
{
|
| 1349 |
+
"epoch": 1.7262773722627736,
|
| 1350 |
+
"grad_norm": 0.2649036336219584,
|
| 1351 |
+
"learning_rate": 3.6432367493785056e-05,
|
| 1352 |
+
"loss": 0.4306,
|
| 1353 |
+
"step": 192
|
| 1354 |
+
},
|
| 1355 |
+
{
|
| 1356 |
+
"epoch": 1.7352610892756877,
|
| 1357 |
+
"grad_norm": 0.20536449694593853,
|
| 1358 |
+
"learning_rate": 3.601396716995375e-05,
|
| 1359 |
+
"loss": 0.4298,
|
| 1360 |
+
"step": 193
|
| 1361 |
+
},
|
| 1362 |
+
{
|
| 1363 |
+
"epoch": 1.7442448062886018,
|
| 1364 |
+
"grad_norm": 0.2752898250202777,
|
| 1365 |
+
"learning_rate": 3.559600688823875e-05,
|
| 1366 |
+
"loss": 0.4213,
|
| 1367 |
+
"step": 194
|
| 1368 |
+
},
|
| 1369 |
+
{
|
| 1370 |
+
"epoch": 1.753228523301516,
|
| 1371 |
+
"grad_norm": 0.2441828080744484,
|
| 1372 |
+
"learning_rate": 3.517853278978708e-05,
|
| 1373 |
+
"loss": 0.4317,
|
| 1374 |
+
"step": 195
|
| 1375 |
+
},
|
| 1376 |
+
{
|
| 1377 |
+
"epoch": 1.76221224031443,
|
| 1378 |
+
"grad_norm": 0.18921602583278035,
|
| 1379 |
+
"learning_rate": 3.4761590962073115e-05,
|
| 1380 |
+
"loss": 0.4352,
|
| 1381 |
+
"step": 196
|
| 1382 |
+
},
|
| 1383 |
+
{
|
| 1384 |
+
"epoch": 1.7711959573273441,
|
| 1385 |
+
"grad_norm": 0.25381845245915774,
|
| 1386 |
+
"learning_rate": 3.434522743381061e-05,
|
| 1387 |
+
"loss": 0.4227,
|
| 1388 |
+
"step": 197
|
| 1389 |
+
},
|
| 1390 |
+
{
|
| 1391 |
+
"epoch": 1.7801796743402583,
|
| 1392 |
+
"grad_norm": 0.22098025447635186,
|
| 1393 |
+
"learning_rate": 3.39294881698714e-05,
|
| 1394 |
+
"loss": 0.4145,
|
| 1395 |
+
"step": 198
|
| 1396 |
+
},
|
| 1397 |
+
{
|
| 1398 |
+
"epoch": 1.7891633913531724,
|
| 1399 |
+
"grad_norm": 0.21253762725360797,
|
| 1400 |
+
"learning_rate": 3.3514419066211025e-05,
|
| 1401 |
+
"loss": 0.4204,
|
| 1402 |
+
"step": 199
|
| 1403 |
+
},
|
| 1404 |
+
{
|
| 1405 |
+
"epoch": 1.7981471083660865,
|
| 1406 |
+
"grad_norm": 0.1993914725116837,
|
| 1407 |
+
"learning_rate": 3.310006594480199e-05,
|
| 1408 |
+
"loss": 0.4212,
|
| 1409 |
+
"step": 200
|
| 1410 |
+
},
|
| 1411 |
+
{
|
| 1412 |
+
"epoch": 1.8071308253790006,
|
| 1413 |
+
"grad_norm": 0.21949029430642605,
|
| 1414 |
+
"learning_rate": 3.268647454857524e-05,
|
| 1415 |
+
"loss": 0.4246,
|
| 1416 |
+
"step": 201
|
| 1417 |
+
},
|
| 1418 |
+
{
|
| 1419 |
+
"epoch": 1.8161145423919147,
|
| 1420 |
+
"grad_norm": 0.18093321704969498,
|
| 1421 |
+
"learning_rate": 3.227369053637028e-05,
|
| 1422 |
+
"loss": 0.4379,
|
| 1423 |
+
"step": 202
|
| 1424 |
+
},
|
| 1425 |
+
{
|
| 1426 |
+
"epoch": 1.8250982594048288,
|
| 1427 |
+
"grad_norm": 0.21469795266380035,
|
| 1428 |
+
"learning_rate": 3.1861759477894656e-05,
|
| 1429 |
+
"loss": 0.4227,
|
| 1430 |
+
"step": 203
|
| 1431 |
+
},
|
| 1432 |
+
{
|
| 1433 |
+
"epoch": 1.834081976417743,
|
| 1434 |
+
"grad_norm": 0.18433389487278787,
|
| 1435 |
+
"learning_rate": 3.145072684869322e-05,
|
| 1436 |
+
"loss": 0.4166,
|
| 1437 |
+
"step": 204
|
| 1438 |
+
},
|
| 1439 |
+
{
|
| 1440 |
+
"epoch": 1.843065693430657,
|
| 1441 |
+
"grad_norm": 0.19979893625833967,
|
| 1442 |
+
"learning_rate": 3.104063802512782e-05,
|
| 1443 |
+
"loss": 0.4281,
|
| 1444 |
+
"step": 205
|
| 1445 |
+
},
|
| 1446 |
+
{
|
| 1447 |
+
"epoch": 1.8520494104435712,
|
| 1448 |
+
"grad_norm": 0.19103290364400521,
|
| 1449 |
+
"learning_rate": 3.063153827936792e-05,
|
| 1450 |
+
"loss": 0.4161,
|
| 1451 |
+
"step": 206
|
| 1452 |
+
},
|
| 1453 |
+
{
|
| 1454 |
+
"epoch": 1.861033127456485,
|
| 1455 |
+
"grad_norm": 0.21106537662314218,
|
| 1456 |
+
"learning_rate": 3.0223472774392753e-05,
|
| 1457 |
+
"loss": 0.4242,
|
| 1458 |
+
"step": 207
|
| 1459 |
+
},
|
| 1460 |
+
{
|
| 1461 |
+
"epoch": 1.8700168444693992,
|
| 1462 |
+
"grad_norm": 0.20844328823177255,
|
| 1463 |
+
"learning_rate": 2.9816486559005482e-05,
|
| 1464 |
+
"loss": 0.4333,
|
| 1465 |
+
"step": 208
|
| 1466 |
+
},
|
| 1467 |
+
{
|
| 1468 |
+
"epoch": 1.8790005614823133,
|
| 1469 |
+
"grad_norm": 0.19732371134712767,
|
| 1470 |
+
"learning_rate": 2.9410624562860026e-05,
|
| 1471 |
+
"loss": 0.4211,
|
| 1472 |
+
"step": 209
|
| 1473 |
+
},
|
| 1474 |
+
{
|
| 1475 |
+
"epoch": 1.8879842784952274,
|
| 1476 |
+
"grad_norm": 0.23608540869194397,
|
| 1477 |
+
"learning_rate": 2.9005931591500974e-05,
|
| 1478 |
+
"loss": 0.422,
|
| 1479 |
+
"step": 210
|
| 1480 |
+
},
|
| 1481 |
+
{
|
| 1482 |
+
"epoch": 1.8969679955081415,
|
| 1483 |
+
"grad_norm": 0.18142970460558788,
|
| 1484 |
+
"learning_rate": 2.860245232141726e-05,
|
| 1485 |
+
"loss": 0.4215,
|
| 1486 |
+
"step": 211
|
| 1487 |
+
},
|
| 1488 |
+
{
|
| 1489 |
+
"epoch": 1.9059517125210554,
|
| 1490 |
+
"grad_norm": 0.21920791583501226,
|
| 1491 |
+
"learning_rate": 2.8200231295110012e-05,
|
| 1492 |
+
"loss": 0.4304,
|
| 1493 |
+
"step": 212
|
| 1494 |
+
},
|
| 1495 |
+
{
|
| 1496 |
+
"epoch": 1.9149354295339696,
|
| 1497 |
+
"grad_norm": 0.1575572213846906,
|
| 1498 |
+
"learning_rate": 2.7799312916175294e-05,
|
| 1499 |
+
"loss": 0.4115,
|
| 1500 |
+
"step": 213
|
| 1501 |
+
},
|
| 1502 |
+
{
|
| 1503 |
+
"epoch": 1.9239191465468837,
|
| 1504 |
+
"grad_norm": 0.21181976577415998,
|
| 1505 |
+
"learning_rate": 2.7399741444402087e-05,
|
| 1506 |
+
"loss": 0.4301,
|
| 1507 |
+
"step": 214
|
| 1508 |
+
},
|
| 1509 |
+
{
|
| 1510 |
+
"epoch": 1.9329028635597978,
|
| 1511 |
+
"grad_norm": 0.1817247821525619,
|
| 1512 |
+
"learning_rate": 2.7001560990886196e-05,
|
| 1513 |
+
"loss": 0.4202,
|
| 1514 |
+
"step": 215
|
| 1515 |
+
},
|
| 1516 |
+
{
|
| 1517 |
+
"epoch": 1.941886580572712,
|
| 1518 |
+
"grad_norm": 0.17883228911738394,
|
| 1519 |
+
"learning_rate": 2.6604815513160556e-05,
|
| 1520 |
+
"loss": 0.4241,
|
| 1521 |
+
"step": 216
|
| 1522 |
+
},
|
| 1523 |
+
{
|
| 1524 |
+
"epoch": 1.950870297585626,
|
| 1525 |
+
"grad_norm": 0.19947193861762386,
|
| 1526 |
+
"learning_rate": 2.6209548810342517e-05,
|
| 1527 |
+
"loss": 0.4137,
|
| 1528 |
+
"step": 217
|
| 1529 |
+
},
|
| 1530 |
+
{
|
| 1531 |
+
"epoch": 1.9598540145985401,
|
| 1532 |
+
"grad_norm": 0.19071364537376845,
|
| 1533 |
+
"learning_rate": 2.5815804518298575e-05,
|
| 1534 |
+
"loss": 0.4327,
|
| 1535 |
+
"step": 218
|
| 1536 |
+
},
|
| 1537 |
+
{
|
| 1538 |
+
"epoch": 1.9688377316114543,
|
| 1539 |
+
"grad_norm": 0.1712467330551581,
|
| 1540 |
+
"learning_rate": 2.542362610482715e-05,
|
| 1541 |
+
"loss": 0.4239,
|
| 1542 |
+
"step": 219
|
| 1543 |
+
},
|
| 1544 |
+
{
|
| 1545 |
+
"epoch": 1.9778214486243684,
|
| 1546 |
+
"grad_norm": 0.16971518326737378,
|
| 1547 |
+
"learning_rate": 2.503305686485991e-05,
|
| 1548 |
+
"loss": 0.4197,
|
| 1549 |
+
"step": 220
|
| 1550 |
+
},
|
| 1551 |
+
{
|
| 1552 |
+
"epoch": 1.9868051656372825,
|
| 1553 |
+
"grad_norm": 0.15655415639126638,
|
| 1554 |
+
"learning_rate": 2.464413991568222e-05,
|
| 1555 |
+
"loss": 0.4085,
|
| 1556 |
+
"step": 221
|
| 1557 |
+
},
|
| 1558 |
+
{
|
| 1559 |
+
"epoch": 1.9957888826501966,
|
| 1560 |
+
"grad_norm": 0.19809705687476845,
|
| 1561 |
+
"learning_rate": 2.4256918192173088e-05,
|
| 1562 |
+
"loss": 0.4325,
|
| 1563 |
+
"step": 222
|
| 1564 |
+
},
|
| 1565 |
+
{
|
| 1566 |
+
"epoch": 2.0061763054463784,
|
| 1567 |
+
"grad_norm": 0.32148951189801367,
|
| 1568 |
+
"learning_rate": 2.3871434442065414e-05,
|
| 1569 |
+
"loss": 0.7473,
|
| 1570 |
+
"step": 223
|
| 1571 |
+
},
|
| 1572 |
+
{
|
| 1573 |
+
"epoch": 2.0151600224592925,
|
| 1574 |
+
"grad_norm": 0.20734628721061774,
|
| 1575 |
+
"learning_rate": 2.3487731221226754e-05,
|
| 1576 |
+
"loss": 0.3817,
|
| 1577 |
+
"step": 224
|
| 1578 |
+
},
|
| 1579 |
+
{
|
| 1580 |
+
"epoch": 2.0241437394722066,
|
| 1581 |
+
"grad_norm": 0.2283335322281334,
|
| 1582 |
+
"learning_rate": 2.3105850888961348e-05,
|
| 1583 |
+
"loss": 0.4043,
|
| 1584 |
+
"step": 225
|
| 1585 |
+
},
|
| 1586 |
+
{
|
| 1587 |
+
"epoch": 2.0331274564851207,
|
| 1588 |
+
"grad_norm": 0.20409956046524397,
|
| 1589 |
+
"learning_rate": 2.272583560333379e-05,
|
| 1590 |
+
"loss": 0.3908,
|
| 1591 |
+
"step": 226
|
| 1592 |
+
},
|
| 1593 |
+
{
|
| 1594 |
+
"epoch": 2.042111173498035,
|
| 1595 |
+
"grad_norm": 0.21233165191280134,
|
| 1596 |
+
"learning_rate": 2.2347727316515e-05,
|
| 1597 |
+
"loss": 0.3963,
|
| 1598 |
+
"step": 227
|
| 1599 |
+
},
|
| 1600 |
+
{
|
| 1601 |
+
"epoch": 2.051094890510949,
|
| 1602 |
+
"grad_norm": 0.2058022752445305,
|
| 1603 |
+
"learning_rate": 2.1971567770150814e-05,
|
| 1604 |
+
"loss": 0.3875,
|
| 1605 |
+
"step": 228
|
| 1606 |
+
},
|
| 1607 |
+
{
|
| 1608 |
+
"epoch": 2.060078607523863,
|
| 1609 |
+
"grad_norm": 0.2202126073153398,
|
| 1610 |
+
"learning_rate": 2.1597398490753917e-05,
|
| 1611 |
+
"loss": 0.4031,
|
| 1612 |
+
"step": 229
|
| 1613 |
+
},
|
| 1614 |
+
{
|
| 1615 |
+
"epoch": 2.069062324536777,
|
| 1616 |
+
"grad_norm": 0.20478388239536333,
|
| 1617 |
+
"learning_rate": 2.1225260785119456e-05,
|
| 1618 |
+
"loss": 0.3891,
|
| 1619 |
+
"step": 230
|
| 1620 |
+
},
|
| 1621 |
+
{
|
| 1622 |
+
"epoch": 2.0780460415496913,
|
| 1623 |
+
"grad_norm": 0.1937488744583445,
|
| 1624 |
+
"learning_rate": 2.0855195735764947e-05,
|
| 1625 |
+
"loss": 0.3906,
|
| 1626 |
+
"step": 231
|
| 1627 |
+
},
|
| 1628 |
+
{
|
| 1629 |
+
"epoch": 2.0870297585626054,
|
| 1630 |
+
"grad_norm": 0.2163161528806818,
|
| 1631 |
+
"learning_rate": 2.0487244196394912e-05,
|
| 1632 |
+
"loss": 0.3941,
|
| 1633 |
+
"step": 232
|
| 1634 |
+
},
|
| 1635 |
+
{
|
| 1636 |
+
"epoch": 2.0960134755755195,
|
| 1637 |
+
"grad_norm": 0.18524457129950797,
|
| 1638 |
+
"learning_rate": 2.0121446787390822e-05,
|
| 1639 |
+
"loss": 0.3799,
|
| 1640 |
+
"step": 233
|
| 1641 |
+
},
|
| 1642 |
+
{
|
| 1643 |
+
"epoch": 2.1049971925884337,
|
| 1644 |
+
"grad_norm": 0.217114553144753,
|
| 1645 |
+
"learning_rate": 1.9757843891326736e-05,
|
| 1646 |
+
"loss": 0.3951,
|
| 1647 |
+
"step": 234
|
| 1648 |
+
},
|
| 1649 |
+
{
|
| 1650 |
+
"epoch": 2.1139809096013478,
|
| 1651 |
+
"grad_norm": 0.2017737198987517,
|
| 1652 |
+
"learning_rate": 1.939647564851127e-05,
|
| 1653 |
+
"loss": 0.3928,
|
| 1654 |
+
"step": 235
|
| 1655 |
+
},
|
| 1656 |
+
{
|
| 1657 |
+
"epoch": 2.1229646266142614,
|
| 1658 |
+
"grad_norm": 0.20060221109136825,
|
| 1659 |
+
"learning_rate": 1.9037381952556217e-05,
|
| 1660 |
+
"loss": 0.3891,
|
| 1661 |
+
"step": 236
|
| 1662 |
+
},
|
| 1663 |
+
{
|
| 1664 |
+
"epoch": 2.1319483436271756,
|
| 1665 |
+
"grad_norm": 0.19362260661168698,
|
| 1666 |
+
"learning_rate": 1.8680602445972463e-05,
|
| 1667 |
+
"loss": 0.3967,
|
| 1668 |
+
"step": 237
|
| 1669 |
+
},
|
| 1670 |
+
{
|
| 1671 |
+
"epoch": 2.1409320606400897,
|
| 1672 |
+
"grad_norm": 0.1898518071335132,
|
| 1673 |
+
"learning_rate": 1.832617651579365e-05,
|
| 1674 |
+
"loss": 0.3859,
|
| 1675 |
+
"step": 238
|
| 1676 |
+
},
|
| 1677 |
+
{
|
| 1678 |
+
"epoch": 2.149915777653004,
|
| 1679 |
+
"grad_norm": 0.1817097392847635,
|
| 1680 |
+
"learning_rate": 1.797414328922797e-05,
|
| 1681 |
+
"loss": 0.3878,
|
| 1682 |
+
"step": 239
|
| 1683 |
+
},
|
| 1684 |
+
{
|
| 1685 |
+
"epoch": 2.158899494665918,
|
| 1686 |
+
"grad_norm": 0.1438401587120875,
|
| 1687 |
+
"learning_rate": 1.7624541629338676e-05,
|
| 1688 |
+
"loss": 0.3902,
|
| 1689 |
+
"step": 240
|
| 1690 |
+
},
|
| 1691 |
+
{
|
| 1692 |
+
"epoch": 2.167883211678832,
|
| 1693 |
+
"grad_norm": 0.1785624789625333,
|
| 1694 |
+
"learning_rate": 1.7277410130753775e-05,
|
| 1695 |
+
"loss": 0.3803,
|
| 1696 |
+
"step": 241
|
| 1697 |
+
},
|
| 1698 |
+
{
|
| 1699 |
+
"epoch": 2.176866928691746,
|
| 1700 |
+
"grad_norm": 0.16410314706005405,
|
| 1701 |
+
"learning_rate": 1.6932787115405318e-05,
|
| 1702 |
+
"loss": 0.4038,
|
| 1703 |
+
"step": 242
|
| 1704 |
+
},
|
| 1705 |
+
{
|
| 1706 |
+
"epoch": 2.1858506457046603,
|
| 1707 |
+
"grad_norm": 0.17495931093245287,
|
| 1708 |
+
"learning_rate": 1.6590710628298826e-05,
|
| 1709 |
+
"loss": 0.4097,
|
| 1710 |
+
"step": 243
|
| 1711 |
+
},
|
| 1712 |
+
{
|
| 1713 |
+
"epoch": 2.1948343627175744,
|
| 1714 |
+
"grad_norm": 0.15732504490550855,
|
| 1715 |
+
"learning_rate": 1.6251218433313254e-05,
|
| 1716 |
+
"loss": 0.3841,
|
| 1717 |
+
"step": 244
|
| 1718 |
+
},
|
| 1719 |
+
{
|
| 1720 |
+
"epoch": 2.2038180797304885,
|
| 1721 |
+
"grad_norm": 0.15370170593613078,
|
| 1722 |
+
"learning_rate": 1.591434800903203e-05,
|
| 1723 |
+
"loss": 0.3879,
|
| 1724 |
+
"step": 245
|
| 1725 |
+
},
|
| 1726 |
+
{
|
| 1727 |
+
"epoch": 2.2128017967434026,
|
| 1728 |
+
"grad_norm": 0.14225993974716095,
|
| 1729 |
+
"learning_rate": 1.558013654460555e-05,
|
| 1730 |
+
"loss": 0.3831,
|
| 1731 |
+
"step": 246
|
| 1732 |
+
},
|
| 1733 |
+
{
|
| 1734 |
+
"epoch": 2.2217855137563167,
|
| 1735 |
+
"grad_norm": 0.16769227472528905,
|
| 1736 |
+
"learning_rate": 1.5248620935645666e-05,
|
| 1737 |
+
"loss": 0.3932,
|
| 1738 |
+
"step": 247
|
| 1739 |
+
},
|
| 1740 |
+
{
|
| 1741 |
+
"epoch": 2.230769230769231,
|
| 1742 |
+
"grad_norm": 0.15697609805604978,
|
| 1743 |
+
"learning_rate": 1.4919837780152544e-05,
|
| 1744 |
+
"loss": 0.3793,
|
| 1745 |
+
"step": 248
|
| 1746 |
+
},
|
| 1747 |
+
{
|
| 1748 |
+
"epoch": 2.239752947782145,
|
| 1749 |
+
"grad_norm": 0.14834997366132227,
|
| 1750 |
+
"learning_rate": 1.4593823374474374e-05,
|
| 1751 |
+
"loss": 0.3884,
|
| 1752 |
+
"step": 249
|
| 1753 |
+
},
|
| 1754 |
+
{
|
| 1755 |
+
"epoch": 2.248736664795059,
|
| 1756 |
+
"grad_norm": 0.1416591291545358,
|
| 1757 |
+
"learning_rate": 1.4270613709300429e-05,
|
| 1758 |
+
"loss": 0.3883,
|
| 1759 |
+
"step": 250
|
| 1760 |
+
},
|
| 1761 |
+
{
|
| 1762 |
+
"epoch": 2.257720381807973,
|
| 1763 |
+
"grad_norm": 0.1416137601943575,
|
| 1764 |
+
"learning_rate": 1.3950244465687833e-05,
|
| 1765 |
+
"loss": 0.3925,
|
| 1766 |
+
"step": 251
|
| 1767 |
+
},
|
| 1768 |
+
{
|
| 1769 |
+
"epoch": 2.2667040988208873,
|
| 1770 |
+
"grad_norm": 0.14486611430866134,
|
| 1771 |
+
"learning_rate": 1.3632751011122497e-05,
|
| 1772 |
+
"loss": 0.3988,
|
| 1773 |
+
"step": 252
|
| 1774 |
+
},
|
| 1775 |
+
{
|
| 1776 |
+
"epoch": 2.2756878158338014,
|
| 1777 |
+
"grad_norm": 0.1438931082489217,
|
| 1778 |
+
"learning_rate": 1.3318168395614697e-05,
|
| 1779 |
+
"loss": 0.3857,
|
| 1780 |
+
"step": 253
|
| 1781 |
+
},
|
| 1782 |
+
{
|
| 1783 |
+
"epoch": 2.2846715328467155,
|
| 1784 |
+
"grad_norm": 0.1376572002883416,
|
| 1785 |
+
"learning_rate": 1.3006531347829699e-05,
|
| 1786 |
+
"loss": 0.4054,
|
| 1787 |
+
"step": 254
|
| 1788 |
+
},
|
| 1789 |
+
{
|
| 1790 |
+
"epoch": 2.293655249859629,
|
| 1791 |
+
"grad_norm": 0.13374618220430773,
|
| 1792 |
+
"learning_rate": 1.2697874271253844e-05,
|
| 1793 |
+
"loss": 0.389,
|
| 1794 |
+
"step": 255
|
| 1795 |
+
},
|
| 1796 |
+
{
|
| 1797 |
+
"epoch": 2.3026389668725433,
|
| 1798 |
+
"grad_norm": 0.1344986927289295,
|
| 1799 |
+
"learning_rate": 1.2392231240396542e-05,
|
| 1800 |
+
"loss": 0.398,
|
| 1801 |
+
"step": 256
|
| 1802 |
+
},
|
| 1803 |
+
{
|
| 1804 |
+
"epoch": 2.3116226838854574,
|
| 1805 |
+
"grad_norm": 0.13248310416193929,
|
| 1806 |
+
"learning_rate": 1.2089635997028592e-05,
|
| 1807 |
+
"loss": 0.3964,
|
| 1808 |
+
"step": 257
|
| 1809 |
+
},
|
| 1810 |
+
{
|
| 1811 |
+
"epoch": 2.3206064008983716,
|
| 1812 |
+
"grad_norm": 0.1250202978575819,
|
| 1813 |
+
"learning_rate": 1.1790121946457212e-05,
|
| 1814 |
+
"loss": 0.3827,
|
| 1815 |
+
"step": 258
|
| 1816 |
+
},
|
| 1817 |
+
{
|
| 1818 |
+
"epoch": 2.3295901179112857,
|
| 1819 |
+
"grad_norm": 0.1267954036736396,
|
| 1820 |
+
"learning_rate": 1.1493722153838239e-05,
|
| 1821 |
+
"loss": 0.403,
|
| 1822 |
+
"step": 259
|
| 1823 |
+
},
|
| 1824 |
+
{
|
| 1825 |
+
"epoch": 2.3385738349242,
|
| 1826 |
+
"grad_norm": 0.11963916536373062,
|
| 1827 |
+
"learning_rate": 1.120046934052585e-05,
|
| 1828 |
+
"loss": 0.3774,
|
| 1829 |
+
"step": 260
|
| 1830 |
+
},
|
| 1831 |
+
{
|
| 1832 |
+
"epoch": 2.347557551937114,
|
| 1833 |
+
"grad_norm": 0.11989536513409248,
|
| 1834 |
+
"learning_rate": 1.0910395880460274e-05,
|
| 1835 |
+
"loss": 0.3994,
|
| 1836 |
+
"step": 261
|
| 1837 |
+
},
|
| 1838 |
+
{
|
| 1839 |
+
"epoch": 2.356541268950028,
|
| 1840 |
+
"grad_norm": 0.13086834428014207,
|
| 1841 |
+
"learning_rate": 1.062353379659383e-05,
|
| 1842 |
+
"loss": 0.3959,
|
| 1843 |
+
"step": 262
|
| 1844 |
+
},
|
| 1845 |
+
{
|
| 1846 |
+
"epoch": 2.365524985962942,
|
| 1847 |
+
"grad_norm": 0.12262091225654496,
|
| 1848 |
+
"learning_rate": 1.0339914757355718e-05,
|
| 1849 |
+
"loss": 0.3909,
|
| 1850 |
+
"step": 263
|
| 1851 |
+
},
|
| 1852 |
+
{
|
| 1853 |
+
"epoch": 2.3745087029758563,
|
| 1854 |
+
"grad_norm": 0.13332341565208733,
|
| 1855 |
+
"learning_rate": 1.0059570073155953e-05,
|
| 1856 |
+
"loss": 0.3964,
|
| 1857 |
+
"step": 264
|
| 1858 |
+
},
|
| 1859 |
+
{
|
| 1860 |
+
"epoch": 2.3834924199887704,
|
| 1861 |
+
"grad_norm": 0.1390720043482773,
|
| 1862 |
+
"learning_rate": 9.782530692928832e-06,
|
| 1863 |
+
"loss": 0.3962,
|
| 1864 |
+
"step": 265
|
| 1865 |
+
},
|
| 1866 |
+
{
|
| 1867 |
+
"epoch": 2.3924761370016845,
|
| 1868 |
+
"grad_norm": 0.12593592732440334,
|
| 1869 |
+
"learning_rate": 9.508827200716273e-06,
|
| 1870 |
+
"loss": 0.3987,
|
| 1871 |
+
"step": 266
|
| 1872 |
+
},
|
| 1873 |
+
{
|
| 1874 |
+
"epoch": 2.4014598540145986,
|
| 1875 |
+
"grad_norm": 0.14058113905698136,
|
| 1876 |
+
"learning_rate": 9.238489812291469e-06,
|
| 1877 |
+
"loss": 0.3851,
|
| 1878 |
+
"step": 267
|
| 1879 |
+
},
|
| 1880 |
+
{
|
| 1881 |
+
"epoch": 2.4104435710275127,
|
| 1882 |
+
"grad_norm": 0.1293391140497839,
|
| 1883 |
+
"learning_rate": 8.971548371823205e-06,
|
| 1884 |
+
"loss": 0.3972,
|
| 1885 |
+
"step": 268
|
| 1886 |
+
},
|
| 1887 |
+
{
|
| 1888 |
+
"epoch": 2.419427288040427,
|
| 1889 |
+
"grad_norm": 0.1312457950118106,
|
| 1890 |
+
"learning_rate": 8.708032348581144e-06,
|
| 1891 |
+
"loss": 0.401,
|
| 1892 |
+
"step": 269
|
| 1893 |
+
},
|
| 1894 |
+
{
|
| 1895 |
+
"epoch": 2.428411005053341,
|
| 1896 |
+
"grad_norm": 0.13054369787139117,
|
| 1897 |
+
"learning_rate": 8.447970833682584e-06,
|
| 1898 |
+
"loss": 0.3801,
|
| 1899 |
+
"step": 270
|
| 1900 |
+
},
|
| 1901 |
+
{
|
| 1902 |
+
"epoch": 2.437394722066255,
|
| 1903 |
+
"grad_norm": 0.12662575874177268,
|
| 1904 |
+
"learning_rate": 8.191392536880852e-06,
|
| 1905 |
+
"loss": 0.3835,
|
| 1906 |
+
"step": 271
|
| 1907 |
+
},
|
| 1908 |
+
{
|
| 1909 |
+
"epoch": 2.446378439079169,
|
| 1910 |
+
"grad_norm": 0.14035374781938695,
|
| 1911 |
+
"learning_rate": 7.938325783395924e-06,
|
| 1912 |
+
"loss": 0.3924,
|
| 1913 |
+
"step": 272
|
| 1914 |
+
},
|
| 1915 |
+
{
|
| 1916 |
+
"epoch": 2.4553621560920833,
|
| 1917 |
+
"grad_norm": 0.11900240575463085,
|
| 1918 |
+
"learning_rate": 7.68879851078741e-06,
|
| 1919 |
+
"loss": 0.3839,
|
| 1920 |
+
"step": 273
|
| 1921 |
+
},
|
| 1922 |
+
{
|
| 1923 |
+
"epoch": 2.464345873104997,
|
| 1924 |
+
"grad_norm": 0.13981444396533035,
|
| 1925 |
+
"learning_rate": 7.442838265870347e-06,
|
| 1926 |
+
"loss": 0.4029,
|
| 1927 |
+
"step": 274
|
| 1928 |
+
},
|
| 1929 |
+
{
|
| 1930 |
+
"epoch": 2.473329590117911,
|
| 1931 |
+
"grad_norm": 0.13223802646070204,
|
| 1932 |
+
"learning_rate": 7.2004722016741605e-06,
|
| 1933 |
+
"loss": 0.3933,
|
| 1934 |
+
"step": 275
|
| 1935 |
+
},
|
| 1936 |
+
{
|
| 1937 |
+
"epoch": 2.482313307130825,
|
| 1938 |
+
"grad_norm": 0.1225885147923774,
|
| 1939 |
+
"learning_rate": 6.961727074445055e-06,
|
| 1940 |
+
"loss": 0.3984,
|
| 1941 |
+
"step": 276
|
| 1942 |
+
},
|
| 1943 |
+
{
|
| 1944 |
+
"epoch": 2.4912970241437393,
|
| 1945 |
+
"grad_norm": 0.11467365860762357,
|
| 1946 |
+
"learning_rate": 6.726629240692255e-06,
|
| 1947 |
+
"loss": 0.378,
|
| 1948 |
+
"step": 277
|
| 1949 |
+
},
|
| 1950 |
+
{
|
| 1951 |
+
"epoch": 2.5002807411566534,
|
| 1952 |
+
"grad_norm": 0.11696089178213367,
|
| 1953 |
+
"learning_rate": 6.4952046542783395e-06,
|
| 1954 |
+
"loss": 0.3794,
|
| 1955 |
+
"step": 278
|
| 1956 |
+
},
|
| 1957 |
+
{
|
| 1958 |
+
"epoch": 2.5092644581695676,
|
| 1959 |
+
"grad_norm": 0.12498140576448633,
|
| 1960 |
+
"learning_rate": 6.2674788635540415e-06,
|
| 1961 |
+
"loss": 0.3905,
|
| 1962 |
+
"step": 279
|
| 1963 |
+
},
|
| 1964 |
+
{
|
| 1965 |
+
"epoch": 2.5182481751824817,
|
| 1966 |
+
"grad_norm": 0.11214127166395674,
|
| 1967 |
+
"learning_rate": 6.04347700853781e-06,
|
| 1968 |
+
"loss": 0.4005,
|
| 1969 |
+
"step": 280
|
| 1970 |
+
},
|
| 1971 |
+
{
|
| 1972 |
+
"epoch": 2.527231892195396,
|
| 1973 |
+
"grad_norm": 0.10515201118000413,
|
| 1974 |
+
"learning_rate": 5.823223818140458e-06,
|
| 1975 |
+
"loss": 0.3826,
|
| 1976 |
+
"step": 281
|
| 1977 |
+
},
|
| 1978 |
+
{
|
| 1979 |
+
"epoch": 2.53621560920831,
|
| 1980 |
+
"grad_norm": 0.10659878155108474,
|
| 1981 |
+
"learning_rate": 5.606743607435183e-06,
|
| 1982 |
+
"loss": 0.3862,
|
| 1983 |
+
"step": 282
|
| 1984 |
+
},
|
| 1985 |
+
{
|
| 1986 |
+
"epoch": 2.545199326221224,
|
| 1987 |
+
"grad_norm": 0.12414903554961848,
|
| 1988 |
+
"learning_rate": 5.394060274973267e-06,
|
| 1989 |
+
"loss": 0.4022,
|
| 1990 |
+
"step": 283
|
| 1991 |
+
},
|
| 1992 |
+
{
|
| 1993 |
+
"epoch": 2.554183043234138,
|
| 1994 |
+
"grad_norm": 0.1086137186577079,
|
| 1995 |
+
"learning_rate": 5.185197300145817e-06,
|
| 1996 |
+
"loss": 0.3772,
|
| 1997 |
+
"step": 284
|
| 1998 |
+
},
|
| 1999 |
+
{
|
| 2000 |
+
"epoch": 2.5631667602470523,
|
| 2001 |
+
"grad_norm": 0.11392438691496676,
|
| 2002 |
+
"learning_rate": 4.980177740591678e-06,
|
| 2003 |
+
"loss": 0.4074,
|
| 2004 |
+
"step": 285
|
| 2005 |
+
},
|
| 2006 |
+
{
|
| 2007 |
+
"epoch": 2.5721504772599664,
|
| 2008 |
+
"grad_norm": 0.10614011676956907,
|
| 2009 |
+
"learning_rate": 4.779024229652005e-06,
|
| 2010 |
+
"loss": 0.3946,
|
| 2011 |
+
"step": 286
|
| 2012 |
+
},
|
| 2013 |
+
{
|
| 2014 |
+
"epoch": 2.5811341942728805,
|
| 2015 |
+
"grad_norm": 0.10589648874830734,
|
| 2016 |
+
"learning_rate": 4.581758973871609e-06,
|
| 2017 |
+
"loss": 0.3948,
|
| 2018 |
+
"step": 287
|
| 2019 |
+
},
|
| 2020 |
+
{
|
| 2021 |
+
"epoch": 2.5901179112857946,
|
| 2022 |
+
"grad_norm": 0.10117190574073207,
|
| 2023 |
+
"learning_rate": 4.3884037505474455e-06,
|
| 2024 |
+
"loss": 0.3834,
|
| 2025 |
+
"step": 288
|
| 2026 |
+
},
|
| 2027 |
+
{
|
| 2028 |
+
"epoch": 2.5991016282987087,
|
| 2029 |
+
"grad_norm": 0.10088190358164044,
|
| 2030 |
+
"learning_rate": 4.198979905324496e-06,
|
| 2031 |
+
"loss": 0.3839,
|
| 2032 |
+
"step": 289
|
| 2033 |
+
},
|
| 2034 |
+
{
|
| 2035 |
+
"epoch": 2.608085345311623,
|
| 2036 |
+
"grad_norm": 0.10998415397637723,
|
| 2037 |
+
"learning_rate": 4.0135083498392905e-06,
|
| 2038 |
+
"loss": 0.4005,
|
| 2039 |
+
"step": 290
|
| 2040 |
+
},
|
| 2041 |
+
{
|
| 2042 |
+
"epoch": 2.6170690623245365,
|
| 2043 |
+
"grad_norm": 0.10291443940485356,
|
| 2044 |
+
"learning_rate": 3.832009559411338e-06,
|
| 2045 |
+
"loss": 0.3946,
|
| 2046 |
+
"step": 291
|
| 2047 |
+
},
|
| 2048 |
+
{
|
| 2049 |
+
"epoch": 2.626052779337451,
|
| 2050 |
+
"grad_norm": 0.0987497217572695,
|
| 2051 |
+
"learning_rate": 3.654503570782755e-06,
|
| 2052 |
+
"loss": 0.3797,
|
| 2053 |
+
"step": 292
|
| 2054 |
+
},
|
| 2055 |
+
{
|
| 2056 |
+
"epoch": 2.6350364963503647,
|
| 2057 |
+
"grad_norm": 0.10073485737297722,
|
| 2058 |
+
"learning_rate": 3.481009979906258e-06,
|
| 2059 |
+
"loss": 0.3977,
|
| 2060 |
+
"step": 293
|
| 2061 |
+
},
|
| 2062 |
+
{
|
| 2063 |
+
"epoch": 2.6440202133632793,
|
| 2064 |
+
"grad_norm": 0.10086816317959164,
|
| 2065 |
+
"learning_rate": 3.311547939781887e-06,
|
| 2066 |
+
"loss": 0.3944,
|
| 2067 |
+
"step": 294
|
| 2068 |
+
},
|
| 2069 |
+
{
|
| 2070 |
+
"epoch": 2.653003930376193,
|
| 2071 |
+
"grad_norm": 0.09803317553764752,
|
| 2072 |
+
"learning_rate": 3.14613615834253e-06,
|
| 2073 |
+
"loss": 0.3851,
|
| 2074 |
+
"step": 295
|
| 2075 |
+
},
|
| 2076 |
+
{
|
| 2077 |
+
"epoch": 2.661987647389107,
|
| 2078 |
+
"grad_norm": 0.10547156290892046,
|
| 2079 |
+
"learning_rate": 2.9847928963887198e-06,
|
| 2080 |
+
"loss": 0.3955,
|
| 2081 |
+
"step": 296
|
| 2082 |
+
},
|
| 2083 |
+
{
|
| 2084 |
+
"epoch": 2.670971364402021,
|
| 2085 |
+
"grad_norm": 0.09842701976978478,
|
| 2086 |
+
"learning_rate": 2.8275359655726586e-06,
|
| 2087 |
+
"loss": 0.3918,
|
| 2088 |
+
"step": 297
|
| 2089 |
+
},
|
| 2090 |
+
{
|
| 2091 |
+
"epoch": 2.6799550814149353,
|
| 2092 |
+
"grad_norm": 0.10279243698030634,
|
| 2093 |
+
"learning_rate": 2.6743827264319012e-06,
|
| 2094 |
+
"loss": 0.3916,
|
| 2095 |
+
"step": 298
|
| 2096 |
+
},
|
| 2097 |
+
{
|
| 2098 |
+
"epoch": 2.6889387984278494,
|
| 2099 |
+
"grad_norm": 0.09391104128159572,
|
| 2100 |
+
"learning_rate": 2.5253500864728155e-06,
|
| 2101 |
+
"loss": 0.3846,
|
| 2102 |
+
"step": 299
|
| 2103 |
+
},
|
| 2104 |
+
{
|
| 2105 |
+
"epoch": 2.6979225154407636,
|
| 2106 |
+
"grad_norm": 0.0914654057659311,
|
| 2107 |
+
"learning_rate": 2.3804544983040724e-06,
|
| 2108 |
+
"loss": 0.3824,
|
| 2109 |
+
"step": 300
|
| 2110 |
+
},
|
| 2111 |
+
{
|
| 2112 |
+
"epoch": 2.7069062324536777,
|
| 2113 |
+
"grad_norm": 0.09870686168033262,
|
| 2114 |
+
"learning_rate": 2.23971195782033e-06,
|
| 2115 |
+
"loss": 0.3927,
|
| 2116 |
+
"step": 301
|
| 2117 |
+
},
|
| 2118 |
+
{
|
| 2119 |
+
"epoch": 2.715889949466592,
|
| 2120 |
+
"grad_norm": 0.1061384997031586,
|
| 2121 |
+
"learning_rate": 2.1031380024363645e-06,
|
| 2122 |
+
"loss": 0.3993,
|
| 2123 |
+
"step": 302
|
| 2124 |
+
},
|
| 2125 |
+
{
|
| 2126 |
+
"epoch": 2.724873666479506,
|
| 2127 |
+
"grad_norm": 0.10190369293256683,
|
| 2128 |
+
"learning_rate": 1.9707477093717786e-06,
|
| 2129 |
+
"loss": 0.386,
|
| 2130 |
+
"step": 303
|
| 2131 |
+
},
|
| 2132 |
+
{
|
| 2133 |
+
"epoch": 2.73385738349242,
|
| 2134 |
+
"grad_norm": 0.09359682387036863,
|
| 2135 |
+
"learning_rate": 1.8425556939865696e-06,
|
| 2136 |
+
"loss": 0.3816,
|
| 2137 |
+
"step": 304
|
| 2138 |
+
},
|
| 2139 |
+
{
|
| 2140 |
+
"epoch": 2.742841100505334,
|
| 2141 |
+
"grad_norm": 0.0951294949828999,
|
| 2142 |
+
"learning_rate": 1.7185761081676222e-06,
|
| 2143 |
+
"loss": 0.3914,
|
| 2144 |
+
"step": 305
|
| 2145 |
+
},
|
| 2146 |
+
{
|
| 2147 |
+
"epoch": 2.7518248175182483,
|
| 2148 |
+
"grad_norm": 0.09488726435556502,
|
| 2149 |
+
"learning_rate": 1.5988226387664151e-06,
|
| 2150 |
+
"loss": 0.3826,
|
| 2151 |
+
"step": 306
|
| 2152 |
+
},
|
| 2153 |
+
{
|
| 2154 |
+
"epoch": 2.7608085345311624,
|
| 2155 |
+
"grad_norm": 0.08946031352013306,
|
| 2156 |
+
"learning_rate": 1.4833085060880349e-06,
|
| 2157 |
+
"loss": 0.3933,
|
| 2158 |
+
"step": 307
|
| 2159 |
+
},
|
| 2160 |
+
{
|
| 2161 |
+
"epoch": 2.7697922515440765,
|
| 2162 |
+
"grad_norm": 0.09936760457984616,
|
| 2163 |
+
"learning_rate": 1.3720464624317108e-06,
|
| 2164 |
+
"loss": 0.3754,
|
| 2165 |
+
"step": 308
|
| 2166 |
+
},
|
| 2167 |
+
{
|
| 2168 |
+
"epoch": 2.7787759685569906,
|
| 2169 |
+
"grad_norm": 0.09605033066883667,
|
| 2170 |
+
"learning_rate": 1.2650487906830234e-06,
|
| 2171 |
+
"loss": 0.4084,
|
| 2172 |
+
"step": 309
|
| 2173 |
+
},
|
| 2174 |
+
{
|
| 2175 |
+
"epoch": 2.7877596855699043,
|
| 2176 |
+
"grad_norm": 0.099913289552659,
|
| 2177 |
+
"learning_rate": 1.1623273029579195e-06,
|
| 2178 |
+
"loss": 0.3757,
|
| 2179 |
+
"step": 310
|
| 2180 |
+
},
|
| 2181 |
+
{
|
| 2182 |
+
"epoch": 2.796743402582819,
|
| 2183 |
+
"grad_norm": 0.09596634894133811,
|
| 2184 |
+
"learning_rate": 1.063893339298674e-06,
|
| 2185 |
+
"loss": 0.3957,
|
| 2186 |
+
"step": 311
|
| 2187 |
+
},
|
| 2188 |
+
{
|
| 2189 |
+
"epoch": 2.8057271195957325,
|
| 2190 |
+
"grad_norm": 0.09539713481428638,
|
| 2191 |
+
"learning_rate": 9.697577664220303e-07,
|
| 2192 |
+
"loss": 0.3943,
|
| 2193 |
+
"step": 312
|
| 2194 |
+
},
|
| 2195 |
+
{
|
| 2196 |
+
"epoch": 2.814710836608647,
|
| 2197 |
+
"grad_norm": 0.09238341661295082,
|
| 2198 |
+
"learning_rate": 8.799309765195452e-07,
|
| 2199 |
+
"loss": 0.3865,
|
| 2200 |
+
"step": 313
|
| 2201 |
+
},
|
| 2202 |
+
{
|
| 2203 |
+
"epoch": 2.8236945536215607,
|
| 2204 |
+
"grad_norm": 0.09215370347773162,
|
| 2205 |
+
"learning_rate": 7.944228861103264e-07,
|
| 2206 |
+
"loss": 0.393,
|
| 2207 |
+
"step": 314
|
| 2208 |
+
},
|
| 2209 |
+
{
|
| 2210 |
+
"epoch": 2.832678270634475,
|
| 2211 |
+
"grad_norm": 0.09454052595038595,
|
| 2212 |
+
"learning_rate": 7.132429349463011e-07,
|
| 2213 |
+
"loss": 0.387,
|
| 2214 |
+
"step": 315
|
| 2215 |
+
},
|
| 2216 |
+
{
|
| 2217 |
+
"epoch": 2.841661987647389,
|
| 2218 |
+
"grad_norm": 0.09250173345082788,
|
| 2219 |
+
"learning_rate": 6.364000849700791e-07,
|
| 2220 |
+
"loss": 0.391,
|
| 2221 |
+
"step": 316
|
| 2222 |
+
},
|
| 2223 |
+
{
|
| 2224 |
+
"epoch": 2.850645704660303,
|
| 2225 |
+
"grad_norm": 0.08839044291030243,
|
| 2226 |
+
"learning_rate": 5.639028193256257e-07,
|
| 2227 |
+
"loss": 0.3974,
|
| 2228 |
+
"step": 317
|
| 2229 |
+
},
|
| 2230 |
+
{
|
| 2231 |
+
"epoch": 2.859629421673217,
|
| 2232 |
+
"grad_norm": 0.08678885222396349,
|
| 2233 |
+
"learning_rate": 4.957591414217344e-07,
|
| 2234 |
+
"loss": 0.3912,
|
| 2235 |
+
"step": 318
|
| 2236 |
+
},
|
| 2237 |
+
{
|
| 2238 |
+
"epoch": 2.8686131386861313,
|
| 2239 |
+
"grad_norm": 0.09688413018501994,
|
| 2240 |
+
"learning_rate": 4.3197657404848935e-07,
|
| 2241 |
+
"loss": 0.3871,
|
| 2242 |
+
"step": 319
|
| 2243 |
+
},
|
| 2244 |
+
{
|
| 2245 |
+
"epoch": 2.8775968556990454,
|
| 2246 |
+
"grad_norm": 0.09442855901495942,
|
| 2247 |
+
"learning_rate": 3.725621585467698e-07,
|
| 2248 |
+
"loss": 0.3955,
|
| 2249 |
+
"step": 320
|
| 2250 |
+
},
|
| 2251 |
+
{
|
| 2252 |
+
"epoch": 2.8865805727119596,
|
| 2253 |
+
"grad_norm": 0.08776447719254343,
|
| 2254 |
+
"learning_rate": 3.1752245403092963e-07,
|
| 2255 |
+
"loss": 0.3889,
|
| 2256 |
+
"step": 321
|
| 2257 |
+
},
|
| 2258 |
+
{
|
| 2259 |
+
"epoch": 2.8955642897248737,
|
| 2260 |
+
"grad_norm": 0.08591645042050655,
|
| 2261 |
+
"learning_rate": 2.6686353666468323e-07,
|
| 2262 |
+
"loss": 0.3876,
|
| 2263 |
+
"step": 322
|
| 2264 |
+
},
|
| 2265 |
+
{
|
| 2266 |
+
"epoch": 2.904548006737788,
|
| 2267 |
+
"grad_norm": 0.08956097003735121,
|
| 2268 |
+
"learning_rate": 2.2059099899033098e-07,
|
| 2269 |
+
"loss": 0.3928,
|
| 2270 |
+
"step": 323
|
| 2271 |
+
},
|
| 2272 |
+
{
|
| 2273 |
+
"epoch": 2.913531723750702,
|
| 2274 |
+
"grad_norm": 0.08874803315016298,
|
| 2275 |
+
"learning_rate": 1.7870994931135977e-07,
|
| 2276 |
+
"loss": 0.3856,
|
| 2277 |
+
"step": 324
|
| 2278 |
+
},
|
| 2279 |
+
{
|
| 2280 |
+
"epoch": 2.922515440763616,
|
| 2281 |
+
"grad_norm": 0.09325131199811001,
|
| 2282 |
+
"learning_rate": 1.412250111285074e-07,
|
| 2283 |
+
"loss": 0.3976,
|
| 2284 |
+
"step": 325
|
| 2285 |
+
},
|
| 2286 |
+
{
|
| 2287 |
+
"epoch": 2.93149915777653,
|
| 2288 |
+
"grad_norm": 0.08572072856798726,
|
| 2289 |
+
"learning_rate": 1.0814032262935315e-07,
|
| 2290 |
+
"loss": 0.3809,
|
| 2291 |
+
"step": 326
|
| 2292 |
+
},
|
| 2293 |
+
{
|
| 2294 |
+
"epoch": 2.9404828747894443,
|
| 2295 |
+
"grad_norm": 0.08771028382043292,
|
| 2296 |
+
"learning_rate": 7.945953623146096e-08,
|
| 2297 |
+
"loss": 0.3791,
|
| 2298 |
+
"step": 327
|
| 2299 |
+
},
|
| 2300 |
+
{
|
| 2301 |
+
"epoch": 2.9494665918023584,
|
| 2302 |
+
"grad_norm": 0.08835820422720063,
|
| 2303 |
+
"learning_rate": 5.518581817918645e-08,
|
| 2304 |
+
"loss": 0.3928,
|
| 2305 |
+
"step": 328
|
| 2306 |
+
},
|
| 2307 |
+
{
|
| 2308 |
+
"epoch": 2.958450308815272,
|
| 2309 |
+
"grad_norm": 0.08971780481475433,
|
| 2310 |
+
"learning_rate": 3.532184819412532e-08,
|
| 2311 |
+
"loss": 0.3875,
|
| 2312 |
+
"step": 329
|
| 2313 |
+
},
|
| 2314 |
+
{
|
| 2315 |
+
"epoch": 2.9674340258281866,
|
| 2316 |
+
"grad_norm": 0.09086168349256409,
|
| 2317 |
+
"learning_rate": 1.9869819179292315e-08,
|
| 2318 |
+
"loss": 0.3902,
|
| 2319 |
+
"step": 330
|
| 2320 |
+
},
|
| 2321 |
+
{
|
| 2322 |
+
"epoch": 2.9764177428411003,
|
| 2323 |
+
"grad_norm": 0.09110615713231646,
|
| 2324 |
+
"learning_rate": 8.83143697702149e-09,
|
| 2325 |
+
"loss": 0.3928,
|
| 2326 |
+
"step": 331
|
| 2327 |
+
},
|
| 2328 |
+
{
|
| 2329 |
+
"epoch": 2.985401459854015,
|
| 2330 |
+
"grad_norm": 0.08717148170316144,
|
| 2331 |
+
"learning_rate": 2.2079201806501916e-09,
|
| 2332 |
+
"loss": 0.3867,
|
| 2333 |
+
"step": 332
|
| 2334 |
+
},
|
| 2335 |
+
{
|
| 2336 |
+
"epoch": 2.9943851768669285,
|
| 2337 |
+
"grad_norm": 0.09194318654950019,
|
| 2338 |
+
"learning_rate": 0.0,
|
| 2339 |
+
"loss": 0.3902,
|
| 2340 |
+
"step": 333
|
| 2341 |
+
},
|
| 2342 |
+
{
|
| 2343 |
+
"epoch": 2.9943851768669285,
|
| 2344 |
+
"step": 333,
|
| 2345 |
+
"total_flos": 4607052149424128.0,
|
| 2346 |
+
"train_loss": 0.4566754535869793,
|
| 2347 |
+
"train_runtime": 115300.1106,
|
| 2348 |
+
"train_samples_per_second": 2.965,
|
| 2349 |
+
"train_steps_per_second": 0.003
|
| 2350 |
+
}
|
| 2351 |
+
],
|
| 2352 |
+
"logging_steps": 1.0,
|
| 2353 |
+
"max_steps": 333,
|
| 2354 |
+
"num_input_tokens_seen": 0,
|
| 2355 |
+
"num_train_epochs": 3,
|
| 2356 |
+
"save_steps": 500,
|
| 2357 |
+
"stateful_callbacks": {
|
| 2358 |
+
"TrainerControl": {
|
| 2359 |
+
"args": {
|
| 2360 |
+
"should_epoch_stop": false,
|
| 2361 |
+
"should_evaluate": false,
|
| 2362 |
+
"should_log": false,
|
| 2363 |
+
"should_save": true,
|
| 2364 |
+
"should_training_stop": true
|
| 2365 |
+
},
|
| 2366 |
+
"attributes": {}
|
| 2367 |
+
}
|
| 2368 |
+
},
|
| 2369 |
+
"total_flos": 4607052149424128.0,
|
| 2370 |
+
"train_batch_size": 1,
|
| 2371 |
+
"trial_name": null,
|
| 2372 |
+
"trial_params": null
|
| 2373 |
+
}
|
training_loss.png
ADDED
|