End of training
Browse files- README.md +2 -1
- all_results.json +8 -0
- train_results.json +8 -0
- trainer_state.json +2107 -0
- training_loss.png +0 -0
README.md
CHANGED
|
@@ -4,6 +4,7 @@ license: apache-2.0
|
|
| 4 |
base_model: Qwen/Qwen2.5-32B-Instruct
|
| 5 |
tags:
|
| 6 |
- llama-factory
|
|
|
|
| 7 |
- generated_from_trainer
|
| 8 |
model-index:
|
| 9 |
- name: openthoughts2_100k_32B
|
|
@@ -15,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 15 |
|
| 16 |
# openthoughts2_100k_32B
|
| 17 |
|
| 18 |
-
This model is a fine-tuned version of [Qwen/Qwen2.5-32B-Instruct](https://huggingface.co/Qwen/Qwen2.5-32B-Instruct) on
|
| 19 |
|
| 20 |
## Model description
|
| 21 |
|
|
|
|
| 4 |
base_model: Qwen/Qwen2.5-32B-Instruct
|
| 5 |
tags:
|
| 6 |
- llama-factory
|
| 7 |
+
- full
|
| 8 |
- generated_from_trainer
|
| 9 |
model-index:
|
| 10 |
- name: openthoughts2_100k_32B
|
|
|
|
| 16 |
|
| 17 |
# openthoughts2_100k_32B
|
| 18 |
|
| 19 |
+
This model is a fine-tuned version of [Qwen/Qwen2.5-32B-Instruct](https://huggingface.co/Qwen/Qwen2.5-32B-Instruct) on the mlfoundations-dev/openthoughts2_100k dataset.
|
| 20 |
|
| 21 |
## Model description
|
| 22 |
|
all_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 4.957983193277311,
|
| 3 |
+
"total_flos": 8.686638104246747e+18,
|
| 4 |
+
"train_loss": 0.3853245627071898,
|
| 5 |
+
"train_runtime": 55023.1167,
|
| 6 |
+
"train_samples_per_second": 2.766,
|
| 7 |
+
"train_steps_per_second": 0.005
|
| 8 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 4.957983193277311,
|
| 3 |
+
"total_flos": 8.686638104246747e+18,
|
| 4 |
+
"train_loss": 0.3853245627071898,
|
| 5 |
+
"train_runtime": 55023.1167,
|
| 6 |
+
"train_samples_per_second": 2.766,
|
| 7 |
+
"train_steps_per_second": 0.005
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,2107 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 4.957983193277311,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 295,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.01680672268907563,
|
| 13 |
+
"grad_norm": 2.691845053029933,
|
| 14 |
+
"learning_rate": 2.666666666666667e-06,
|
| 15 |
+
"loss": 0.7667,
|
| 16 |
+
"step": 1
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.03361344537815126,
|
| 20 |
+
"grad_norm": 2.68760562160016,
|
| 21 |
+
"learning_rate": 5.333333333333334e-06,
|
| 22 |
+
"loss": 0.7613,
|
| 23 |
+
"step": 2
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.05042016806722689,
|
| 27 |
+
"grad_norm": 2.5059318657737664,
|
| 28 |
+
"learning_rate": 8.000000000000001e-06,
|
| 29 |
+
"loss": 0.7694,
|
| 30 |
+
"step": 3
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.06722689075630252,
|
| 34 |
+
"grad_norm": 1.4120973662715042,
|
| 35 |
+
"learning_rate": 1.0666666666666667e-05,
|
| 36 |
+
"loss": 0.7052,
|
| 37 |
+
"step": 4
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.08403361344537816,
|
| 41 |
+
"grad_norm": 1.1598849553761785,
|
| 42 |
+
"learning_rate": 1.3333333333333333e-05,
|
| 43 |
+
"loss": 0.6791,
|
| 44 |
+
"step": 5
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.10084033613445378,
|
| 48 |
+
"grad_norm": 1.0934198113532667,
|
| 49 |
+
"learning_rate": 1.6000000000000003e-05,
|
| 50 |
+
"loss": 0.6622,
|
| 51 |
+
"step": 6
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.11764705882352941,
|
| 55 |
+
"grad_norm": 1.341847205599009,
|
| 56 |
+
"learning_rate": 1.866666666666667e-05,
|
| 57 |
+
"loss": 0.6278,
|
| 58 |
+
"step": 7
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.13445378151260504,
|
| 62 |
+
"grad_norm": 1.0069713241648435,
|
| 63 |
+
"learning_rate": 2.1333333333333335e-05,
|
| 64 |
+
"loss": 0.5896,
|
| 65 |
+
"step": 8
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.15126050420168066,
|
| 69 |
+
"grad_norm": 1.407705426454466,
|
| 70 |
+
"learning_rate": 2.4e-05,
|
| 71 |
+
"loss": 0.593,
|
| 72 |
+
"step": 9
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.16806722689075632,
|
| 76 |
+
"grad_norm": 1.0425063442232185,
|
| 77 |
+
"learning_rate": 2.6666666666666667e-05,
|
| 78 |
+
"loss": 0.5643,
|
| 79 |
+
"step": 10
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.18487394957983194,
|
| 83 |
+
"grad_norm": 0.7384483820761029,
|
| 84 |
+
"learning_rate": 2.9333333333333333e-05,
|
| 85 |
+
"loss": 0.5691,
|
| 86 |
+
"step": 11
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.20168067226890757,
|
| 90 |
+
"grad_norm": 0.8833333481110107,
|
| 91 |
+
"learning_rate": 3.2000000000000005e-05,
|
| 92 |
+
"loss": 0.5488,
|
| 93 |
+
"step": 12
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.2184873949579832,
|
| 97 |
+
"grad_norm": 0.6334637256226143,
|
| 98 |
+
"learning_rate": 3.466666666666667e-05,
|
| 99 |
+
"loss": 0.528,
|
| 100 |
+
"step": 13
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.23529411764705882,
|
| 104 |
+
"grad_norm": 0.7111526801057929,
|
| 105 |
+
"learning_rate": 3.733333333333334e-05,
|
| 106 |
+
"loss": 0.5334,
|
| 107 |
+
"step": 14
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.25210084033613445,
|
| 111 |
+
"grad_norm": 0.6335250585966727,
|
| 112 |
+
"learning_rate": 4e-05,
|
| 113 |
+
"loss": 0.5191,
|
| 114 |
+
"step": 15
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.2689075630252101,
|
| 118 |
+
"grad_norm": 0.4176665648822708,
|
| 119 |
+
"learning_rate": 4.266666666666667e-05,
|
| 120 |
+
"loss": 0.5111,
|
| 121 |
+
"step": 16
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.2857142857142857,
|
| 125 |
+
"grad_norm": 0.48345873630902336,
|
| 126 |
+
"learning_rate": 4.5333333333333335e-05,
|
| 127 |
+
"loss": 0.5106,
|
| 128 |
+
"step": 17
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.3025210084033613,
|
| 132 |
+
"grad_norm": 0.4346895304692354,
|
| 133 |
+
"learning_rate": 4.8e-05,
|
| 134 |
+
"loss": 0.5062,
|
| 135 |
+
"step": 18
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.31932773109243695,
|
| 139 |
+
"grad_norm": 0.406087088494166,
|
| 140 |
+
"learning_rate": 5.066666666666667e-05,
|
| 141 |
+
"loss": 0.4986,
|
| 142 |
+
"step": 19
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.33613445378151263,
|
| 146 |
+
"grad_norm": 0.41630827905929374,
|
| 147 |
+
"learning_rate": 5.333333333333333e-05,
|
| 148 |
+
"loss": 0.503,
|
| 149 |
+
"step": 20
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.35294117647058826,
|
| 153 |
+
"grad_norm": 0.43530126632220195,
|
| 154 |
+
"learning_rate": 5.6e-05,
|
| 155 |
+
"loss": 0.4878,
|
| 156 |
+
"step": 21
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.3697478991596639,
|
| 160 |
+
"grad_norm": 0.38169963753892966,
|
| 161 |
+
"learning_rate": 5.8666666666666665e-05,
|
| 162 |
+
"loss": 0.4974,
|
| 163 |
+
"step": 22
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.3865546218487395,
|
| 167 |
+
"grad_norm": 0.36783843498888424,
|
| 168 |
+
"learning_rate": 6.133333333333334e-05,
|
| 169 |
+
"loss": 0.486,
|
| 170 |
+
"step": 23
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.40336134453781514,
|
| 174 |
+
"grad_norm": 0.4235245053309515,
|
| 175 |
+
"learning_rate": 6.400000000000001e-05,
|
| 176 |
+
"loss": 0.477,
|
| 177 |
+
"step": 24
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.42016806722689076,
|
| 181 |
+
"grad_norm": 0.46640033262366404,
|
| 182 |
+
"learning_rate": 6.666666666666667e-05,
|
| 183 |
+
"loss": 0.4777,
|
| 184 |
+
"step": 25
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.4369747899159664,
|
| 188 |
+
"grad_norm": 0.5472563079921927,
|
| 189 |
+
"learning_rate": 6.933333333333334e-05,
|
| 190 |
+
"loss": 0.4744,
|
| 191 |
+
"step": 26
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.453781512605042,
|
| 195 |
+
"grad_norm": 0.6687308726522384,
|
| 196 |
+
"learning_rate": 7.2e-05,
|
| 197 |
+
"loss": 0.4767,
|
| 198 |
+
"step": 27
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.47058823529411764,
|
| 202 |
+
"grad_norm": 0.7476270057305555,
|
| 203 |
+
"learning_rate": 7.466666666666667e-05,
|
| 204 |
+
"loss": 0.4753,
|
| 205 |
+
"step": 28
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.48739495798319327,
|
| 209 |
+
"grad_norm": 0.5741702259748745,
|
| 210 |
+
"learning_rate": 7.733333333333333e-05,
|
| 211 |
+
"loss": 0.4712,
|
| 212 |
+
"step": 29
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.5042016806722689,
|
| 216 |
+
"grad_norm": 0.40851930843562606,
|
| 217 |
+
"learning_rate": 8e-05,
|
| 218 |
+
"loss": 0.4577,
|
| 219 |
+
"step": 30
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.5210084033613446,
|
| 223 |
+
"grad_norm": 0.6154048464174249,
|
| 224 |
+
"learning_rate": 7.99971891808302e-05,
|
| 225 |
+
"loss": 0.4671,
|
| 226 |
+
"step": 31
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.5378151260504201,
|
| 230 |
+
"grad_norm": 0.5955513054069845,
|
| 231 |
+
"learning_rate": 7.998875711835601e-05,
|
| 232 |
+
"loss": 0.4584,
|
| 233 |
+
"step": 32
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.5546218487394958,
|
| 237 |
+
"grad_norm": 0.5794668150962397,
|
| 238 |
+
"learning_rate": 7.997470499762758e-05,
|
| 239 |
+
"loss": 0.4707,
|
| 240 |
+
"step": 33
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.5714285714285714,
|
| 244 |
+
"grad_norm": 0.725604120307356,
|
| 245 |
+
"learning_rate": 7.995503479354342e-05,
|
| 246 |
+
"loss": 0.4586,
|
| 247 |
+
"step": 34
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.5882352941176471,
|
| 251 |
+
"grad_norm": 0.8805000042332701,
|
| 252 |
+
"learning_rate": 7.992974927057287e-05,
|
| 253 |
+
"loss": 0.4586,
|
| 254 |
+
"step": 35
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.6050420168067226,
|
| 258 |
+
"grad_norm": 0.9301160541600448,
|
| 259 |
+
"learning_rate": 7.989885198236755e-05,
|
| 260 |
+
"loss": 0.461,
|
| 261 |
+
"step": 36
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.6218487394957983,
|
| 265 |
+
"grad_norm": 0.6687959491557314,
|
| 266 |
+
"learning_rate": 7.986234727126197e-05,
|
| 267 |
+
"loss": 0.4567,
|
| 268 |
+
"step": 37
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.6386554621848739,
|
| 272 |
+
"grad_norm": 0.48276052476406184,
|
| 273 |
+
"learning_rate": 7.982024026766325e-05,
|
| 274 |
+
"loss": 0.4509,
|
| 275 |
+
"step": 38
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.6554621848739496,
|
| 279 |
+
"grad_norm": 0.6994251235569624,
|
| 280 |
+
"learning_rate": 7.977253688932997e-05,
|
| 281 |
+
"loss": 0.4468,
|
| 282 |
+
"step": 39
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.6722689075630253,
|
| 286 |
+
"grad_norm": 0.4944245393194192,
|
| 287 |
+
"learning_rate": 7.971924384054068e-05,
|
| 288 |
+
"loss": 0.4375,
|
| 289 |
+
"step": 40
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.6890756302521008,
|
| 293 |
+
"grad_norm": 0.5667967297706823,
|
| 294 |
+
"learning_rate": 7.966036861115155e-05,
|
| 295 |
+
"loss": 0.4454,
|
| 296 |
+
"step": 41
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.7058823529411765,
|
| 300 |
+
"grad_norm": 0.6394864064046843,
|
| 301 |
+
"learning_rate": 7.959591947554373e-05,
|
| 302 |
+
"loss": 0.4527,
|
| 303 |
+
"step": 42
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.7226890756302521,
|
| 307 |
+
"grad_norm": 0.4788837795601491,
|
| 308 |
+
"learning_rate": 7.952590549146052e-05,
|
| 309 |
+
"loss": 0.4442,
|
| 310 |
+
"step": 43
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.7394957983193278,
|
| 314 |
+
"grad_norm": 0.535989134201356,
|
| 315 |
+
"learning_rate": 7.945033649873435e-05,
|
| 316 |
+
"loss": 0.4466,
|
| 317 |
+
"step": 44
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.7563025210084033,
|
| 321 |
+
"grad_norm": 0.5073709650588538,
|
| 322 |
+
"learning_rate": 7.936922311790388e-05,
|
| 323 |
+
"loss": 0.4399,
|
| 324 |
+
"step": 45
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.773109243697479,
|
| 328 |
+
"grad_norm": 0.4801591037445421,
|
| 329 |
+
"learning_rate": 7.928257674872142e-05,
|
| 330 |
+
"loss": 0.4397,
|
| 331 |
+
"step": 46
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.7899159663865546,
|
| 335 |
+
"grad_norm": 0.39577976012924415,
|
| 336 |
+
"learning_rate": 7.919040956855071e-05,
|
| 337 |
+
"loss": 0.4347,
|
| 338 |
+
"step": 47
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.8067226890756303,
|
| 342 |
+
"grad_norm": 0.425532007101897,
|
| 343 |
+
"learning_rate": 7.909273453065563e-05,
|
| 344 |
+
"loss": 0.4322,
|
| 345 |
+
"step": 48
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.8235294117647058,
|
| 349 |
+
"grad_norm": 0.4044140035245334,
|
| 350 |
+
"learning_rate": 7.89895653623796e-05,
|
| 351 |
+
"loss": 0.4359,
|
| 352 |
+
"step": 49
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.8403361344537815,
|
| 356 |
+
"grad_norm": 0.31786491157153846,
|
| 357 |
+
"learning_rate": 7.888091656321644e-05,
|
| 358 |
+
"loss": 0.438,
|
| 359 |
+
"step": 50
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.8571428571428571,
|
| 363 |
+
"grad_norm": 0.48289081143530643,
|
| 364 |
+
"learning_rate": 7.87668034027725e-05,
|
| 365 |
+
"loss": 0.4386,
|
| 366 |
+
"step": 51
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.8739495798319328,
|
| 370 |
+
"grad_norm": 0.28421440219587246,
|
| 371 |
+
"learning_rate": 7.864724191862074e-05,
|
| 372 |
+
"loss": 0.4387,
|
| 373 |
+
"step": 52
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.8907563025210085,
|
| 377 |
+
"grad_norm": 0.5138694221384514,
|
| 378 |
+
"learning_rate": 7.852224891404673e-05,
|
| 379 |
+
"loss": 0.4287,
|
| 380 |
+
"step": 53
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.907563025210084,
|
| 384 |
+
"grad_norm": 0.40831253443546706,
|
| 385 |
+
"learning_rate": 7.839184195568714e-05,
|
| 386 |
+
"loss": 0.433,
|
| 387 |
+
"step": 54
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.9243697478991597,
|
| 391 |
+
"grad_norm": 0.33975871394180684,
|
| 392 |
+
"learning_rate": 7.82560393710609e-05,
|
| 393 |
+
"loss": 0.4265,
|
| 394 |
+
"step": 55
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.9411764705882353,
|
| 398 |
+
"grad_norm": 0.4213615019709387,
|
| 399 |
+
"learning_rate": 7.811486024599342e-05,
|
| 400 |
+
"loss": 0.4305,
|
| 401 |
+
"step": 56
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.957983193277311,
|
| 405 |
+
"grad_norm": 0.4395502467895831,
|
| 406 |
+
"learning_rate": 7.796832442193425e-05,
|
| 407 |
+
"loss": 0.4189,
|
| 408 |
+
"step": 57
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.9747899159663865,
|
| 412 |
+
"grad_norm": 0.6743521862656765,
|
| 413 |
+
"learning_rate": 7.781645249316853e-05,
|
| 414 |
+
"loss": 0.4321,
|
| 415 |
+
"step": 58
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.9915966386554622,
|
| 419 |
+
"grad_norm": 0.5817014550697496,
|
| 420 |
+
"learning_rate": 7.765926580392275e-05,
|
| 421 |
+
"loss": 0.4279,
|
| 422 |
+
"step": 59
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 1.0084033613445378,
|
| 426 |
+
"grad_norm": 0.6307832215969053,
|
| 427 |
+
"learning_rate": 7.749678644536485e-05,
|
| 428 |
+
"loss": 0.635,
|
| 429 |
+
"step": 60
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 1.0252100840336134,
|
| 433 |
+
"grad_norm": 0.842944326776837,
|
| 434 |
+
"learning_rate": 7.73290372524996e-05,
|
| 435 |
+
"loss": 0.4154,
|
| 436 |
+
"step": 61
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 1.0420168067226891,
|
| 440 |
+
"grad_norm": 0.5917893972831788,
|
| 441 |
+
"learning_rate": 7.715604180095937e-05,
|
| 442 |
+
"loss": 0.4147,
|
| 443 |
+
"step": 62
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 1.0588235294117647,
|
| 447 |
+
"grad_norm": 0.6525263797558681,
|
| 448 |
+
"learning_rate": 7.697782440369076e-05,
|
| 449 |
+
"loss": 0.4113,
|
| 450 |
+
"step": 63
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 1.0756302521008403,
|
| 454 |
+
"grad_norm": 0.7505675324963623,
|
| 455 |
+
"learning_rate": 7.679441010753755e-05,
|
| 456 |
+
"loss": 0.411,
|
| 457 |
+
"step": 64
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 1.092436974789916,
|
| 461 |
+
"grad_norm": 0.6641125952248763,
|
| 462 |
+
"learning_rate": 7.660582468972074e-05,
|
| 463 |
+
"loss": 0.4058,
|
| 464 |
+
"step": 65
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 1.1092436974789917,
|
| 468 |
+
"grad_norm": 0.7288877221732883,
|
| 469 |
+
"learning_rate": 7.641209465421573e-05,
|
| 470 |
+
"loss": 0.4032,
|
| 471 |
+
"step": 66
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 1.1260504201680672,
|
| 475 |
+
"grad_norm": 0.4638673139752868,
|
| 476 |
+
"learning_rate": 7.621324722802738e-05,
|
| 477 |
+
"loss": 0.4086,
|
| 478 |
+
"step": 67
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 1.1428571428571428,
|
| 482 |
+
"grad_norm": 0.5860864612437695,
|
| 483 |
+
"learning_rate": 7.600931035736356e-05,
|
| 484 |
+
"loss": 0.3992,
|
| 485 |
+
"step": 68
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 1.1596638655462184,
|
| 489 |
+
"grad_norm": 0.43794259370767846,
|
| 490 |
+
"learning_rate": 7.580031270370755e-05,
|
| 491 |
+
"loss": 0.406,
|
| 492 |
+
"step": 69
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 1.1764705882352942,
|
| 496 |
+
"grad_norm": 0.5363944495188004,
|
| 497 |
+
"learning_rate": 7.558628363978991e-05,
|
| 498 |
+
"loss": 0.407,
|
| 499 |
+
"step": 70
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 1.1932773109243697,
|
| 503 |
+
"grad_norm": 0.403621325503068,
|
| 504 |
+
"learning_rate": 7.536725324546042e-05,
|
| 505 |
+
"loss": 0.4162,
|
| 506 |
+
"step": 71
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 1.2100840336134453,
|
| 510 |
+
"grad_norm": 0.5865691342672088,
|
| 511 |
+
"learning_rate": 7.514325230346066e-05,
|
| 512 |
+
"loss": 0.4025,
|
| 513 |
+
"step": 72
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 1.226890756302521,
|
| 517 |
+
"grad_norm": 0.32292970243716784,
|
| 518 |
+
"learning_rate": 7.491431229509771e-05,
|
| 519 |
+
"loss": 0.4065,
|
| 520 |
+
"step": 73
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 1.2436974789915967,
|
| 524 |
+
"grad_norm": 0.5101659400008781,
|
| 525 |
+
"learning_rate": 7.468046539581978e-05,
|
| 526 |
+
"loss": 0.4024,
|
| 527 |
+
"step": 74
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 1.2605042016806722,
|
| 531 |
+
"grad_norm": 0.3748012445079814,
|
| 532 |
+
"learning_rate": 7.444174447069423e-05,
|
| 533 |
+
"loss": 0.4119,
|
| 534 |
+
"step": 75
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 1.2773109243697478,
|
| 538 |
+
"grad_norm": 0.3653692632226436,
|
| 539 |
+
"learning_rate": 7.41981830697887e-05,
|
| 540 |
+
"loss": 0.4026,
|
| 541 |
+
"step": 76
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 1.2941176470588236,
|
| 545 |
+
"grad_norm": 0.33893355997710367,
|
| 546 |
+
"learning_rate": 7.394981542345592e-05,
|
| 547 |
+
"loss": 0.4015,
|
| 548 |
+
"step": 77
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 1.3109243697478992,
|
| 552 |
+
"grad_norm": 0.3416281786091713,
|
| 553 |
+
"learning_rate": 7.369667643752296e-05,
|
| 554 |
+
"loss": 0.3971,
|
| 555 |
+
"step": 78
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 1.3277310924369747,
|
| 559 |
+
"grad_norm": 0.4129645250532624,
|
| 560 |
+
"learning_rate": 7.343880168838554e-05,
|
| 561 |
+
"loss": 0.4068,
|
| 562 |
+
"step": 79
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 1.3445378151260505,
|
| 566 |
+
"grad_norm": 0.335965073499183,
|
| 567 |
+
"learning_rate": 7.317622741800808e-05,
|
| 568 |
+
"loss": 0.4059,
|
| 569 |
+
"step": 80
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 1.361344537815126,
|
| 573 |
+
"grad_norm": 0.4235285007706537,
|
| 574 |
+
"learning_rate": 7.29089905288302e-05,
|
| 575 |
+
"loss": 0.3947,
|
| 576 |
+
"step": 81
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 1.3781512605042017,
|
| 580 |
+
"grad_norm": 0.3999182510004518,
|
| 581 |
+
"learning_rate": 7.263712857858044e-05,
|
| 582 |
+
"loss": 0.4051,
|
| 583 |
+
"step": 82
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 1.3949579831932772,
|
| 587 |
+
"grad_norm": 0.5060305893035955,
|
| 588 |
+
"learning_rate": 7.236067977499791e-05,
|
| 589 |
+
"loss": 0.3898,
|
| 590 |
+
"step": 83
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 1.4117647058823528,
|
| 594 |
+
"grad_norm": 0.398862809618769,
|
| 595 |
+
"learning_rate": 7.207968297046238e-05,
|
| 596 |
+
"loss": 0.4019,
|
| 597 |
+
"step": 84
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 1.4285714285714286,
|
| 601 |
+
"grad_norm": 0.43100792196157717,
|
| 602 |
+
"learning_rate": 7.179417765653413e-05,
|
| 603 |
+
"loss": 0.4059,
|
| 604 |
+
"step": 85
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 1.4453781512605042,
|
| 608 |
+
"grad_norm": 0.39592305732412314,
|
| 609 |
+
"learning_rate": 7.150420395840363e-05,
|
| 610 |
+
"loss": 0.394,
|
| 611 |
+
"step": 86
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 1.46218487394958,
|
| 615 |
+
"grad_norm": 0.3442118106877621,
|
| 616 |
+
"learning_rate": 7.120980262925234e-05,
|
| 617 |
+
"loss": 0.395,
|
| 618 |
+
"step": 87
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 1.4789915966386555,
|
| 622 |
+
"grad_norm": 0.3206211870301796,
|
| 623 |
+
"learning_rate": 7.091101504452525e-05,
|
| 624 |
+
"loss": 0.4003,
|
| 625 |
+
"step": 88
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 1.495798319327731,
|
| 629 |
+
"grad_norm": 0.383410424859833,
|
| 630 |
+
"learning_rate": 7.06078831961159e-05,
|
| 631 |
+
"loss": 0.3912,
|
| 632 |
+
"step": 89
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 1.5126050420168067,
|
| 636 |
+
"grad_norm": 0.44849560625543744,
|
| 637 |
+
"learning_rate": 7.030044968646481e-05,
|
| 638 |
+
"loss": 0.3985,
|
| 639 |
+
"step": 90
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 1.5294117647058822,
|
| 643 |
+
"grad_norm": 0.38400073581396393,
|
| 644 |
+
"learning_rate": 6.99887577225721e-05,
|
| 645 |
+
"loss": 0.395,
|
| 646 |
+
"step": 91
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 1.5462184873949578,
|
| 650 |
+
"grad_norm": 0.24198297755580098,
|
| 651 |
+
"learning_rate": 6.967285110992512e-05,
|
| 652 |
+
"loss": 0.4004,
|
| 653 |
+
"step": 92
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 1.5630252100840336,
|
| 657 |
+
"grad_norm": 0.29207821438059633,
|
| 658 |
+
"learning_rate": 6.935277424634203e-05,
|
| 659 |
+
"loss": 0.399,
|
| 660 |
+
"step": 93
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 1.5798319327731094,
|
| 664 |
+
"grad_norm": 0.35899575244109855,
|
| 665 |
+
"learning_rate": 6.902857211573202e-05,
|
| 666 |
+
"loss": 0.4002,
|
| 667 |
+
"step": 94
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 1.596638655462185,
|
| 671 |
+
"grad_norm": 0.32099355054216155,
|
| 672 |
+
"learning_rate": 6.870029028177324e-05,
|
| 673 |
+
"loss": 0.3965,
|
| 674 |
+
"step": 95
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 1.6134453781512605,
|
| 678 |
+
"grad_norm": 0.29012138581501923,
|
| 679 |
+
"learning_rate": 6.836797488150935e-05,
|
| 680 |
+
"loss": 0.3872,
|
| 681 |
+
"step": 96
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 1.6302521008403361,
|
| 685 |
+
"grad_norm": 0.2748296924769867,
|
| 686 |
+
"learning_rate": 6.803167261886515e-05,
|
| 687 |
+
"loss": 0.3962,
|
| 688 |
+
"step": 97
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 1.6470588235294117,
|
| 692 |
+
"grad_norm": 0.26822786043560753,
|
| 693 |
+
"learning_rate": 6.769143075808304e-05,
|
| 694 |
+
"loss": 0.4003,
|
| 695 |
+
"step": 98
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 1.6638655462184873,
|
| 699 |
+
"grad_norm": 0.17852130806611444,
|
| 700 |
+
"learning_rate": 6.734729711708021e-05,
|
| 701 |
+
"loss": 0.3916,
|
| 702 |
+
"step": 99
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 1.680672268907563,
|
| 706 |
+
"grad_norm": 0.28162955924515753,
|
| 707 |
+
"learning_rate": 6.699932006072842e-05,
|
| 708 |
+
"loss": 0.3898,
|
| 709 |
+
"step": 100
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 1.6974789915966386,
|
| 713 |
+
"grad_norm": 0.28961502105683556,
|
| 714 |
+
"learning_rate": 6.664754849405673e-05,
|
| 715 |
+
"loss": 0.3947,
|
| 716 |
+
"step": 101
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 1.7142857142857144,
|
| 720 |
+
"grad_norm": 0.25691484707796003,
|
| 721 |
+
"learning_rate": 6.629203185537827e-05,
|
| 722 |
+
"loss": 0.3926,
|
| 723 |
+
"step": 102
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 1.73109243697479,
|
| 727 |
+
"grad_norm": 0.23255282584695108,
|
| 728 |
+
"learning_rate": 6.59328201093422e-05,
|
| 729 |
+
"loss": 0.3944,
|
| 730 |
+
"step": 103
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 1.7478991596638656,
|
| 734 |
+
"grad_norm": 0.1976889085047533,
|
| 735 |
+
"learning_rate": 6.55699637399116e-05,
|
| 736 |
+
"loss": 0.3933,
|
| 737 |
+
"step": 104
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 1.7647058823529411,
|
| 741 |
+
"grad_norm": 0.2646202937784542,
|
| 742 |
+
"learning_rate": 6.520351374326846e-05,
|
| 743 |
+
"loss": 0.4022,
|
| 744 |
+
"step": 105
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 1.7815126050420167,
|
| 748 |
+
"grad_norm": 0.3171970494610042,
|
| 749 |
+
"learning_rate": 6.483352162064649e-05,
|
| 750 |
+
"loss": 0.4014,
|
| 751 |
+
"step": 106
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 1.7983193277310925,
|
| 755 |
+
"grad_norm": 0.37334159682738505,
|
| 756 |
+
"learning_rate": 6.446003937109329e-05,
|
| 757 |
+
"loss": 0.3911,
|
| 758 |
+
"step": 107
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 1.815126050420168,
|
| 762 |
+
"grad_norm": 0.3609029090894194,
|
| 763 |
+
"learning_rate": 6.408311948416215e-05,
|
| 764 |
+
"loss": 0.3934,
|
| 765 |
+
"step": 108
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 1.8319327731092439,
|
| 769 |
+
"grad_norm": 0.3181436318155162,
|
| 770 |
+
"learning_rate": 6.370281493253529e-05,
|
| 771 |
+
"loss": 0.4023,
|
| 772 |
+
"step": 109
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 1.8487394957983194,
|
| 776 |
+
"grad_norm": 0.27702770109669783,
|
| 777 |
+
"learning_rate": 6.331917916457889e-05,
|
| 778 |
+
"loss": 0.4024,
|
| 779 |
+
"step": 110
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 1.865546218487395,
|
| 783 |
+
"grad_norm": 0.25379167905152156,
|
| 784 |
+
"learning_rate": 6.29322660968315e-05,
|
| 785 |
+
"loss": 0.3922,
|
| 786 |
+
"step": 111
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 1.8823529411764706,
|
| 790 |
+
"grad_norm": 0.282020046055504,
|
| 791 |
+
"learning_rate": 6.25421301064265e-05,
|
| 792 |
+
"loss": 0.3935,
|
| 793 |
+
"step": 112
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 1.8991596638655461,
|
| 797 |
+
"grad_norm": 0.3309510391544018,
|
| 798 |
+
"learning_rate": 6.214882602344995e-05,
|
| 799 |
+
"loss": 0.3945,
|
| 800 |
+
"step": 113
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 1.9159663865546217,
|
| 804 |
+
"grad_norm": 0.2845751702835833,
|
| 805 |
+
"learning_rate": 6.175240912323461e-05,
|
| 806 |
+
"loss": 0.3881,
|
| 807 |
+
"step": 114
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 1.9327731092436975,
|
| 811 |
+
"grad_norm": 0.2118436549339309,
|
| 812 |
+
"learning_rate": 6.135293511859164e-05,
|
| 813 |
+
"loss": 0.3948,
|
| 814 |
+
"step": 115
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 1.949579831932773,
|
| 818 |
+
"grad_norm": 0.23008393569547847,
|
| 819 |
+
"learning_rate": 6.0950460151980515e-05,
|
| 820 |
+
"loss": 0.3881,
|
| 821 |
+
"step": 116
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 1.9663865546218489,
|
| 825 |
+
"grad_norm": 0.23553865314276004,
|
| 826 |
+
"learning_rate": 6.054504078761882e-05,
|
| 827 |
+
"loss": 0.3863,
|
| 828 |
+
"step": 117
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 1.9831932773109244,
|
| 832 |
+
"grad_norm": 0.22921488970577752,
|
| 833 |
+
"learning_rate": 6.013673400353261e-05,
|
| 834 |
+
"loss": 0.3917,
|
| 835 |
+
"step": 118
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 2.0,
|
| 839 |
+
"grad_norm": 0.2616084911144252,
|
| 840 |
+
"learning_rate": 5.972559718354869e-05,
|
| 841 |
+
"loss": 0.5859,
|
| 842 |
+
"step": 119
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 2.0168067226890756,
|
| 846 |
+
"grad_norm": 0.3076389532615379,
|
| 847 |
+
"learning_rate": 5.93116881092298e-05,
|
| 848 |
+
"loss": 0.3702,
|
| 849 |
+
"step": 120
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 2.033613445378151,
|
| 853 |
+
"grad_norm": 0.31907041291254,
|
| 854 |
+
"learning_rate": 5.889506495175398e-05,
|
| 855 |
+
"loss": 0.3681,
|
| 856 |
+
"step": 121
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 2.0504201680672267,
|
| 860 |
+
"grad_norm": 0.4791426552198182,
|
| 861 |
+
"learning_rate": 5.84757862637391e-05,
|
| 862 |
+
"loss": 0.3673,
|
| 863 |
+
"step": 122
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 2.0672268907563027,
|
| 867 |
+
"grad_norm": 0.5778056426757773,
|
| 868 |
+
"learning_rate": 5.805391097101386e-05,
|
| 869 |
+
"loss": 0.3661,
|
| 870 |
+
"step": 123
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 2.0840336134453783,
|
| 874 |
+
"grad_norm": 0.5507339006232655,
|
| 875 |
+
"learning_rate": 5.7629498364336256e-05,
|
| 876 |
+
"loss": 0.3663,
|
| 877 |
+
"step": 124
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 2.100840336134454,
|
| 881 |
+
"grad_norm": 0.6328726056353446,
|
| 882 |
+
"learning_rate": 5.720260809106083e-05,
|
| 883 |
+
"loss": 0.3729,
|
| 884 |
+
"step": 125
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 2.1176470588235294,
|
| 888 |
+
"grad_norm": 0.7429629034960262,
|
| 889 |
+
"learning_rate": 5.677330014675576e-05,
|
| 890 |
+
"loss": 0.3658,
|
| 891 |
+
"step": 126
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 2.134453781512605,
|
| 895 |
+
"grad_norm": 0.6678103909472105,
|
| 896 |
+
"learning_rate": 5.6341634866771e-05,
|
| 897 |
+
"loss": 0.367,
|
| 898 |
+
"step": 127
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 2.1512605042016806,
|
| 902 |
+
"grad_norm": 0.3388603239941812,
|
| 903 |
+
"learning_rate": 5.590767291775879e-05,
|
| 904 |
+
"loss": 0.3694,
|
| 905 |
+
"step": 128
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 2.168067226890756,
|
| 909 |
+
"grad_norm": 0.35517335938167693,
|
| 910 |
+
"learning_rate": 5.547147528914735e-05,
|
| 911 |
+
"loss": 0.365,
|
| 912 |
+
"step": 129
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 2.184873949579832,
|
| 916 |
+
"grad_norm": 0.49335712782993274,
|
| 917 |
+
"learning_rate": 5.503310328456953e-05,
|
| 918 |
+
"loss": 0.3675,
|
| 919 |
+
"step": 130
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 2.2016806722689077,
|
| 923 |
+
"grad_norm": 0.3332416734487142,
|
| 924 |
+
"learning_rate": 5.4592618513247016e-05,
|
| 925 |
+
"loss": 0.3689,
|
| 926 |
+
"step": 131
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 2.2184873949579833,
|
| 930 |
+
"grad_norm": 0.28185687496125533,
|
| 931 |
+
"learning_rate": 5.415008288133177e-05,
|
| 932 |
+
"loss": 0.3688,
|
| 933 |
+
"step": 132
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 2.235294117647059,
|
| 937 |
+
"grad_norm": 0.46259549178722464,
|
| 938 |
+
"learning_rate": 5.370555858320566e-05,
|
| 939 |
+
"loss": 0.3668,
|
| 940 |
+
"step": 133
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 2.2521008403361344,
|
| 944 |
+
"grad_norm": 0.238264701230805,
|
| 945 |
+
"learning_rate": 5.3259108092739626e-05,
|
| 946 |
+
"loss": 0.3749,
|
| 947 |
+
"step": 134
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 2.26890756302521,
|
| 951 |
+
"grad_norm": 0.3292616365501503,
|
| 952 |
+
"learning_rate": 5.2810794154513503e-05,
|
| 953 |
+
"loss": 0.3588,
|
| 954 |
+
"step": 135
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 2.2857142857142856,
|
| 958 |
+
"grad_norm": 0.3911314797881199,
|
| 959 |
+
"learning_rate": 5.23606797749979e-05,
|
| 960 |
+
"loss": 0.3629,
|
| 961 |
+
"step": 136
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 2.302521008403361,
|
| 965 |
+
"grad_norm": 0.19918751894156442,
|
| 966 |
+
"learning_rate": 5.1908828213699125e-05,
|
| 967 |
+
"loss": 0.3652,
|
| 968 |
+
"step": 137
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 2.3193277310924367,
|
| 972 |
+
"grad_norm": 0.35182385992516935,
|
| 973 |
+
"learning_rate": 5.1455302974268694e-05,
|
| 974 |
+
"loss": 0.3661,
|
| 975 |
+
"step": 138
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 2.3361344537815127,
|
| 979 |
+
"grad_norm": 0.272728906608522,
|
| 980 |
+
"learning_rate": 5.1000167795578484e-05,
|
| 981 |
+
"loss": 0.3624,
|
| 982 |
+
"step": 139
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 2.3529411764705883,
|
| 986 |
+
"grad_norm": 0.25068511541253613,
|
| 987 |
+
"learning_rate": 5.054348664276271e-05,
|
| 988 |
+
"loss": 0.3693,
|
| 989 |
+
"step": 140
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 2.369747899159664,
|
| 993 |
+
"grad_norm": 0.31733904110572775,
|
| 994 |
+
"learning_rate": 5.008532369822835e-05,
|
| 995 |
+
"loss": 0.3649,
|
| 996 |
+
"step": 141
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 2.3865546218487395,
|
| 1000 |
+
"grad_norm": 0.17972907825746567,
|
| 1001 |
+
"learning_rate": 4.962574335263474e-05,
|
| 1002 |
+
"loss": 0.3662,
|
| 1003 |
+
"step": 142
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 2.403361344537815,
|
| 1007 |
+
"grad_norm": 0.3020836624778438,
|
| 1008 |
+
"learning_rate": 4.916481019584417e-05,
|
| 1009 |
+
"loss": 0.3624,
|
| 1010 |
+
"step": 143
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 2.4201680672268906,
|
| 1014 |
+
"grad_norm": 0.2308850828209494,
|
| 1015 |
+
"learning_rate": 4.870258900784432e-05,
|
| 1016 |
+
"loss": 0.3552,
|
| 1017 |
+
"step": 144
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 2.4369747899159666,
|
| 1021 |
+
"grad_norm": 0.20193774141580698,
|
| 1022 |
+
"learning_rate": 4.8239144749643936e-05,
|
| 1023 |
+
"loss": 0.3564,
|
| 1024 |
+
"step": 145
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 2.453781512605042,
|
| 1028 |
+
"grad_norm": 0.250331662781215,
|
| 1029 |
+
"learning_rate": 4.777454255414331e-05,
|
| 1030 |
+
"loss": 0.3543,
|
| 1031 |
+
"step": 146
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 2.4705882352941178,
|
| 1035 |
+
"grad_norm": 0.17546824261254554,
|
| 1036 |
+
"learning_rate": 4.73088477169803e-05,
|
| 1037 |
+
"loss": 0.3574,
|
| 1038 |
+
"step": 147
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 2.4873949579831933,
|
| 1042 |
+
"grad_norm": 0.19673285966819015,
|
| 1043 |
+
"learning_rate": 4.684212568735369e-05,
|
| 1044 |
+
"loss": 0.3667,
|
| 1045 |
+
"step": 148
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 2.504201680672269,
|
| 1049 |
+
"grad_norm": 0.1842996207044742,
|
| 1050 |
+
"learning_rate": 4.637444205882487e-05,
|
| 1051 |
+
"loss": 0.3655,
|
| 1052 |
+
"step": 149
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 2.5210084033613445,
|
| 1056 |
+
"grad_norm": 0.18687884258666848,
|
| 1057 |
+
"learning_rate": 4.5905862560099255e-05,
|
| 1058 |
+
"loss": 0.3603,
|
| 1059 |
+
"step": 150
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 2.53781512605042,
|
| 1063 |
+
"grad_norm": 0.20685281920056947,
|
| 1064 |
+
"learning_rate": 4.5436453045788733e-05,
|
| 1065 |
+
"loss": 0.3566,
|
| 1066 |
+
"step": 151
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 2.5546218487394956,
|
| 1070 |
+
"grad_norm": 0.16022750696310714,
|
| 1071 |
+
"learning_rate": 4.496627948715638e-05,
|
| 1072 |
+
"loss": 0.364,
|
| 1073 |
+
"step": 152
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 2.571428571428571,
|
| 1077 |
+
"grad_norm": 0.1943611017642303,
|
| 1078 |
+
"learning_rate": 4.449540796284476e-05,
|
| 1079 |
+
"loss": 0.3674,
|
| 1080 |
+
"step": 153
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 2.588235294117647,
|
| 1084 |
+
"grad_norm": 0.1714929488039447,
|
| 1085 |
+
"learning_rate": 4.4023904649589237e-05,
|
| 1086 |
+
"loss": 0.357,
|
| 1087 |
+
"step": 154
|
| 1088 |
+
},
|
| 1089 |
+
{
|
| 1090 |
+
"epoch": 2.6050420168067228,
|
| 1091 |
+
"grad_norm": 0.15384773411105176,
|
| 1092 |
+
"learning_rate": 4.3551835812917395e-05,
|
| 1093 |
+
"loss": 0.3591,
|
| 1094 |
+
"step": 155
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 2.6218487394957983,
|
| 1098 |
+
"grad_norm": 0.19073166965123248,
|
| 1099 |
+
"learning_rate": 4.307926779783601e-05,
|
| 1100 |
+
"loss": 0.3591,
|
| 1101 |
+
"step": 156
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 2.638655462184874,
|
| 1105 |
+
"grad_norm": 0.1335703992060258,
|
| 1106 |
+
"learning_rate": 4.260626701950686e-05,
|
| 1107 |
+
"loss": 0.3602,
|
| 1108 |
+
"step": 157
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"epoch": 2.6554621848739495,
|
| 1112 |
+
"grad_norm": 0.17349520044174563,
|
| 1113 |
+
"learning_rate": 4.2132899953912705e-05,
|
| 1114 |
+
"loss": 0.3667,
|
| 1115 |
+
"step": 158
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 2.6722689075630255,
|
| 1119 |
+
"grad_norm": 0.1775150978811181,
|
| 1120 |
+
"learning_rate": 4.165923312851467e-05,
|
| 1121 |
+
"loss": 0.3589,
|
| 1122 |
+
"step": 159
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 2.689075630252101,
|
| 1126 |
+
"grad_norm": 0.14407278924806977,
|
| 1127 |
+
"learning_rate": 4.1185333112902394e-05,
|
| 1128 |
+
"loss": 0.357,
|
| 1129 |
+
"step": 160
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 2.7058823529411766,
|
| 1133 |
+
"grad_norm": 0.15231547266435208,
|
| 1134 |
+
"learning_rate": 4.07112665094383e-05,
|
| 1135 |
+
"loss": 0.36,
|
| 1136 |
+
"step": 161
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 2.722689075630252,
|
| 1140 |
+
"grad_norm": 0.14252315824535286,
|
| 1141 |
+
"learning_rate": 4.023709994389722e-05,
|
| 1142 |
+
"loss": 0.3573,
|
| 1143 |
+
"step": 162
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 2.7394957983193278,
|
| 1147 |
+
"grad_norm": 0.1378441265699476,
|
| 1148 |
+
"learning_rate": 3.9762900056102785e-05,
|
| 1149 |
+
"loss": 0.3572,
|
| 1150 |
+
"step": 163
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 2.7563025210084033,
|
| 1154 |
+
"grad_norm": 0.14347929675561036,
|
| 1155 |
+
"learning_rate": 3.9288733490561705e-05,
|
| 1156 |
+
"loss": 0.3608,
|
| 1157 |
+
"step": 164
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 2.773109243697479,
|
| 1161 |
+
"grad_norm": 0.14849331510833153,
|
| 1162 |
+
"learning_rate": 3.881466688709761e-05,
|
| 1163 |
+
"loss": 0.3591,
|
| 1164 |
+
"step": 165
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 2.7899159663865545,
|
| 1168 |
+
"grad_norm": 0.13906139555698402,
|
| 1169 |
+
"learning_rate": 3.834076687148534e-05,
|
| 1170 |
+
"loss": 0.3558,
|
| 1171 |
+
"step": 166
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 2.80672268907563,
|
| 1175 |
+
"grad_norm": 0.14260008374258937,
|
| 1176 |
+
"learning_rate": 3.78671000460873e-05,
|
| 1177 |
+
"loss": 0.3575,
|
| 1178 |
+
"step": 167
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 2.8235294117647056,
|
| 1182 |
+
"grad_norm": 0.12436252855299343,
|
| 1183 |
+
"learning_rate": 3.739373298049316e-05,
|
| 1184 |
+
"loss": 0.3586,
|
| 1185 |
+
"step": 168
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 2.8403361344537816,
|
| 1189 |
+
"grad_norm": 0.13020395224817283,
|
| 1190 |
+
"learning_rate": 3.6920732202164004e-05,
|
| 1191 |
+
"loss": 0.3676,
|
| 1192 |
+
"step": 169
|
| 1193 |
+
},
|
| 1194 |
+
{
|
| 1195 |
+
"epoch": 2.857142857142857,
|
| 1196 |
+
"grad_norm": 0.11254986991287481,
|
| 1197 |
+
"learning_rate": 3.644816418708261e-05,
|
| 1198 |
+
"loss": 0.3618,
|
| 1199 |
+
"step": 170
|
| 1200 |
+
},
|
| 1201 |
+
{
|
| 1202 |
+
"epoch": 2.8739495798319328,
|
| 1203 |
+
"grad_norm": 0.12963966045485173,
|
| 1204 |
+
"learning_rate": 3.597609535041078e-05,
|
| 1205 |
+
"loss": 0.3573,
|
| 1206 |
+
"step": 171
|
| 1207 |
+
},
|
| 1208 |
+
{
|
| 1209 |
+
"epoch": 2.8907563025210083,
|
| 1210 |
+
"grad_norm": 0.11011973634738455,
|
| 1211 |
+
"learning_rate": 3.550459203715525e-05,
|
| 1212 |
+
"loss": 0.3542,
|
| 1213 |
+
"step": 172
|
| 1214 |
+
},
|
| 1215 |
+
{
|
| 1216 |
+
"epoch": 2.907563025210084,
|
| 1217 |
+
"grad_norm": 0.11758458272262749,
|
| 1218 |
+
"learning_rate": 3.503372051284363e-05,
|
| 1219 |
+
"loss": 0.3636,
|
| 1220 |
+
"step": 173
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 2.92436974789916,
|
| 1224 |
+
"grad_norm": 0.11410514339329306,
|
| 1225 |
+
"learning_rate": 3.456354695421128e-05,
|
| 1226 |
+
"loss": 0.364,
|
| 1227 |
+
"step": 174
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"epoch": 2.9411764705882355,
|
| 1231 |
+
"grad_norm": 0.10012225550387552,
|
| 1232 |
+
"learning_rate": 3.409413743990076e-05,
|
| 1233 |
+
"loss": 0.3551,
|
| 1234 |
+
"step": 175
|
| 1235 |
+
},
|
| 1236 |
+
{
|
| 1237 |
+
"epoch": 2.957983193277311,
|
| 1238 |
+
"grad_norm": 0.1043683097492942,
|
| 1239 |
+
"learning_rate": 3.3625557941175145e-05,
|
| 1240 |
+
"loss": 0.3661,
|
| 1241 |
+
"step": 176
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 2.9747899159663866,
|
| 1245 |
+
"grad_norm": 0.09527314217220456,
|
| 1246 |
+
"learning_rate": 3.3157874312646324e-05,
|
| 1247 |
+
"loss": 0.3603,
|
| 1248 |
+
"step": 177
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 2.991596638655462,
|
| 1252 |
+
"grad_norm": 0.09160058059615397,
|
| 1253 |
+
"learning_rate": 3.2691152283019707e-05,
|
| 1254 |
+
"loss": 0.3669,
|
| 1255 |
+
"step": 178
|
| 1256 |
+
},
|
| 1257 |
+
{
|
| 1258 |
+
"epoch": 3.008403361344538,
|
| 1259 |
+
"grad_norm": 0.1673063143193419,
|
| 1260 |
+
"learning_rate": 3.2225457445856704e-05,
|
| 1261 |
+
"loss": 0.5331,
|
| 1262 |
+
"step": 179
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 3.0252100840336134,
|
| 1266 |
+
"grad_norm": 0.15269449288549453,
|
| 1267 |
+
"learning_rate": 3.176085525035607e-05,
|
| 1268 |
+
"loss": 0.3318,
|
| 1269 |
+
"step": 180
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 3.042016806722689,
|
| 1273 |
+
"grad_norm": 0.1424068532393887,
|
| 1274 |
+
"learning_rate": 3.1297410992155695e-05,
|
| 1275 |
+
"loss": 0.3372,
|
| 1276 |
+
"step": 181
|
| 1277 |
+
},
|
| 1278 |
+
{
|
| 1279 |
+
"epoch": 3.0588235294117645,
|
| 1280 |
+
"grad_norm": 0.16281662608901373,
|
| 1281 |
+
"learning_rate": 3.083518980415584e-05,
|
| 1282 |
+
"loss": 0.3382,
|
| 1283 |
+
"step": 182
|
| 1284 |
+
},
|
| 1285 |
+
{
|
| 1286 |
+
"epoch": 3.0756302521008405,
|
| 1287 |
+
"grad_norm": 0.14963568531065155,
|
| 1288 |
+
"learning_rate": 3.0374256647365273e-05,
|
| 1289 |
+
"loss": 0.3309,
|
| 1290 |
+
"step": 183
|
| 1291 |
+
},
|
| 1292 |
+
{
|
| 1293 |
+
"epoch": 3.092436974789916,
|
| 1294 |
+
"grad_norm": 0.15470662750592085,
|
| 1295 |
+
"learning_rate": 2.9914676301771667e-05,
|
| 1296 |
+
"loss": 0.3278,
|
| 1297 |
+
"step": 184
|
| 1298 |
+
},
|
| 1299 |
+
{
|
| 1300 |
+
"epoch": 3.1092436974789917,
|
| 1301 |
+
"grad_norm": 0.14918104899050125,
|
| 1302 |
+
"learning_rate": 2.9456513357237305e-05,
|
| 1303 |
+
"loss": 0.3346,
|
| 1304 |
+
"step": 185
|
| 1305 |
+
},
|
| 1306 |
+
{
|
| 1307 |
+
"epoch": 3.1260504201680672,
|
| 1308 |
+
"grad_norm": 0.11986806248988924,
|
| 1309 |
+
"learning_rate": 2.8999832204421533e-05,
|
| 1310 |
+
"loss": 0.3357,
|
| 1311 |
+
"step": 186
|
| 1312 |
+
},
|
| 1313 |
+
{
|
| 1314 |
+
"epoch": 3.142857142857143,
|
| 1315 |
+
"grad_norm": 0.1401413508252755,
|
| 1316 |
+
"learning_rate": 2.8544697025731312e-05,
|
| 1317 |
+
"loss": 0.3298,
|
| 1318 |
+
"step": 187
|
| 1319 |
+
},
|
| 1320 |
+
{
|
| 1321 |
+
"epoch": 3.1596638655462184,
|
| 1322 |
+
"grad_norm": 0.11688946361187755,
|
| 1323 |
+
"learning_rate": 2.809117178630089e-05,
|
| 1324 |
+
"loss": 0.3289,
|
| 1325 |
+
"step": 188
|
| 1326 |
+
},
|
| 1327 |
+
{
|
| 1328 |
+
"epoch": 3.176470588235294,
|
| 1329 |
+
"grad_norm": 0.11513857072611866,
|
| 1330 |
+
"learning_rate": 2.7639320225002108e-05,
|
| 1331 |
+
"loss": 0.3349,
|
| 1332 |
+
"step": 189
|
| 1333 |
+
},
|
| 1334 |
+
{
|
| 1335 |
+
"epoch": 3.19327731092437,
|
| 1336 |
+
"grad_norm": 0.13295332062587695,
|
| 1337 |
+
"learning_rate": 2.7189205845486503e-05,
|
| 1338 |
+
"loss": 0.3258,
|
| 1339 |
+
"step": 190
|
| 1340 |
+
},
|
| 1341 |
+
{
|
| 1342 |
+
"epoch": 3.2100840336134455,
|
| 1343 |
+
"grad_norm": 0.10654787208820445,
|
| 1344 |
+
"learning_rate": 2.674089190726039e-05,
|
| 1345 |
+
"loss": 0.3289,
|
| 1346 |
+
"step": 191
|
| 1347 |
+
},
|
| 1348 |
+
{
|
| 1349 |
+
"epoch": 3.226890756302521,
|
| 1350 |
+
"grad_norm": 0.11708698026472689,
|
| 1351 |
+
"learning_rate": 2.6294441416794348e-05,
|
| 1352 |
+
"loss": 0.3462,
|
| 1353 |
+
"step": 192
|
| 1354 |
+
},
|
| 1355 |
+
{
|
| 1356 |
+
"epoch": 3.2436974789915967,
|
| 1357 |
+
"grad_norm": 0.10661916872672404,
|
| 1358 |
+
"learning_rate": 2.584991711866825e-05,
|
| 1359 |
+
"loss": 0.3362,
|
| 1360 |
+
"step": 193
|
| 1361 |
+
},
|
| 1362 |
+
{
|
| 1363 |
+
"epoch": 3.2605042016806722,
|
| 1364 |
+
"grad_norm": 0.10963466274242117,
|
| 1365 |
+
"learning_rate": 2.5407381486753e-05,
|
| 1366 |
+
"loss": 0.3362,
|
| 1367 |
+
"step": 194
|
| 1368 |
+
},
|
| 1369 |
+
{
|
| 1370 |
+
"epoch": 3.277310924369748,
|
| 1371 |
+
"grad_norm": 0.10063035794234076,
|
| 1372 |
+
"learning_rate": 2.4966896715430484e-05,
|
| 1373 |
+
"loss": 0.327,
|
| 1374 |
+
"step": 195
|
| 1375 |
+
},
|
| 1376 |
+
{
|
| 1377 |
+
"epoch": 3.2941176470588234,
|
| 1378 |
+
"grad_norm": 0.08888027946992373,
|
| 1379 |
+
"learning_rate": 2.452852471085266e-05,
|
| 1380 |
+
"loss": 0.3298,
|
| 1381 |
+
"step": 196
|
| 1382 |
+
},
|
| 1383 |
+
{
|
| 1384 |
+
"epoch": 3.310924369747899,
|
| 1385 |
+
"grad_norm": 0.09286682252301849,
|
| 1386 |
+
"learning_rate": 2.4092327082241228e-05,
|
| 1387 |
+
"loss": 0.3342,
|
| 1388 |
+
"step": 197
|
| 1389 |
+
},
|
| 1390 |
+
{
|
| 1391 |
+
"epoch": 3.327731092436975,
|
| 1392 |
+
"grad_norm": 0.09055460205682467,
|
| 1393 |
+
"learning_rate": 2.3658365133229014e-05,
|
| 1394 |
+
"loss": 0.3339,
|
| 1395 |
+
"step": 198
|
| 1396 |
+
},
|
| 1397 |
+
{
|
| 1398 |
+
"epoch": 3.3445378151260505,
|
| 1399 |
+
"grad_norm": 0.09155888552387885,
|
| 1400 |
+
"learning_rate": 2.322669985324427e-05,
|
| 1401 |
+
"loss": 0.3334,
|
| 1402 |
+
"step": 199
|
| 1403 |
+
},
|
| 1404 |
+
{
|
| 1405 |
+
"epoch": 3.361344537815126,
|
| 1406 |
+
"grad_norm": 0.08505084693441671,
|
| 1407 |
+
"learning_rate": 2.2797391908939196e-05,
|
| 1408 |
+
"loss": 0.342,
|
| 1409 |
+
"step": 200
|
| 1410 |
+
},
|
| 1411 |
+
{
|
| 1412 |
+
"epoch": 3.3781512605042017,
|
| 1413 |
+
"grad_norm": 0.08273754791035418,
|
| 1414 |
+
"learning_rate": 2.2370501635663758e-05,
|
| 1415 |
+
"loss": 0.329,
|
| 1416 |
+
"step": 201
|
| 1417 |
+
},
|
| 1418 |
+
{
|
| 1419 |
+
"epoch": 3.3949579831932772,
|
| 1420 |
+
"grad_norm": 0.08079562060991577,
|
| 1421 |
+
"learning_rate": 2.194608902898615e-05,
|
| 1422 |
+
"loss": 0.3384,
|
| 1423 |
+
"step": 202
|
| 1424 |
+
},
|
| 1425 |
+
{
|
| 1426 |
+
"epoch": 3.411764705882353,
|
| 1427 |
+
"grad_norm": 0.07617281044869882,
|
| 1428 |
+
"learning_rate": 2.1524213736260914e-05,
|
| 1429 |
+
"loss": 0.3316,
|
| 1430 |
+
"step": 203
|
| 1431 |
+
},
|
| 1432 |
+
{
|
| 1433 |
+
"epoch": 3.4285714285714284,
|
| 1434 |
+
"grad_norm": 0.07890968188047062,
|
| 1435 |
+
"learning_rate": 2.110493504824603e-05,
|
| 1436 |
+
"loss": 0.331,
|
| 1437 |
+
"step": 204
|
| 1438 |
+
},
|
| 1439 |
+
{
|
| 1440 |
+
"epoch": 3.4453781512605044,
|
| 1441 |
+
"grad_norm": 0.08236740880146626,
|
| 1442 |
+
"learning_rate": 2.068831189077021e-05,
|
| 1443 |
+
"loss": 0.3235,
|
| 1444 |
+
"step": 205
|
| 1445 |
+
},
|
| 1446 |
+
{
|
| 1447 |
+
"epoch": 3.46218487394958,
|
| 1448 |
+
"grad_norm": 0.08010300188311276,
|
| 1449 |
+
"learning_rate": 2.0274402816451327e-05,
|
| 1450 |
+
"loss": 0.3354,
|
| 1451 |
+
"step": 206
|
| 1452 |
+
},
|
| 1453 |
+
{
|
| 1454 |
+
"epoch": 3.4789915966386555,
|
| 1455 |
+
"grad_norm": 0.07556676839360087,
|
| 1456 |
+
"learning_rate": 1.9863265996467395e-05,
|
| 1457 |
+
"loss": 0.3325,
|
| 1458 |
+
"step": 207
|
| 1459 |
+
},
|
| 1460 |
+
{
|
| 1461 |
+
"epoch": 3.495798319327731,
|
| 1462 |
+
"grad_norm": 0.08644829354649453,
|
| 1463 |
+
"learning_rate": 1.9454959212381196e-05,
|
| 1464 |
+
"loss": 0.3343,
|
| 1465 |
+
"step": 208
|
| 1466 |
+
},
|
| 1467 |
+
{
|
| 1468 |
+
"epoch": 3.5126050420168067,
|
| 1469 |
+
"grad_norm": 0.06933463203713842,
|
| 1470 |
+
"learning_rate": 1.9049539848019508e-05,
|
| 1471 |
+
"loss": 0.3267,
|
| 1472 |
+
"step": 209
|
| 1473 |
+
},
|
| 1474 |
+
{
|
| 1475 |
+
"epoch": 3.5294117647058822,
|
| 1476 |
+
"grad_norm": 0.07883663206049663,
|
| 1477 |
+
"learning_rate": 1.864706488140839e-05,
|
| 1478 |
+
"loss": 0.3381,
|
| 1479 |
+
"step": 210
|
| 1480 |
+
},
|
| 1481 |
+
{
|
| 1482 |
+
"epoch": 3.546218487394958,
|
| 1483 |
+
"grad_norm": 0.07440943171429551,
|
| 1484 |
+
"learning_rate": 1.8247590876765397e-05,
|
| 1485 |
+
"loss": 0.3328,
|
| 1486 |
+
"step": 211
|
| 1487 |
+
},
|
| 1488 |
+
{
|
| 1489 |
+
"epoch": 3.5630252100840334,
|
| 1490 |
+
"grad_norm": 0.07845358987260048,
|
| 1491 |
+
"learning_rate": 1.7851173976550074e-05,
|
| 1492 |
+
"loss": 0.3338,
|
| 1493 |
+
"step": 212
|
| 1494 |
+
},
|
| 1495 |
+
{
|
| 1496 |
+
"epoch": 3.5798319327731094,
|
| 1497 |
+
"grad_norm": 0.07667469429104415,
|
| 1498 |
+
"learning_rate": 1.745786989357352e-05,
|
| 1499 |
+
"loss": 0.3312,
|
| 1500 |
+
"step": 213
|
| 1501 |
+
},
|
| 1502 |
+
{
|
| 1503 |
+
"epoch": 3.596638655462185,
|
| 1504 |
+
"grad_norm": 0.06882240569281524,
|
| 1505 |
+
"learning_rate": 1.7067733903168514e-05,
|
| 1506 |
+
"loss": 0.3399,
|
| 1507 |
+
"step": 214
|
| 1508 |
+
},
|
| 1509 |
+
{
|
| 1510 |
+
"epoch": 3.6134453781512605,
|
| 1511 |
+
"grad_norm": 0.08858258063226424,
|
| 1512 |
+
"learning_rate": 1.6680820835421124e-05,
|
| 1513 |
+
"loss": 0.3283,
|
| 1514 |
+
"step": 215
|
| 1515 |
+
},
|
| 1516 |
+
{
|
| 1517 |
+
"epoch": 3.630252100840336,
|
| 1518 |
+
"grad_norm": 0.07785484814115333,
|
| 1519 |
+
"learning_rate": 1.629718506746473e-05,
|
| 1520 |
+
"loss": 0.3355,
|
| 1521 |
+
"step": 216
|
| 1522 |
+
},
|
| 1523 |
+
{
|
| 1524 |
+
"epoch": 3.6470588235294117,
|
| 1525 |
+
"grad_norm": 0.07404409723650303,
|
| 1526 |
+
"learning_rate": 1.5916880515837853e-05,
|
| 1527 |
+
"loss": 0.3358,
|
| 1528 |
+
"step": 217
|
| 1529 |
+
},
|
| 1530 |
+
{
|
| 1531 |
+
"epoch": 3.6638655462184873,
|
| 1532 |
+
"grad_norm": 0.06323103004624403,
|
| 1533 |
+
"learning_rate": 1.553996062890673e-05,
|
| 1534 |
+
"loss": 0.3296,
|
| 1535 |
+
"step": 218
|
| 1536 |
+
},
|
| 1537 |
+
{
|
| 1538 |
+
"epoch": 3.6806722689075633,
|
| 1539 |
+
"grad_norm": 0.08092456496616264,
|
| 1540 |
+
"learning_rate": 1.5166478379353527e-05,
|
| 1541 |
+
"loss": 0.3337,
|
| 1542 |
+
"step": 219
|
| 1543 |
+
},
|
| 1544 |
+
{
|
| 1545 |
+
"epoch": 3.697478991596639,
|
| 1546 |
+
"grad_norm": 0.07724817468137635,
|
| 1547 |
+
"learning_rate": 1.4796486256731561e-05,
|
| 1548 |
+
"loss": 0.3305,
|
| 1549 |
+
"step": 220
|
| 1550 |
+
},
|
| 1551 |
+
{
|
| 1552 |
+
"epoch": 3.7142857142857144,
|
| 1553 |
+
"grad_norm": 0.06304818954878921,
|
| 1554 |
+
"learning_rate": 1.4430036260088405e-05,
|
| 1555 |
+
"loss": 0.3373,
|
| 1556 |
+
"step": 221
|
| 1557 |
+
},
|
| 1558 |
+
{
|
| 1559 |
+
"epoch": 3.73109243697479,
|
| 1560 |
+
"grad_norm": 0.08283160664741292,
|
| 1561 |
+
"learning_rate": 1.4067179890657818e-05,
|
| 1562 |
+
"loss": 0.335,
|
| 1563 |
+
"step": 222
|
| 1564 |
+
},
|
| 1565 |
+
{
|
| 1566 |
+
"epoch": 3.7478991596638656,
|
| 1567 |
+
"grad_norm": 0.07071809987938908,
|
| 1568 |
+
"learning_rate": 1.3707968144621737e-05,
|
| 1569 |
+
"loss": 0.3335,
|
| 1570 |
+
"step": 223
|
| 1571 |
+
},
|
| 1572 |
+
{
|
| 1573 |
+
"epoch": 3.764705882352941,
|
| 1574 |
+
"grad_norm": 0.06519271417777431,
|
| 1575 |
+
"learning_rate": 1.3352451505943283e-05,
|
| 1576 |
+
"loss": 0.3285,
|
| 1577 |
+
"step": 224
|
| 1578 |
+
},
|
| 1579 |
+
{
|
| 1580 |
+
"epoch": 3.7815126050420167,
|
| 1581 |
+
"grad_norm": 0.08085939224611428,
|
| 1582 |
+
"learning_rate": 1.3000679939271588e-05,
|
| 1583 |
+
"loss": 0.3431,
|
| 1584 |
+
"step": 225
|
| 1585 |
+
},
|
| 1586 |
+
{
|
| 1587 |
+
"epoch": 3.7983193277310923,
|
| 1588 |
+
"grad_norm": 0.07763603372328087,
|
| 1589 |
+
"learning_rate": 1.2652702882919816e-05,
|
| 1590 |
+
"loss": 0.3213,
|
| 1591 |
+
"step": 226
|
| 1592 |
+
},
|
| 1593 |
+
{
|
| 1594 |
+
"epoch": 3.815126050420168,
|
| 1595 |
+
"grad_norm": 0.0743425344338226,
|
| 1596 |
+
"learning_rate": 1.2308569241916982e-05,
|
| 1597 |
+
"loss": 0.3274,
|
| 1598 |
+
"step": 227
|
| 1599 |
+
},
|
| 1600 |
+
{
|
| 1601 |
+
"epoch": 3.831932773109244,
|
| 1602 |
+
"grad_norm": 0.07404639932153928,
|
| 1603 |
+
"learning_rate": 1.196832738113486e-05,
|
| 1604 |
+
"loss": 0.331,
|
| 1605 |
+
"step": 228
|
| 1606 |
+
},
|
| 1607 |
+
{
|
| 1608 |
+
"epoch": 3.8487394957983194,
|
| 1609 |
+
"grad_norm": 0.07593531342958786,
|
| 1610 |
+
"learning_rate": 1.1632025118490668e-05,
|
| 1611 |
+
"loss": 0.3318,
|
| 1612 |
+
"step": 229
|
| 1613 |
+
},
|
| 1614 |
+
{
|
| 1615 |
+
"epoch": 3.865546218487395,
|
| 1616 |
+
"grad_norm": 0.07077344758948155,
|
| 1617 |
+
"learning_rate": 1.1299709718226745e-05,
|
| 1618 |
+
"loss": 0.3231,
|
| 1619 |
+
"step": 230
|
| 1620 |
+
},
|
| 1621 |
+
{
|
| 1622 |
+
"epoch": 3.8823529411764706,
|
| 1623 |
+
"grad_norm": 0.06548660169085967,
|
| 1624 |
+
"learning_rate": 1.0971427884267984e-05,
|
| 1625 |
+
"loss": 0.3352,
|
| 1626 |
+
"step": 231
|
| 1627 |
+
},
|
| 1628 |
+
{
|
| 1629 |
+
"epoch": 3.899159663865546,
|
| 1630 |
+
"grad_norm": 0.0799510965954409,
|
| 1631 |
+
"learning_rate": 1.064722575365797e-05,
|
| 1632 |
+
"loss": 0.3297,
|
| 1633 |
+
"step": 232
|
| 1634 |
+
},
|
| 1635 |
+
{
|
| 1636 |
+
"epoch": 3.9159663865546217,
|
| 1637 |
+
"grad_norm": 0.06903199628223637,
|
| 1638 |
+
"learning_rate": 1.032714889007488e-05,
|
| 1639 |
+
"loss": 0.3282,
|
| 1640 |
+
"step": 233
|
| 1641 |
+
},
|
| 1642 |
+
{
|
| 1643 |
+
"epoch": 3.9327731092436977,
|
| 1644 |
+
"grad_norm": 0.06314341603757684,
|
| 1645 |
+
"learning_rate": 1.0011242277427903e-05,
|
| 1646 |
+
"loss": 0.3365,
|
| 1647 |
+
"step": 234
|
| 1648 |
+
},
|
| 1649 |
+
{
|
| 1650 |
+
"epoch": 3.9495798319327733,
|
| 1651 |
+
"grad_norm": 0.0666298928944856,
|
| 1652 |
+
"learning_rate": 9.699550313535196e-06,
|
| 1653 |
+
"loss": 0.3362,
|
| 1654 |
+
"step": 235
|
| 1655 |
+
},
|
| 1656 |
+
{
|
| 1657 |
+
"epoch": 3.966386554621849,
|
| 1658 |
+
"grad_norm": 0.0627011771597765,
|
| 1659 |
+
"learning_rate": 9.392116803884113e-06,
|
| 1660 |
+
"loss": 0.331,
|
| 1661 |
+
"step": 236
|
| 1662 |
+
},
|
| 1663 |
+
{
|
| 1664 |
+
"epoch": 3.9831932773109244,
|
| 1665 |
+
"grad_norm": 0.060464466125568515,
|
| 1666 |
+
"learning_rate": 9.088984955474749e-06,
|
| 1667 |
+
"loss": 0.3368,
|
| 1668 |
+
"step": 237
|
| 1669 |
+
},
|
| 1670 |
+
{
|
| 1671 |
+
"epoch": 4.0,
|
| 1672 |
+
"grad_norm": 0.09356586036089908,
|
| 1673 |
+
"learning_rate": 8.790197370747662e-06,
|
| 1674 |
+
"loss": 0.4923,
|
| 1675 |
+
"step": 238
|
| 1676 |
+
},
|
| 1677 |
+
{
|
| 1678 |
+
"epoch": 4.016806722689076,
|
| 1679 |
+
"grad_norm": 0.10735324292426798,
|
| 1680 |
+
"learning_rate": 8.495796041596379e-06,
|
| 1681 |
+
"loss": 0.3164,
|
| 1682 |
+
"step": 239
|
| 1683 |
+
},
|
| 1684 |
+
{
|
| 1685 |
+
"epoch": 4.033613445378151,
|
| 1686 |
+
"grad_norm": 0.08906434120221782,
|
| 1687 |
+
"learning_rate": 8.205822343465865e-06,
|
| 1688 |
+
"loss": 0.3188,
|
| 1689 |
+
"step": 240
|
| 1690 |
+
},
|
| 1691 |
+
{
|
| 1692 |
+
"epoch": 4.050420168067227,
|
| 1693 |
+
"grad_norm": 0.07143739190636415,
|
| 1694 |
+
"learning_rate": 7.92031702953762e-06,
|
| 1695 |
+
"loss": 0.3163,
|
| 1696 |
+
"step": 241
|
| 1697 |
+
},
|
| 1698 |
+
{
|
| 1699 |
+
"epoch": 4.067226890756302,
|
| 1700 |
+
"grad_norm": 0.08129146282140881,
|
| 1701 |
+
"learning_rate": 7.639320225002106e-06,
|
| 1702 |
+
"loss": 0.3105,
|
| 1703 |
+
"step": 242
|
| 1704 |
+
},
|
| 1705 |
+
{
|
| 1706 |
+
"epoch": 4.084033613445378,
|
| 1707 |
+
"grad_norm": 0.09324166409313432,
|
| 1708 |
+
"learning_rate": 7.3628714214195505e-06,
|
| 1709 |
+
"loss": 0.3055,
|
| 1710 |
+
"step": 243
|
| 1711 |
+
},
|
| 1712 |
+
{
|
| 1713 |
+
"epoch": 4.100840336134453,
|
| 1714 |
+
"grad_norm": 0.0823634803270004,
|
| 1715 |
+
"learning_rate": 7.091009471169808e-06,
|
| 1716 |
+
"loss": 0.3236,
|
| 1717 |
+
"step": 244
|
| 1718 |
+
},
|
| 1719 |
+
{
|
| 1720 |
+
"epoch": 4.117647058823529,
|
| 1721 |
+
"grad_norm": 0.07870663627808307,
|
| 1722 |
+
"learning_rate": 6.82377258199193e-06,
|
| 1723 |
+
"loss": 0.3193,
|
| 1724 |
+
"step": 245
|
| 1725 |
+
},
|
| 1726 |
+
{
|
| 1727 |
+
"epoch": 4.1344537815126055,
|
| 1728 |
+
"grad_norm": 0.07946395327977603,
|
| 1729 |
+
"learning_rate": 6.561198311614458e-06,
|
| 1730 |
+
"loss": 0.3203,
|
| 1731 |
+
"step": 246
|
| 1732 |
+
},
|
| 1733 |
+
{
|
| 1734 |
+
"epoch": 4.151260504201681,
|
| 1735 |
+
"grad_norm": 0.07558498538686588,
|
| 1736 |
+
"learning_rate": 6.303323562477044e-06,
|
| 1737 |
+
"loss": 0.3158,
|
| 1738 |
+
"step": 247
|
| 1739 |
+
},
|
| 1740 |
+
{
|
| 1741 |
+
"epoch": 4.168067226890757,
|
| 1742 |
+
"grad_norm": 0.0709953014247398,
|
| 1743 |
+
"learning_rate": 6.050184576544089e-06,
|
| 1744 |
+
"loss": 0.3164,
|
| 1745 |
+
"step": 248
|
| 1746 |
+
},
|
| 1747 |
+
{
|
| 1748 |
+
"epoch": 4.184873949579832,
|
| 1749 |
+
"grad_norm": 0.07696195377606617,
|
| 1750 |
+
"learning_rate": 5.801816930211304e-06,
|
| 1751 |
+
"loss": 0.3098,
|
| 1752 |
+
"step": 249
|
| 1753 |
+
},
|
| 1754 |
+
{
|
| 1755 |
+
"epoch": 4.201680672268908,
|
| 1756 |
+
"grad_norm": 0.07621903426030113,
|
| 1757 |
+
"learning_rate": 5.558255529305779e-06,
|
| 1758 |
+
"loss": 0.3172,
|
| 1759 |
+
"step": 250
|
| 1760 |
+
},
|
| 1761 |
+
{
|
| 1762 |
+
"epoch": 4.218487394957983,
|
| 1763 |
+
"grad_norm": 0.0696988945230777,
|
| 1764 |
+
"learning_rate": 5.319534604180239e-06,
|
| 1765 |
+
"loss": 0.3101,
|
| 1766 |
+
"step": 251
|
| 1767 |
+
},
|
| 1768 |
+
{
|
| 1769 |
+
"epoch": 4.235294117647059,
|
| 1770 |
+
"grad_norm": 0.07047347116426476,
|
| 1771 |
+
"learning_rate": 5.085687704902311e-06,
|
| 1772 |
+
"loss": 0.3088,
|
| 1773 |
+
"step": 252
|
| 1774 |
+
},
|
| 1775 |
+
{
|
| 1776 |
+
"epoch": 4.2521008403361344,
|
| 1777 |
+
"grad_norm": 0.06567501377074629,
|
| 1778 |
+
"learning_rate": 4.856747696539352e-06,
|
| 1779 |
+
"loss": 0.317,
|
| 1780 |
+
"step": 253
|
| 1781 |
+
},
|
| 1782 |
+
{
|
| 1783 |
+
"epoch": 4.26890756302521,
|
| 1784 |
+
"grad_norm": 0.07033603665228617,
|
| 1785 |
+
"learning_rate": 4.632746754539592e-06,
|
| 1786 |
+
"loss": 0.3114,
|
| 1787 |
+
"step": 254
|
| 1788 |
+
},
|
| 1789 |
+
{
|
| 1790 |
+
"epoch": 4.285714285714286,
|
| 1791 |
+
"grad_norm": 0.0698122927823858,
|
| 1792 |
+
"learning_rate": 4.4137163602101114e-06,
|
| 1793 |
+
"loss": 0.315,
|
| 1794 |
+
"step": 255
|
| 1795 |
+
},
|
| 1796 |
+
{
|
| 1797 |
+
"epoch": 4.302521008403361,
|
| 1798 |
+
"grad_norm": 0.06239567204088085,
|
| 1799 |
+
"learning_rate": 4.199687296292463e-06,
|
| 1800 |
+
"loss": 0.3167,
|
| 1801 |
+
"step": 256
|
| 1802 |
+
},
|
| 1803 |
+
{
|
| 1804 |
+
"epoch": 4.319327731092437,
|
| 1805 |
+
"grad_norm": 0.0633239857016661,
|
| 1806 |
+
"learning_rate": 3.990689642636451e-06,
|
| 1807 |
+
"loss": 0.3209,
|
| 1808 |
+
"step": 257
|
| 1809 |
+
},
|
| 1810 |
+
{
|
| 1811 |
+
"epoch": 4.336134453781512,
|
| 1812 |
+
"grad_norm": 0.05877452888023069,
|
| 1813 |
+
"learning_rate": 3.786752771972633e-06,
|
| 1814 |
+
"loss": 0.3166,
|
| 1815 |
+
"step": 258
|
| 1816 |
+
},
|
| 1817 |
+
{
|
| 1818 |
+
"epoch": 4.352941176470588,
|
| 1819 |
+
"grad_norm": 0.06855271577553523,
|
| 1820 |
+
"learning_rate": 3.5879053457842727e-06,
|
| 1821 |
+
"loss": 0.3127,
|
| 1822 |
+
"step": 259
|
| 1823 |
+
},
|
| 1824 |
+
{
|
| 1825 |
+
"epoch": 4.369747899159664,
|
| 1826 |
+
"grad_norm": 0.06267028425256276,
|
| 1827 |
+
"learning_rate": 3.3941753102792617e-06,
|
| 1828 |
+
"loss": 0.3207,
|
| 1829 |
+
"step": 260
|
| 1830 |
+
},
|
| 1831 |
+
{
|
| 1832 |
+
"epoch": 4.38655462184874,
|
| 1833 |
+
"grad_norm": 0.057762863026921,
|
| 1834 |
+
"learning_rate": 3.205589892462464e-06,
|
| 1835 |
+
"loss": 0.3097,
|
| 1836 |
+
"step": 261
|
| 1837 |
+
},
|
| 1838 |
+
{
|
| 1839 |
+
"epoch": 4.4033613445378155,
|
| 1840 |
+
"grad_norm": 0.05916728390340108,
|
| 1841 |
+
"learning_rate": 3.0221755963092535e-06,
|
| 1842 |
+
"loss": 0.3086,
|
| 1843 |
+
"step": 262
|
| 1844 |
+
},
|
| 1845 |
+
{
|
| 1846 |
+
"epoch": 4.420168067226891,
|
| 1847 |
+
"grad_norm": 0.05839719330633518,
|
| 1848 |
+
"learning_rate": 2.8439581990406282e-06,
|
| 1849 |
+
"loss": 0.3235,
|
| 1850 |
+
"step": 263
|
| 1851 |
+
},
|
| 1852 |
+
{
|
| 1853 |
+
"epoch": 4.436974789915967,
|
| 1854 |
+
"grad_norm": 0.058111529162375726,
|
| 1855 |
+
"learning_rate": 2.670962747500405e-06,
|
| 1856 |
+
"loss": 0.3113,
|
| 1857 |
+
"step": 264
|
| 1858 |
+
},
|
| 1859 |
+
{
|
| 1860 |
+
"epoch": 4.453781512605042,
|
| 1861 |
+
"grad_norm": 0.05155427136563161,
|
| 1862 |
+
"learning_rate": 2.5032135546351644e-06,
|
| 1863 |
+
"loss": 0.3207,
|
| 1864 |
+
"step": 265
|
| 1865 |
+
},
|
| 1866 |
+
{
|
| 1867 |
+
"epoch": 4.470588235294118,
|
| 1868 |
+
"grad_norm": 0.054806384161329866,
|
| 1869 |
+
"learning_rate": 2.3407341960772543e-06,
|
| 1870 |
+
"loss": 0.3185,
|
| 1871 |
+
"step": 266
|
| 1872 |
+
},
|
| 1873 |
+
{
|
| 1874 |
+
"epoch": 4.487394957983193,
|
| 1875 |
+
"grad_norm": 0.05379784341548756,
|
| 1876 |
+
"learning_rate": 2.1835475068314692e-06,
|
| 1877 |
+
"loss": 0.3148,
|
| 1878 |
+
"step": 267
|
| 1879 |
+
},
|
| 1880 |
+
{
|
| 1881 |
+
"epoch": 4.504201680672269,
|
| 1882 |
+
"grad_norm": 0.05055903263153311,
|
| 1883 |
+
"learning_rate": 2.0316755780657705e-06,
|
| 1884 |
+
"loss": 0.3202,
|
| 1885 |
+
"step": 268
|
| 1886 |
+
},
|
| 1887 |
+
{
|
| 1888 |
+
"epoch": 4.5210084033613445,
|
| 1889 |
+
"grad_norm": 0.05165925203081087,
|
| 1890 |
+
"learning_rate": 1.8851397540065886e-06,
|
| 1891 |
+
"loss": 0.3099,
|
| 1892 |
+
"step": 269
|
| 1893 |
+
},
|
| 1894 |
+
{
|
| 1895 |
+
"epoch": 4.53781512605042,
|
| 1896 |
+
"grad_norm": 0.05954585796253012,
|
| 1897 |
+
"learning_rate": 1.7439606289391032e-06,
|
| 1898 |
+
"loss": 0.3147,
|
| 1899 |
+
"step": 270
|
| 1900 |
+
},
|
| 1901 |
+
{
|
| 1902 |
+
"epoch": 4.554621848739496,
|
| 1903 |
+
"grad_norm": 0.05086549032881178,
|
| 1904 |
+
"learning_rate": 1.608158044312864e-06,
|
| 1905 |
+
"loss": 0.308,
|
| 1906 |
+
"step": 271
|
| 1907 |
+
},
|
| 1908 |
+
{
|
| 1909 |
+
"epoch": 4.571428571428571,
|
| 1910 |
+
"grad_norm": 0.05118616843393958,
|
| 1911 |
+
"learning_rate": 1.4777510859532763e-06,
|
| 1912 |
+
"loss": 0.3141,
|
| 1913 |
+
"step": 272
|
| 1914 |
+
},
|
| 1915 |
+
{
|
| 1916 |
+
"epoch": 4.588235294117647,
|
| 1917 |
+
"grad_norm": 0.050068941732144816,
|
| 1918 |
+
"learning_rate": 1.3527580813792728e-06,
|
| 1919 |
+
"loss": 0.3141,
|
| 1920 |
+
"step": 273
|
| 1921 |
+
},
|
| 1922 |
+
{
|
| 1923 |
+
"epoch": 4.605042016806722,
|
| 1924 |
+
"grad_norm": 0.050986087169577816,
|
| 1925 |
+
"learning_rate": 1.2331965972275107e-06,
|
| 1926 |
+
"loss": 0.319,
|
| 1927 |
+
"step": 274
|
| 1928 |
+
},
|
| 1929 |
+
{
|
| 1930 |
+
"epoch": 4.621848739495798,
|
| 1931 |
+
"grad_norm": 0.048541058817604836,
|
| 1932 |
+
"learning_rate": 1.1190834367835701e-06,
|
| 1933 |
+
"loss": 0.3131,
|
| 1934 |
+
"step": 275
|
| 1935 |
+
},
|
| 1936 |
+
{
|
| 1937 |
+
"epoch": 4.6386554621848735,
|
| 1938 |
+
"grad_norm": 0.04755798444633252,
|
| 1939 |
+
"learning_rate": 1.010434637620401e-06,
|
| 1940 |
+
"loss": 0.3211,
|
| 1941 |
+
"step": 276
|
| 1942 |
+
},
|
| 1943 |
+
{
|
| 1944 |
+
"epoch": 4.65546218487395,
|
| 1945 |
+
"grad_norm": 0.05166544989005396,
|
| 1946 |
+
"learning_rate": 9.072654693443783e-07,
|
| 1947 |
+
"loss": 0.3087,
|
| 1948 |
+
"step": 277
|
| 1949 |
+
},
|
| 1950 |
+
{
|
| 1951 |
+
"epoch": 4.6722689075630255,
|
| 1952 |
+
"grad_norm": 0.05005903779456428,
|
| 1953 |
+
"learning_rate": 8.095904314492942e-07,
|
| 1954 |
+
"loss": 0.3155,
|
| 1955 |
+
"step": 278
|
| 1956 |
+
},
|
| 1957 |
+
{
|
| 1958 |
+
"epoch": 4.689075630252101,
|
| 1959 |
+
"grad_norm": 0.05107913700639846,
|
| 1960 |
+
"learning_rate": 7.174232512785973e-07,
|
| 1961 |
+
"loss": 0.3158,
|
| 1962 |
+
"step": 279
|
| 1963 |
+
},
|
| 1964 |
+
{
|
| 1965 |
+
"epoch": 4.705882352941177,
|
| 1966 |
+
"grad_norm": 0.046696470698089916,
|
| 1967 |
+
"learning_rate": 6.307768820961269e-07,
|
| 1968 |
+
"loss": 0.3155,
|
| 1969 |
+
"step": 280
|
| 1970 |
+
},
|
| 1971 |
+
{
|
| 1972 |
+
"epoch": 4.722689075630252,
|
| 1973 |
+
"grad_norm": 0.04921954144549168,
|
| 1974 |
+
"learning_rate": 5.496635012656626e-07,
|
| 1975 |
+
"loss": 0.3205,
|
| 1976 |
+
"step": 281
|
| 1977 |
+
},
|
| 1978 |
+
{
|
| 1979 |
+
"epoch": 4.739495798319328,
|
| 1980 |
+
"grad_norm": 0.049051214749879606,
|
| 1981 |
+
"learning_rate": 4.740945085394888e-07,
|
| 1982 |
+
"loss": 0.3102,
|
| 1983 |
+
"step": 282
|
| 1984 |
+
},
|
| 1985 |
+
{
|
| 1986 |
+
"epoch": 4.756302521008403,
|
| 1987 |
+
"grad_norm": 0.050177911281723414,
|
| 1988 |
+
"learning_rate": 4.040805244562762e-07,
|
| 1989 |
+
"loss": 0.3197,
|
| 1990 |
+
"step": 283
|
| 1991 |
+
},
|
| 1992 |
+
{
|
| 1993 |
+
"epoch": 4.773109243697479,
|
| 1994 |
+
"grad_norm": 0.049320398857828206,
|
| 1995 |
+
"learning_rate": 3.3963138884845817e-07,
|
| 1996 |
+
"loss": 0.3099,
|
| 1997 |
+
"step": 284
|
| 1998 |
+
},
|
| 1999 |
+
{
|
| 2000 |
+
"epoch": 4.7899159663865545,
|
| 2001 |
+
"grad_norm": 0.053838116706168226,
|
| 2002 |
+
"learning_rate": 2.8075615945932333e-07,
|
| 2003 |
+
"loss": 0.3174,
|
| 2004 |
+
"step": 285
|
| 2005 |
+
},
|
| 2006 |
+
{
|
| 2007 |
+
"epoch": 4.80672268907563,
|
| 2008 |
+
"grad_norm": 0.04745995537684286,
|
| 2009 |
+
"learning_rate": 2.2746311067004312e-07,
|
| 2010 |
+
"loss": 0.3195,
|
| 2011 |
+
"step": 286
|
| 2012 |
+
},
|
| 2013 |
+
{
|
| 2014 |
+
"epoch": 4.823529411764706,
|
| 2015 |
+
"grad_norm": 0.05360166057822884,
|
| 2016 |
+
"learning_rate": 1.7975973233677058e-07,
|
| 2017 |
+
"loss": 0.3181,
|
| 2018 |
+
"step": 287
|
| 2019 |
+
},
|
| 2020 |
+
{
|
| 2021 |
+
"epoch": 4.840336134453781,
|
| 2022 |
+
"grad_norm": 0.04830776523177845,
|
| 2023 |
+
"learning_rate": 1.3765272873802914e-07,
|
| 2024 |
+
"loss": 0.31,
|
| 2025 |
+
"step": 288
|
| 2026 |
+
},
|
| 2027 |
+
{
|
| 2028 |
+
"epoch": 4.857142857142857,
|
| 2029 |
+
"grad_norm": 0.047076064205271655,
|
| 2030 |
+
"learning_rate": 1.0114801763245751e-07,
|
| 2031 |
+
"loss": 0.3206,
|
| 2032 |
+
"step": 289
|
| 2033 |
+
},
|
| 2034 |
+
{
|
| 2035 |
+
"epoch": 4.873949579831933,
|
| 2036 |
+
"grad_norm": 0.04663873732271828,
|
| 2037 |
+
"learning_rate": 7.025072942714595e-08,
|
| 2038 |
+
"loss": 0.31,
|
| 2039 |
+
"step": 290
|
| 2040 |
+
},
|
| 2041 |
+
{
|
| 2042 |
+
"epoch": 4.890756302521009,
|
| 2043 |
+
"grad_norm": 0.050541819941751935,
|
| 2044 |
+
"learning_rate": 4.496520645659086e-08,
|
| 2045 |
+
"loss": 0.3187,
|
| 2046 |
+
"step": 291
|
| 2047 |
+
},
|
| 2048 |
+
{
|
| 2049 |
+
"epoch": 4.907563025210084,
|
| 2050 |
+
"grad_norm": 0.0480153474982601,
|
| 2051 |
+
"learning_rate": 2.5295002372427436e-08,
|
| 2052 |
+
"loss": 0.3135,
|
| 2053 |
+
"step": 292
|
| 2054 |
+
},
|
| 2055 |
+
{
|
| 2056 |
+
"epoch": 4.92436974789916,
|
| 2057 |
+
"grad_norm": 0.046945149107523426,
|
| 2058 |
+
"learning_rate": 1.1242881643993563e-08,
|
| 2059 |
+
"loss": 0.3209,
|
| 2060 |
+
"step": 293
|
| 2061 |
+
},
|
| 2062 |
+
{
|
| 2063 |
+
"epoch": 4.9411764705882355,
|
| 2064 |
+
"grad_norm": 0.04646328576480192,
|
| 2065 |
+
"learning_rate": 2.8108191698050793e-09,
|
| 2066 |
+
"loss": 0.3105,
|
| 2067 |
+
"step": 294
|
| 2068 |
+
},
|
| 2069 |
+
{
|
| 2070 |
+
"epoch": 4.957983193277311,
|
| 2071 |
+
"grad_norm": 0.04680709825119643,
|
| 2072 |
+
"learning_rate": 0.0,
|
| 2073 |
+
"loss": 0.316,
|
| 2074 |
+
"step": 295
|
| 2075 |
+
},
|
| 2076 |
+
{
|
| 2077 |
+
"epoch": 4.957983193277311,
|
| 2078 |
+
"step": 295,
|
| 2079 |
+
"total_flos": 8.686638104246747e+18,
|
| 2080 |
+
"train_loss": 0.3853245627071898,
|
| 2081 |
+
"train_runtime": 55023.1167,
|
| 2082 |
+
"train_samples_per_second": 2.766,
|
| 2083 |
+
"train_steps_per_second": 0.005
|
| 2084 |
+
}
|
| 2085 |
+
],
|
| 2086 |
+
"logging_steps": 1.0,
|
| 2087 |
+
"max_steps": 295,
|
| 2088 |
+
"num_input_tokens_seen": 0,
|
| 2089 |
+
"num_train_epochs": 5,
|
| 2090 |
+
"save_steps": 500,
|
| 2091 |
+
"stateful_callbacks": {
|
| 2092 |
+
"TrainerControl": {
|
| 2093 |
+
"args": {
|
| 2094 |
+
"should_epoch_stop": false,
|
| 2095 |
+
"should_evaluate": false,
|
| 2096 |
+
"should_log": false,
|
| 2097 |
+
"should_save": true,
|
| 2098 |
+
"should_training_stop": true
|
| 2099 |
+
},
|
| 2100 |
+
"attributes": {}
|
| 2101 |
+
}
|
| 2102 |
+
},
|
| 2103 |
+
"total_flos": 8.686638104246747e+18,
|
| 2104 |
+
"train_batch_size": 1,
|
| 2105 |
+
"trial_name": null,
|
| 2106 |
+
"trial_params": null
|
| 2107 |
+
}
|
training_loss.png
ADDED
|