Commit
•
ef4cb37
1
Parent(s):
642eb33
End of training
Browse files- README.md +2 -1
- all_results.json +8 -0
- train_results.json +8 -0
- trainer_state.json +1351 -0
- training_loss.png +0 -0
README.md
CHANGED
|
@@ -4,6 +4,7 @@ license: llama3.1
|
|
| 4 |
base_model: meta-llama/Llama-Guard-3-8B
|
| 5 |
tags:
|
| 6 |
- llama-factory
|
|
|
|
| 7 |
- generated_from_trainer
|
| 8 |
model-index:
|
| 9 |
- name: laaj_5k_3epochs_all_Scode
|
|
@@ -15,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 15 |
|
| 16 |
# laaj_5k_3epochs_all_Scode
|
| 17 |
|
| 18 |
-
This model is a fine-tuned version of [meta-llama/Llama-Guard-3-8B](https://huggingface.co/meta-llama/Llama-Guard-3-8B) on
|
| 19 |
|
| 20 |
## Model description
|
| 21 |
|
|
|
|
| 4 |
base_model: meta-llama/Llama-Guard-3-8B
|
| 5 |
tags:
|
| 6 |
- llama-factory
|
| 7 |
+
- lora
|
| 8 |
- generated_from_trainer
|
| 9 |
model-index:
|
| 10 |
- name: laaj_5k_3epochs_all_Scode
|
|
|
|
| 16 |
|
| 17 |
# laaj_5k_3epochs_all_Scode
|
| 18 |
|
| 19 |
+
This model is a fine-tuned version of [meta-llama/Llama-Guard-3-8B](https://huggingface.co/meta-llama/Llama-Guard-3-8B) on the laaj_5k_allbatch_Scode dataset.
|
| 20 |
|
| 21 |
## Model description
|
| 22 |
|
all_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"total_flos": 299543991877632.0,
|
| 4 |
+
"train_loss": 0.3980048195521037,
|
| 5 |
+
"train_runtime": 15877.4798,
|
| 6 |
+
"train_samples_per_second": 0.945,
|
| 7 |
+
"train_steps_per_second": 0.118
|
| 8 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"total_flos": 299543991877632.0,
|
| 4 |
+
"train_loss": 0.3980048195521037,
|
| 5 |
+
"train_runtime": 15877.4798,
|
| 6 |
+
"train_samples_per_second": 0.945,
|
| 7 |
+
"train_steps_per_second": 0.118
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1351 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 3.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 1875,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.016,
|
| 13 |
+
"grad_norm": 1.5132097727696663,
|
| 14 |
+
"learning_rate": 9.999298177883903e-05,
|
| 15 |
+
"loss": 1.6713,
|
| 16 |
+
"step": 10
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.032,
|
| 20 |
+
"grad_norm": 1.0288849642854194,
|
| 21 |
+
"learning_rate": 9.997192908557323e-05,
|
| 22 |
+
"loss": 0.9562,
|
| 23 |
+
"step": 20
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.048,
|
| 27 |
+
"grad_norm": 2.628046163977058,
|
| 28 |
+
"learning_rate": 9.993684783030088e-05,
|
| 29 |
+
"loss": 0.7998,
|
| 30 |
+
"step": 30
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.064,
|
| 34 |
+
"grad_norm": 1.0977335920429718,
|
| 35 |
+
"learning_rate": 9.988774786134234e-05,
|
| 36 |
+
"loss": 0.7019,
|
| 37 |
+
"step": 40
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.08,
|
| 41 |
+
"grad_norm": 1.143511217142594,
|
| 42 |
+
"learning_rate": 9.982464296247522e-05,
|
| 43 |
+
"loss": 0.6965,
|
| 44 |
+
"step": 50
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.096,
|
| 48 |
+
"grad_norm": 1.1886880844299708,
|
| 49 |
+
"learning_rate": 9.974755084906502e-05,
|
| 50 |
+
"loss": 0.6417,
|
| 51 |
+
"step": 60
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.112,
|
| 55 |
+
"grad_norm": 1.3088809099781331,
|
| 56 |
+
"learning_rate": 9.965649316309178e-05,
|
| 57 |
+
"loss": 0.6388,
|
| 58 |
+
"step": 70
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.128,
|
| 62 |
+
"grad_norm": 2.0299623866899035,
|
| 63 |
+
"learning_rate": 9.955149546707465e-05,
|
| 64 |
+
"loss": 0.6665,
|
| 65 |
+
"step": 80
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.144,
|
| 69 |
+
"grad_norm": 1.5251129511711958,
|
| 70 |
+
"learning_rate": 9.94325872368957e-05,
|
| 71 |
+
"loss": 0.6014,
|
| 72 |
+
"step": 90
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.16,
|
| 76 |
+
"grad_norm": 1.307242452061001,
|
| 77 |
+
"learning_rate": 9.929980185352526e-05,
|
| 78 |
+
"loss": 0.5971,
|
| 79 |
+
"step": 100
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.176,
|
| 83 |
+
"grad_norm": 1.0674735946460283,
|
| 84 |
+
"learning_rate": 9.915317659365077e-05,
|
| 85 |
+
"loss": 0.6206,
|
| 86 |
+
"step": 110
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.192,
|
| 90 |
+
"grad_norm": 1.3671834842056045,
|
| 91 |
+
"learning_rate": 9.899275261921234e-05,
|
| 92 |
+
"loss": 0.5597,
|
| 93 |
+
"step": 120
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.208,
|
| 97 |
+
"grad_norm": 1.3041158193336297,
|
| 98 |
+
"learning_rate": 9.881857496584726e-05,
|
| 99 |
+
"loss": 0.5774,
|
| 100 |
+
"step": 130
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.224,
|
| 104 |
+
"grad_norm": 1.2739151782924372,
|
| 105 |
+
"learning_rate": 9.863069253024719e-05,
|
| 106 |
+
"loss": 0.5656,
|
| 107 |
+
"step": 140
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.24,
|
| 111 |
+
"grad_norm": 1.1009566652519438,
|
| 112 |
+
"learning_rate": 9.842915805643155e-05,
|
| 113 |
+
"loss": 0.5724,
|
| 114 |
+
"step": 150
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.256,
|
| 118 |
+
"grad_norm": 1.2004741768824123,
|
| 119 |
+
"learning_rate": 9.821402812094073e-05,
|
| 120 |
+
"loss": 0.5796,
|
| 121 |
+
"step": 160
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.272,
|
| 125 |
+
"grad_norm": 1.5111413381395908,
|
| 126 |
+
"learning_rate": 9.798536311695334e-05,
|
| 127 |
+
"loss": 0.5556,
|
| 128 |
+
"step": 170
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.288,
|
| 132 |
+
"grad_norm": 1.0331976579035735,
|
| 133 |
+
"learning_rate": 9.774322723733216e-05,
|
| 134 |
+
"loss": 0.5451,
|
| 135 |
+
"step": 180
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.304,
|
| 139 |
+
"grad_norm": 1.1745428366022237,
|
| 140 |
+
"learning_rate": 9.748768845660334e-05,
|
| 141 |
+
"loss": 0.5605,
|
| 142 |
+
"step": 190
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.32,
|
| 146 |
+
"grad_norm": 1.067749681555933,
|
| 147 |
+
"learning_rate": 9.721881851187406e-05,
|
| 148 |
+
"loss": 0.5644,
|
| 149 |
+
"step": 200
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.336,
|
| 153 |
+
"grad_norm": 1.1773627454789788,
|
| 154 |
+
"learning_rate": 9.693669288269372e-05,
|
| 155 |
+
"loss": 0.5658,
|
| 156 |
+
"step": 210
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.352,
|
| 160 |
+
"grad_norm": 1.4049555802409854,
|
| 161 |
+
"learning_rate": 9.664139076986473e-05,
|
| 162 |
+
"loss": 0.5483,
|
| 163 |
+
"step": 220
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.368,
|
| 167 |
+
"grad_norm": 1.436137944296166,
|
| 168 |
+
"learning_rate": 9.63329950732086e-05,
|
| 169 |
+
"loss": 0.5376,
|
| 170 |
+
"step": 230
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.384,
|
| 174 |
+
"grad_norm": 1.2686595633739197,
|
| 175 |
+
"learning_rate": 9.601159236829352e-05,
|
| 176 |
+
"loss": 0.5402,
|
| 177 |
+
"step": 240
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.4,
|
| 181 |
+
"grad_norm": 1.4062419829256017,
|
| 182 |
+
"learning_rate": 9.567727288213005e-05,
|
| 183 |
+
"loss": 0.5552,
|
| 184 |
+
"step": 250
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.416,
|
| 188 |
+
"grad_norm": 0.9610102580305002,
|
| 189 |
+
"learning_rate": 9.533013046784189e-05,
|
| 190 |
+
"loss": 0.5118,
|
| 191 |
+
"step": 260
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.432,
|
| 195 |
+
"grad_norm": 1.2008075396073077,
|
| 196 |
+
"learning_rate": 9.497026257831855e-05,
|
| 197 |
+
"loss": 0.5479,
|
| 198 |
+
"step": 270
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.448,
|
| 202 |
+
"grad_norm": 1.116478910009909,
|
| 203 |
+
"learning_rate": 9.459777023885755e-05,
|
| 204 |
+
"loss": 0.4773,
|
| 205 |
+
"step": 280
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.464,
|
| 209 |
+
"grad_norm": 1.2287813088079407,
|
| 210 |
+
"learning_rate": 9.421275801880362e-05,
|
| 211 |
+
"loss": 0.5149,
|
| 212 |
+
"step": 290
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.48,
|
| 216 |
+
"grad_norm": 1.3153026196754682,
|
| 217 |
+
"learning_rate": 9.381533400219318e-05,
|
| 218 |
+
"loss": 0.5209,
|
| 219 |
+
"step": 300
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.496,
|
| 223 |
+
"grad_norm": 1.6158648577050707,
|
| 224 |
+
"learning_rate": 9.340560975741197e-05,
|
| 225 |
+
"loss": 0.5065,
|
| 226 |
+
"step": 310
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.512,
|
| 230 |
+
"grad_norm": 1.1601862022387133,
|
| 231 |
+
"learning_rate": 9.298370030587456e-05,
|
| 232 |
+
"loss": 0.515,
|
| 233 |
+
"step": 320
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.528,
|
| 237 |
+
"grad_norm": 1.2591523341167126,
|
| 238 |
+
"learning_rate": 9.254972408973461e-05,
|
| 239 |
+
"loss": 0.5375,
|
| 240 |
+
"step": 330
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.544,
|
| 244 |
+
"grad_norm": 1.0361296810068354,
|
| 245 |
+
"learning_rate": 9.210380293863462e-05,
|
| 246 |
+
"loss": 0.5085,
|
| 247 |
+
"step": 340
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.56,
|
| 251 |
+
"grad_norm": 1.0654093627590127,
|
| 252 |
+
"learning_rate": 9.164606203550497e-05,
|
| 253 |
+
"loss": 0.5137,
|
| 254 |
+
"step": 350
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.576,
|
| 258 |
+
"grad_norm": 1.2697508248750302,
|
| 259 |
+
"learning_rate": 9.117662988142138e-05,
|
| 260 |
+
"loss": 0.4906,
|
| 261 |
+
"step": 360
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.592,
|
| 265 |
+
"grad_norm": 1.2602467399423862,
|
| 266 |
+
"learning_rate": 9.069563825953092e-05,
|
| 267 |
+
"loss": 0.4966,
|
| 268 |
+
"step": 370
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.608,
|
| 272 |
+
"grad_norm": 2.0044411177134918,
|
| 273 |
+
"learning_rate": 9.020322219805674e-05,
|
| 274 |
+
"loss": 0.5063,
|
| 275 |
+
"step": 380
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.624,
|
| 279 |
+
"grad_norm": 1.148187026295081,
|
| 280 |
+
"learning_rate": 8.969951993239177e-05,
|
| 281 |
+
"loss": 0.4864,
|
| 282 |
+
"step": 390
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.64,
|
| 286 |
+
"grad_norm": 1.0529856886801567,
|
| 287 |
+
"learning_rate": 8.9184672866292e-05,
|
| 288 |
+
"loss": 0.4632,
|
| 289 |
+
"step": 400
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.656,
|
| 293 |
+
"grad_norm": 1.1830984654167522,
|
| 294 |
+
"learning_rate": 8.865882553218037e-05,
|
| 295 |
+
"loss": 0.5097,
|
| 296 |
+
"step": 410
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.672,
|
| 300 |
+
"grad_norm": 1.1495869848318996,
|
| 301 |
+
"learning_rate": 8.81221255505724e-05,
|
| 302 |
+
"loss": 0.4795,
|
| 303 |
+
"step": 420
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.688,
|
| 307 |
+
"grad_norm": 1.1235900402260404,
|
| 308 |
+
"learning_rate": 8.757472358863481e-05,
|
| 309 |
+
"loss": 0.5006,
|
| 310 |
+
"step": 430
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.704,
|
| 314 |
+
"grad_norm": 1.2883600054948787,
|
| 315 |
+
"learning_rate": 8.701677331788891e-05,
|
| 316 |
+
"loss": 0.4822,
|
| 317 |
+
"step": 440
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.72,
|
| 321 |
+
"grad_norm": 1.0725324329031838,
|
| 322 |
+
"learning_rate": 8.644843137107059e-05,
|
| 323 |
+
"loss": 0.4856,
|
| 324 |
+
"step": 450
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.736,
|
| 328 |
+
"grad_norm": 2.9167440508075955,
|
| 329 |
+
"learning_rate": 8.586985729815894e-05,
|
| 330 |
+
"loss": 0.4648,
|
| 331 |
+
"step": 460
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.752,
|
| 335 |
+
"grad_norm": 0.9522185207229502,
|
| 336 |
+
"learning_rate": 8.528121352158604e-05,
|
| 337 |
+
"loss": 0.488,
|
| 338 |
+
"step": 470
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.768,
|
| 342 |
+
"grad_norm": 1.5907707346034905,
|
| 343 |
+
"learning_rate": 8.468266529064025e-05,
|
| 344 |
+
"loss": 0.5039,
|
| 345 |
+
"step": 480
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.784,
|
| 349 |
+
"grad_norm": 1.1370580239916313,
|
| 350 |
+
"learning_rate": 8.4074380635076e-05,
|
| 351 |
+
"loss": 0.4737,
|
| 352 |
+
"step": 490
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.8,
|
| 356 |
+
"grad_norm": 0.9782419363730104,
|
| 357 |
+
"learning_rate": 8.345653031794292e-05,
|
| 358 |
+
"loss": 0.5296,
|
| 359 |
+
"step": 500
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.816,
|
| 363 |
+
"grad_norm": 1.1919631254125505,
|
| 364 |
+
"learning_rate": 8.282928778764783e-05,
|
| 365 |
+
"loss": 0.4688,
|
| 366 |
+
"step": 510
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.832,
|
| 370 |
+
"grad_norm": 1.2994447479210165,
|
| 371 |
+
"learning_rate": 8.21928291292627e-05,
|
| 372 |
+
"loss": 0.4675,
|
| 373 |
+
"step": 520
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.848,
|
| 377 |
+
"grad_norm": 0.9689627531529147,
|
| 378 |
+
"learning_rate": 8.154733301509248e-05,
|
| 379 |
+
"loss": 0.4408,
|
| 380 |
+
"step": 530
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.864,
|
| 384 |
+
"grad_norm": 1.1649720718750445,
|
| 385 |
+
"learning_rate": 8.089298065451672e-05,
|
| 386 |
+
"loss": 0.4608,
|
| 387 |
+
"step": 540
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.88,
|
| 391 |
+
"grad_norm": 1.3438713492048824,
|
| 392 |
+
"learning_rate": 8.022995574311876e-05,
|
| 393 |
+
"loss": 0.4476,
|
| 394 |
+
"step": 550
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.896,
|
| 398 |
+
"grad_norm": 1.0737706296113363,
|
| 399 |
+
"learning_rate": 7.95584444111171e-05,
|
| 400 |
+
"loss": 0.5027,
|
| 401 |
+
"step": 560
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.912,
|
| 405 |
+
"grad_norm": 1.3469637742497171,
|
| 406 |
+
"learning_rate": 7.887863517111338e-05,
|
| 407 |
+
"loss": 0.4929,
|
| 408 |
+
"step": 570
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.928,
|
| 412 |
+
"grad_norm": 1.1251117260635928,
|
| 413 |
+
"learning_rate": 7.819071886517134e-05,
|
| 414 |
+
"loss": 0.4739,
|
| 415 |
+
"step": 580
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.944,
|
| 419 |
+
"grad_norm": 1.0579726403654415,
|
| 420 |
+
"learning_rate": 7.7494888611242e-05,
|
| 421 |
+
"loss": 0.4529,
|
| 422 |
+
"step": 590
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.96,
|
| 426 |
+
"grad_norm": 1.0498899044694738,
|
| 427 |
+
"learning_rate": 7.679133974894983e-05,
|
| 428 |
+
"loss": 0.4619,
|
| 429 |
+
"step": 600
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.976,
|
| 433 |
+
"grad_norm": 1.2353618604203835,
|
| 434 |
+
"learning_rate": 7.60802697847554e-05,
|
| 435 |
+
"loss": 0.462,
|
| 436 |
+
"step": 610
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.992,
|
| 440 |
+
"grad_norm": 1.4618575476184537,
|
| 441 |
+
"learning_rate": 7.536187833650947e-05,
|
| 442 |
+
"loss": 0.4478,
|
| 443 |
+
"step": 620
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 1.008,
|
| 447 |
+
"grad_norm": 0.9313196650505406,
|
| 448 |
+
"learning_rate": 7.463636707741458e-05,
|
| 449 |
+
"loss": 0.4304,
|
| 450 |
+
"step": 630
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 1.024,
|
| 454 |
+
"grad_norm": 1.1239041555920188,
|
| 455 |
+
"learning_rate": 7.390393967940962e-05,
|
| 456 |
+
"loss": 0.4138,
|
| 457 |
+
"step": 640
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 1.04,
|
| 461 |
+
"grad_norm": 1.8736342455754906,
|
| 462 |
+
"learning_rate": 7.316480175599309e-05,
|
| 463 |
+
"loss": 0.3909,
|
| 464 |
+
"step": 650
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 1.056,
|
| 468 |
+
"grad_norm": 1.3075877434881118,
|
| 469 |
+
"learning_rate": 7.241916080450163e-05,
|
| 470 |
+
"loss": 0.3873,
|
| 471 |
+
"step": 660
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 1.072,
|
| 475 |
+
"grad_norm": 1.1366815390360774,
|
| 476 |
+
"learning_rate": 7.166722614785937e-05,
|
| 477 |
+
"loss": 0.3744,
|
| 478 |
+
"step": 670
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 1.088,
|
| 482 |
+
"grad_norm": 1.1482160099998533,
|
| 483 |
+
"learning_rate": 7.090920887581506e-05,
|
| 484 |
+
"loss": 0.3641,
|
| 485 |
+
"step": 680
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 1.104,
|
| 489 |
+
"grad_norm": 1.4903122013208903,
|
| 490 |
+
"learning_rate": 7.014532178568314e-05,
|
| 491 |
+
"loss": 0.3708,
|
| 492 |
+
"step": 690
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 1.12,
|
| 496 |
+
"grad_norm": 1.2283466862099195,
|
| 497 |
+
"learning_rate": 6.937577932260515e-05,
|
| 498 |
+
"loss": 0.4147,
|
| 499 |
+
"step": 700
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 1.1360000000000001,
|
| 503 |
+
"grad_norm": 1.044360506141548,
|
| 504 |
+
"learning_rate": 6.860079751934908e-05,
|
| 505 |
+
"loss": 0.3698,
|
| 506 |
+
"step": 710
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 1.152,
|
| 510 |
+
"grad_norm": 1.9659986049038847,
|
| 511 |
+
"learning_rate": 6.782059393566253e-05,
|
| 512 |
+
"loss": 0.3768,
|
| 513 |
+
"step": 720
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 1.168,
|
| 517 |
+
"grad_norm": 1.1559700548705432,
|
| 518 |
+
"learning_rate": 6.70353875971976e-05,
|
| 519 |
+
"loss": 0.3713,
|
| 520 |
+
"step": 730
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 1.184,
|
| 524 |
+
"grad_norm": 1.3638599004402776,
|
| 525 |
+
"learning_rate": 6.624539893402382e-05,
|
| 526 |
+
"loss": 0.376,
|
| 527 |
+
"step": 740
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 1.2,
|
| 531 |
+
"grad_norm": 1.2823520816693201,
|
| 532 |
+
"learning_rate": 6.545084971874738e-05,
|
| 533 |
+
"loss": 0.3762,
|
| 534 |
+
"step": 750
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 1.216,
|
| 538 |
+
"grad_norm": 1.1889832443284072,
|
| 539 |
+
"learning_rate": 6.465196300425287e-05,
|
| 540 |
+
"loss": 0.3891,
|
| 541 |
+
"step": 760
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 1.232,
|
| 545 |
+
"grad_norm": 1.128194765929284,
|
| 546 |
+
"learning_rate": 6.384896306108612e-05,
|
| 547 |
+
"loss": 0.3772,
|
| 548 |
+
"step": 770
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 1.248,
|
| 552 |
+
"grad_norm": 1.4491197750693638,
|
| 553 |
+
"learning_rate": 6.304207531449486e-05,
|
| 554 |
+
"loss": 0.3678,
|
| 555 |
+
"step": 780
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 1.264,
|
| 559 |
+
"grad_norm": 1.5686675701147237,
|
| 560 |
+
"learning_rate": 6.223152628114537e-05,
|
| 561 |
+
"loss": 0.3648,
|
| 562 |
+
"step": 790
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 1.28,
|
| 566 |
+
"grad_norm": 1.0902715474500886,
|
| 567 |
+
"learning_rate": 6.141754350553279e-05,
|
| 568 |
+
"loss": 0.3909,
|
| 569 |
+
"step": 800
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 1.296,
|
| 573 |
+
"grad_norm": 1.6337999430835446,
|
| 574 |
+
"learning_rate": 6.0600355496102745e-05,
|
| 575 |
+
"loss": 0.383,
|
| 576 |
+
"step": 810
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 1.312,
|
| 580 |
+
"grad_norm": 1.321659745725057,
|
| 581 |
+
"learning_rate": 5.9780191661102415e-05,
|
| 582 |
+
"loss": 0.3802,
|
| 583 |
+
"step": 820
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 1.328,
|
| 587 |
+
"grad_norm": 1.6867441595334431,
|
| 588 |
+
"learning_rate": 5.8957282244179124e-05,
|
| 589 |
+
"loss": 0.423,
|
| 590 |
+
"step": 830
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 1.3439999999999999,
|
| 594 |
+
"grad_norm": 1.2441664958096805,
|
| 595 |
+
"learning_rate": 5.813185825974419e-05,
|
| 596 |
+
"loss": 0.3641,
|
| 597 |
+
"step": 840
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 1.3599999999999999,
|
| 601 |
+
"grad_norm": 1.6422801866133996,
|
| 602 |
+
"learning_rate": 5.730415142812059e-05,
|
| 603 |
+
"loss": 0.3519,
|
| 604 |
+
"step": 850
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 1.376,
|
| 608 |
+
"grad_norm": 1.7154734289625684,
|
| 609 |
+
"learning_rate": 5.6474394110492344e-05,
|
| 610 |
+
"loss": 0.3732,
|
| 611 |
+
"step": 860
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 1.392,
|
| 615 |
+
"grad_norm": 1.3529634625288804,
|
| 616 |
+
"learning_rate": 5.564281924367408e-05,
|
| 617 |
+
"loss": 0.3954,
|
| 618 |
+
"step": 870
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 1.408,
|
| 622 |
+
"grad_norm": 1.4693831158149995,
|
| 623 |
+
"learning_rate": 5.480966027471889e-05,
|
| 624 |
+
"loss": 0.3669,
|
| 625 |
+
"step": 880
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 1.424,
|
| 629 |
+
"grad_norm": 1.5051265025599196,
|
| 630 |
+
"learning_rate": 5.3975151095382995e-05,
|
| 631 |
+
"loss": 0.3994,
|
| 632 |
+
"step": 890
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 1.44,
|
| 636 |
+
"grad_norm": 1.2893604881333178,
|
| 637 |
+
"learning_rate": 5.313952597646568e-05,
|
| 638 |
+
"loss": 0.3904,
|
| 639 |
+
"step": 900
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 1.456,
|
| 643 |
+
"grad_norm": 1.5665270237662434,
|
| 644 |
+
"learning_rate": 5.230301950204262e-05,
|
| 645 |
+
"loss": 0.3821,
|
| 646 |
+
"step": 910
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 1.472,
|
| 650 |
+
"grad_norm": 1.1847281856196625,
|
| 651 |
+
"learning_rate": 5.1465866503611426e-05,
|
| 652 |
+
"loss": 0.3713,
|
| 653 |
+
"step": 920
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 1.488,
|
| 657 |
+
"grad_norm": 1.6027343131881702,
|
| 658 |
+
"learning_rate": 5.062830199416764e-05,
|
| 659 |
+
"loss": 0.3785,
|
| 660 |
+
"step": 930
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 1.504,
|
| 664 |
+
"grad_norm": 1.1745245908384976,
|
| 665 |
+
"learning_rate": 4.979056110222981e-05,
|
| 666 |
+
"loss": 0.3851,
|
| 667 |
+
"step": 940
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 1.52,
|
| 671 |
+
"grad_norm": 1.2252730329400867,
|
| 672 |
+
"learning_rate": 4.895287900583216e-05,
|
| 673 |
+
"loss": 0.3991,
|
| 674 |
+
"step": 950
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 1.536,
|
| 678 |
+
"grad_norm": 1.9189789923421308,
|
| 679 |
+
"learning_rate": 4.811549086650327e-05,
|
| 680 |
+
"loss": 0.3726,
|
| 681 |
+
"step": 960
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 1.552,
|
| 685 |
+
"grad_norm": 1.9319344659060065,
|
| 686 |
+
"learning_rate": 4.7278631763249554e-05,
|
| 687 |
+
"loss": 0.3703,
|
| 688 |
+
"step": 970
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 1.568,
|
| 692 |
+
"grad_norm": 1.3155775136973578,
|
| 693 |
+
"learning_rate": 4.6442536626561675e-05,
|
| 694 |
+
"loss": 0.3462,
|
| 695 |
+
"step": 980
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 1.584,
|
| 699 |
+
"grad_norm": 1.3059644425743502,
|
| 700 |
+
"learning_rate": 4.560744017246284e-05,
|
| 701 |
+
"loss": 0.3486,
|
| 702 |
+
"step": 990
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 1.6,
|
| 706 |
+
"grad_norm": 1.1918137712202796,
|
| 707 |
+
"learning_rate": 4.477357683661734e-05,
|
| 708 |
+
"loss": 0.3384,
|
| 709 |
+
"step": 1000
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 1.616,
|
| 713 |
+
"grad_norm": 1.3051153832513638,
|
| 714 |
+
"learning_rate": 4.394118070851749e-05,
|
| 715 |
+
"loss": 0.3976,
|
| 716 |
+
"step": 1010
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 1.6320000000000001,
|
| 720 |
+
"grad_norm": 2.1135344091319266,
|
| 721 |
+
"learning_rate": 4.31104854657681e-05,
|
| 722 |
+
"loss": 0.3607,
|
| 723 |
+
"step": 1020
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 1.6480000000000001,
|
| 727 |
+
"grad_norm": 1.3948892979651148,
|
| 728 |
+
"learning_rate": 4.228172430848644e-05,
|
| 729 |
+
"loss": 0.371,
|
| 730 |
+
"step": 1030
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 1.6640000000000001,
|
| 734 |
+
"grad_norm": 1.3934093132103698,
|
| 735 |
+
"learning_rate": 4.1455129893836174e-05,
|
| 736 |
+
"loss": 0.3627,
|
| 737 |
+
"step": 1040
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 1.6800000000000002,
|
| 741 |
+
"grad_norm": 2.099040035068272,
|
| 742 |
+
"learning_rate": 4.063093427071376e-05,
|
| 743 |
+
"loss": 0.3725,
|
| 744 |
+
"step": 1050
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 1.696,
|
| 748 |
+
"grad_norm": 1.1171081747364493,
|
| 749 |
+
"learning_rate": 3.9809368814605766e-05,
|
| 750 |
+
"loss": 0.3466,
|
| 751 |
+
"step": 1060
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 1.712,
|
| 755 |
+
"grad_norm": 1.2557303206572503,
|
| 756 |
+
"learning_rate": 3.899066416263493e-05,
|
| 757 |
+
"loss": 0.3629,
|
| 758 |
+
"step": 1070
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 1.728,
|
| 762 |
+
"grad_norm": 1.2612524984380895,
|
| 763 |
+
"learning_rate": 3.817505014881378e-05,
|
| 764 |
+
"loss": 0.3941,
|
| 765 |
+
"step": 1080
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 1.744,
|
| 769 |
+
"grad_norm": 1.4536732919602855,
|
| 770 |
+
"learning_rate": 3.736275573952354e-05,
|
| 771 |
+
"loss": 0.3362,
|
| 772 |
+
"step": 1090
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 1.76,
|
| 776 |
+
"grad_norm": 1.5347325608557592,
|
| 777 |
+
"learning_rate": 3.655400896923672e-05,
|
| 778 |
+
"loss": 0.3197,
|
| 779 |
+
"step": 1100
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 1.776,
|
| 783 |
+
"grad_norm": 1.5306810075691544,
|
| 784 |
+
"learning_rate": 3.5749036876501194e-05,
|
| 785 |
+
"loss": 0.3447,
|
| 786 |
+
"step": 1110
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 1.792,
|
| 790 |
+
"grad_norm": 1.2300188758500825,
|
| 791 |
+
"learning_rate": 3.494806544020398e-05,
|
| 792 |
+
"loss": 0.3535,
|
| 793 |
+
"step": 1120
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 1.808,
|
| 797 |
+
"grad_norm": 1.4275688891160825,
|
| 798 |
+
"learning_rate": 3.4151319516132416e-05,
|
| 799 |
+
"loss": 0.3447,
|
| 800 |
+
"step": 1130
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 1.8239999999999998,
|
| 804 |
+
"grad_norm": 1.397634927032295,
|
| 805 |
+
"learning_rate": 3.335902277385067e-05,
|
| 806 |
+
"loss": 0.338,
|
| 807 |
+
"step": 1140
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 1.8399999999999999,
|
| 811 |
+
"grad_norm": 1.8570000554042922,
|
| 812 |
+
"learning_rate": 3.257139763390925e-05,
|
| 813 |
+
"loss": 0.3915,
|
| 814 |
+
"step": 1150
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 1.8559999999999999,
|
| 818 |
+
"grad_norm": 1.0694824076607796,
|
| 819 |
+
"learning_rate": 3.178866520540509e-05,
|
| 820 |
+
"loss": 0.332,
|
| 821 |
+
"step": 1160
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 1.8719999999999999,
|
| 825 |
+
"grad_norm": 1.9227266591369816,
|
| 826 |
+
"learning_rate": 3.101104522390995e-05,
|
| 827 |
+
"loss": 0.3734,
|
| 828 |
+
"step": 1170
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 1.888,
|
| 832 |
+
"grad_norm": 1.4206748536497713,
|
| 833 |
+
"learning_rate": 3.023875598978419e-05,
|
| 834 |
+
"loss": 0.3409,
|
| 835 |
+
"step": 1180
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 1.904,
|
| 839 |
+
"grad_norm": 1.4359353896901201,
|
| 840 |
+
"learning_rate": 2.9472014306893603e-05,
|
| 841 |
+
"loss": 0.3333,
|
| 842 |
+
"step": 1190
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 1.92,
|
| 846 |
+
"grad_norm": 1.2517700720828175,
|
| 847 |
+
"learning_rate": 2.8711035421746367e-05,
|
| 848 |
+
"loss": 0.369,
|
| 849 |
+
"step": 1200
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 1.936,
|
| 853 |
+
"grad_norm": 1.3801398040671706,
|
| 854 |
+
"learning_rate": 2.795603296306708e-05,
|
| 855 |
+
"loss": 0.3449,
|
| 856 |
+
"step": 1210
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 1.952,
|
| 860 |
+
"grad_norm": 1.8786886188992349,
|
| 861 |
+
"learning_rate": 2.7207218881825014e-05,
|
| 862 |
+
"loss": 0.3662,
|
| 863 |
+
"step": 1220
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 1.968,
|
| 867 |
+
"grad_norm": 1.2874259666120602,
|
| 868 |
+
"learning_rate": 2.6464803391733374e-05,
|
| 869 |
+
"loss": 0.3773,
|
| 870 |
+
"step": 1230
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 1.984,
|
| 874 |
+
"grad_norm": 1.1788438157215508,
|
| 875 |
+
"learning_rate": 2.5728994910236304e-05,
|
| 876 |
+
"loss": 0.3245,
|
| 877 |
+
"step": 1240
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 2.0,
|
| 881 |
+
"grad_norm": 1.5631045640027712,
|
| 882 |
+
"learning_rate": 2.500000000000001e-05,
|
| 883 |
+
"loss": 0.3412,
|
| 884 |
+
"step": 1250
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 2.016,
|
| 888 |
+
"grad_norm": 1.4435057320180134,
|
| 889 |
+
"learning_rate": 2.4278023310924673e-05,
|
| 890 |
+
"loss": 0.3027,
|
| 891 |
+
"step": 1260
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 2.032,
|
| 895 |
+
"grad_norm": 1.690073924846414,
|
| 896 |
+
"learning_rate": 2.3563267522693415e-05,
|
| 897 |
+
"loss": 0.2745,
|
| 898 |
+
"step": 1270
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 2.048,
|
| 902 |
+
"grad_norm": 1.571802508559373,
|
| 903 |
+
"learning_rate": 2.2855933287874138e-05,
|
| 904 |
+
"loss": 0.2686,
|
| 905 |
+
"step": 1280
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 2.064,
|
| 909 |
+
"grad_norm": 1.2829706151133036,
|
| 910 |
+
"learning_rate": 2.215621917559062e-05,
|
| 911 |
+
"loss": 0.2628,
|
| 912 |
+
"step": 1290
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 2.08,
|
| 916 |
+
"grad_norm": 1.5440500775973827,
|
| 917 |
+
"learning_rate": 2.1464321615778422e-05,
|
| 918 |
+
"loss": 0.274,
|
| 919 |
+
"step": 1300
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 2.096,
|
| 923 |
+
"grad_norm": 2.0539984960007627,
|
| 924 |
+
"learning_rate": 2.07804348440414e-05,
|
| 925 |
+
"loss": 0.2711,
|
| 926 |
+
"step": 1310
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 2.112,
|
| 930 |
+
"grad_norm": 1.6138729128750948,
|
| 931 |
+
"learning_rate": 2.0104750847124075e-05,
|
| 932 |
+
"loss": 0.2662,
|
| 933 |
+
"step": 1320
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 2.128,
|
| 937 |
+
"grad_norm": 1.4560623181879653,
|
| 938 |
+
"learning_rate": 1.9437459309015427e-05,
|
| 939 |
+
"loss": 0.2722,
|
| 940 |
+
"step": 1330
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 2.144,
|
| 944 |
+
"grad_norm": 1.4659611175991962,
|
| 945 |
+
"learning_rate": 1.8778747557699224e-05,
|
| 946 |
+
"loss": 0.2965,
|
| 947 |
+
"step": 1340
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 2.16,
|
| 951 |
+
"grad_norm": 1.375382006257737,
|
| 952 |
+
"learning_rate": 1.8128800512565513e-05,
|
| 953 |
+
"loss": 0.2789,
|
| 954 |
+
"step": 1350
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 2.176,
|
| 958 |
+
"grad_norm": 1.599179489252925,
|
| 959 |
+
"learning_rate": 1.7487800632498545e-05,
|
| 960 |
+
"loss": 0.2727,
|
| 961 |
+
"step": 1360
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 2.192,
|
| 965 |
+
"grad_norm": 1.517535594054785,
|
| 966 |
+
"learning_rate": 1.685592786465524e-05,
|
| 967 |
+
"loss": 0.2579,
|
| 968 |
+
"step": 1370
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 2.208,
|
| 972 |
+
"grad_norm": 2.1592276480492414,
|
| 973 |
+
"learning_rate": 1.6233359593948777e-05,
|
| 974 |
+
"loss": 0.2639,
|
| 975 |
+
"step": 1380
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 2.224,
|
| 979 |
+
"grad_norm": 1.7544518130502498,
|
| 980 |
+
"learning_rate": 1.5620270593251635e-05,
|
| 981 |
+
"loss": 0.2909,
|
| 982 |
+
"step": 1390
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 2.24,
|
| 986 |
+
"grad_norm": 1.44885168688834,
|
| 987 |
+
"learning_rate": 1.5016832974331724e-05,
|
| 988 |
+
"loss": 0.2675,
|
| 989 |
+
"step": 1400
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 2.2560000000000002,
|
| 993 |
+
"grad_norm": 1.6827124662772142,
|
| 994 |
+
"learning_rate": 1.4423216139535734e-05,
|
| 995 |
+
"loss": 0.2501,
|
| 996 |
+
"step": 1410
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 2.2720000000000002,
|
| 1000 |
+
"grad_norm": 1.8488651881141411,
|
| 1001 |
+
"learning_rate": 1.3839586734232906e-05,
|
| 1002 |
+
"loss": 0.28,
|
| 1003 |
+
"step": 1420
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 2.288,
|
| 1007 |
+
"grad_norm": 1.8044696205561654,
|
| 1008 |
+
"learning_rate": 1.3266108600032929e-05,
|
| 1009 |
+
"loss": 0.2946,
|
| 1010 |
+
"step": 1430
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 2.304,
|
| 1014 |
+
"grad_norm": 1.5680732030217428,
|
| 1015 |
+
"learning_rate": 1.2702942728790895e-05,
|
| 1016 |
+
"loss": 0.2801,
|
| 1017 |
+
"step": 1440
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 2.32,
|
| 1021 |
+
"grad_norm": 1.8180785380951152,
|
| 1022 |
+
"learning_rate": 1.2150247217412186e-05,
|
| 1023 |
+
"loss": 0.2686,
|
| 1024 |
+
"step": 1450
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 2.336,
|
| 1028 |
+
"grad_norm": 1.4756615703317884,
|
| 1029 |
+
"learning_rate": 1.160817722347014e-05,
|
| 1030 |
+
"loss": 0.2623,
|
| 1031 |
+
"step": 1460
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 2.352,
|
| 1035 |
+
"grad_norm": 1.4951599689749675,
|
| 1036 |
+
"learning_rate": 1.1076884921648834e-05,
|
| 1037 |
+
"loss": 0.2865,
|
| 1038 |
+
"step": 1470
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 2.368,
|
| 1042 |
+
"grad_norm": 1.8073936077115729,
|
| 1043 |
+
"learning_rate": 1.0556519461023301e-05,
|
| 1044 |
+
"loss": 0.2677,
|
| 1045 |
+
"step": 1480
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 2.384,
|
| 1049 |
+
"grad_norm": 1.6787640384880986,
|
| 1050 |
+
"learning_rate": 1.0047226923189024e-05,
|
| 1051 |
+
"loss": 0.2821,
|
| 1052 |
+
"step": 1490
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 2.4,
|
| 1056 |
+
"grad_norm": 1.7736081937737844,
|
| 1057 |
+
"learning_rate": 9.549150281252633e-06,
|
| 1058 |
+
"loss": 0.2628,
|
| 1059 |
+
"step": 1500
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 2.416,
|
| 1063 |
+
"grad_norm": 1.8163690827077064,
|
| 1064 |
+
"learning_rate": 9.06242935969528e-06,
|
| 1065 |
+
"loss": 0.2648,
|
| 1066 |
+
"step": 1510
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 2.432,
|
| 1070 |
+
"grad_norm": 1.914925459991879,
|
| 1071 |
+
"learning_rate": 8.587200795119793e-06,
|
| 1072 |
+
"loss": 0.2718,
|
| 1073 |
+
"step": 1520
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 2.448,
|
| 1077 |
+
"grad_norm": 1.557572030989793,
|
| 1078 |
+
"learning_rate": 8.123597997892918e-06,
|
| 1079 |
+
"loss": 0.2779,
|
| 1080 |
+
"step": 1530
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 2.464,
|
| 1084 |
+
"grad_norm": 2.4189475486471412,
|
| 1085 |
+
"learning_rate": 7.671751114693104e-06,
|
| 1086 |
+
"loss": 0.2522,
|
| 1087 |
+
"step": 1540
|
| 1088 |
+
},
|
| 1089 |
+
{
|
| 1090 |
+
"epoch": 2.48,
|
| 1091 |
+
"grad_norm": 1.5979978703748492,
|
| 1092 |
+
"learning_rate": 7.2317869919746705e-06,
|
| 1093 |
+
"loss": 0.2668,
|
| 1094 |
+
"step": 1550
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 2.496,
|
| 1098 |
+
"grad_norm": 1.5867221943137244,
|
| 1099 |
+
"learning_rate": 6.803829140358237e-06,
|
| 1100 |
+
"loss": 0.2582,
|
| 1101 |
+
"step": 1560
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 2.512,
|
| 1105 |
+
"grad_norm": 2.0088219455242866,
|
| 1106 |
+
"learning_rate": 6.3879976999578154e-06,
|
| 1107 |
+
"loss": 0.2509,
|
| 1108 |
+
"step": 1570
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"epoch": 2.528,
|
| 1112 |
+
"grad_norm": 1.7960925687722837,
|
| 1113 |
+
"learning_rate": 5.98440940665399e-06,
|
| 1114 |
+
"loss": 0.2743,
|
| 1115 |
+
"step": 1580
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 2.544,
|
| 1119 |
+
"grad_norm": 1.5572986827037758,
|
| 1120 |
+
"learning_rate": 5.593177559322777e-06,
|
| 1121 |
+
"loss": 0.2833,
|
| 1122 |
+
"step": 1590
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 2.56,
|
| 1126 |
+
"grad_norm": 1.370850993490327,
|
| 1127 |
+
"learning_rate": 5.214411988029355e-06,
|
| 1128 |
+
"loss": 0.2693,
|
| 1129 |
+
"step": 1600
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 2.576,
|
| 1133 |
+
"grad_norm": 1.3820976570272918,
|
| 1134 |
+
"learning_rate": 4.848219023195644e-06,
|
| 1135 |
+
"loss": 0.2607,
|
| 1136 |
+
"step": 1610
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 2.592,
|
| 1140 |
+
"grad_norm": 1.7956675550522907,
|
| 1141 |
+
"learning_rate": 4.494701465750217e-06,
|
| 1142 |
+
"loss": 0.2632,
|
| 1143 |
+
"step": 1620
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 2.608,
|
| 1147 |
+
"grad_norm": 1.657852197945445,
|
| 1148 |
+
"learning_rate": 4.153958558269189e-06,
|
| 1149 |
+
"loss": 0.2573,
|
| 1150 |
+
"step": 1630
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 2.624,
|
| 1154 |
+
"grad_norm": 1.6754996668537725,
|
| 1155 |
+
"learning_rate": 3.826085957115888e-06,
|
| 1156 |
+
"loss": 0.2411,
|
| 1157 |
+
"step": 1640
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 2.64,
|
| 1161 |
+
"grad_norm": 1.3160903872279537,
|
| 1162 |
+
"learning_rate": 3.511175705587433e-06,
|
| 1163 |
+
"loss": 0.2601,
|
| 1164 |
+
"step": 1650
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 2.656,
|
| 1168 |
+
"grad_norm": 1.4417765456611342,
|
| 1169 |
+
"learning_rate": 3.2093162080754637e-06,
|
| 1170 |
+
"loss": 0.2832,
|
| 1171 |
+
"step": 1660
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 2.672,
|
| 1175 |
+
"grad_norm": 1.5418721821515773,
|
| 1176 |
+
"learning_rate": 2.9205922052484958e-06,
|
| 1177 |
+
"loss": 0.2725,
|
| 1178 |
+
"step": 1670
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 2.6879999999999997,
|
| 1182 |
+
"grad_norm": 1.3747599439593585,
|
| 1183 |
+
"learning_rate": 2.6450847502627884e-06,
|
| 1184 |
+
"loss": 0.2654,
|
| 1185 |
+
"step": 1680
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 2.7039999999999997,
|
| 1189 |
+
"grad_norm": 1.528133223907777,
|
| 1190 |
+
"learning_rate": 2.3828711860083674e-06,
|
| 1191 |
+
"loss": 0.2784,
|
| 1192 |
+
"step": 1690
|
| 1193 |
+
},
|
| 1194 |
+
{
|
| 1195 |
+
"epoch": 2.7199999999999998,
|
| 1196 |
+
"grad_norm": 2.3538184370373534,
|
| 1197 |
+
"learning_rate": 2.134025123396638e-06,
|
| 1198 |
+
"loss": 0.2758,
|
| 1199 |
+
"step": 1700
|
| 1200 |
+
},
|
| 1201 |
+
{
|
| 1202 |
+
"epoch": 2.7359999999999998,
|
| 1203 |
+
"grad_norm": 1.6198355766213741,
|
| 1204 |
+
"learning_rate": 1.8986164206957035e-06,
|
| 1205 |
+
"loss": 0.2533,
|
| 1206 |
+
"step": 1710
|
| 1207 |
+
},
|
| 1208 |
+
{
|
| 1209 |
+
"epoch": 2.752,
|
| 1210 |
+
"grad_norm": 1.8976100233670337,
|
| 1211 |
+
"learning_rate": 1.6767111639191202e-06,
|
| 1212 |
+
"loss": 0.268,
|
| 1213 |
+
"step": 1720
|
| 1214 |
+
},
|
| 1215 |
+
{
|
| 1216 |
+
"epoch": 2.768,
|
| 1217 |
+
"grad_norm": 1.7822120577723535,
|
| 1218 |
+
"learning_rate": 1.4683716482736366e-06,
|
| 1219 |
+
"loss": 0.28,
|
| 1220 |
+
"step": 1730
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 2.784,
|
| 1224 |
+
"grad_norm": 2.2850018497482183,
|
| 1225 |
+
"learning_rate": 1.2736563606711382e-06,
|
| 1226 |
+
"loss": 0.2797,
|
| 1227 |
+
"step": 1740
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"epoch": 2.8,
|
| 1231 |
+
"grad_norm": 1.79121827664844,
|
| 1232 |
+
"learning_rate": 1.0926199633097157e-06,
|
| 1233 |
+
"loss": 0.2831,
|
| 1234 |
+
"step": 1750
|
| 1235 |
+
},
|
| 1236 |
+
{
|
| 1237 |
+
"epoch": 2.816,
|
| 1238 |
+
"grad_norm": 1.7925072153699964,
|
| 1239 |
+
"learning_rate": 9.253132783283547e-07,
|
| 1240 |
+
"loss": 0.2558,
|
| 1241 |
+
"step": 1760
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 2.832,
|
| 1245 |
+
"grad_norm": 1.2894709082534288,
|
| 1246 |
+
"learning_rate": 7.717832735397335e-07,
|
| 1247 |
+
"loss": 0.2635,
|
| 1248 |
+
"step": 1770
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 2.848,
|
| 1252 |
+
"grad_norm": 1.4531115959909715,
|
| 1253 |
+
"learning_rate": 6.3207304924498e-07,
|
| 1254 |
+
"loss": 0.2639,
|
| 1255 |
+
"step": 1780
|
| 1256 |
+
},
|
| 1257 |
+
{
|
| 1258 |
+
"epoch": 2.864,
|
| 1259 |
+
"grad_norm": 1.9570336366099832,
|
| 1260 |
+
"learning_rate": 5.062218261342122e-07,
|
| 1261 |
+
"loss": 0.2532,
|
| 1262 |
+
"step": 1790
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 2.88,
|
| 1266 |
+
"grad_norm": 1.3421965286430086,
|
| 1267 |
+
"learning_rate": 3.9426493427611177e-07,
|
| 1268 |
+
"loss": 0.2327,
|
| 1269 |
+
"step": 1800
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 2.896,
|
| 1273 |
+
"grad_norm": 2.3252685172502465,
|
| 1274 |
+
"learning_rate": 2.962338031997691e-07,
|
| 1275 |
+
"loss": 0.2621,
|
| 1276 |
+
"step": 1810
|
| 1277 |
+
},
|
| 1278 |
+
{
|
| 1279 |
+
"epoch": 2.912,
|
| 1280 |
+
"grad_norm": 1.7539361758237508,
|
| 1281 |
+
"learning_rate": 2.1215595307154667e-07,
|
| 1282 |
+
"loss": 0.256,
|
| 1283 |
+
"step": 1820
|
| 1284 |
+
},
|
| 1285 |
+
{
|
| 1286 |
+
"epoch": 2.928,
|
| 1287 |
+
"grad_norm": 1.417074579458341,
|
| 1288 |
+
"learning_rate": 1.420549869693033e-07,
|
| 1289 |
+
"loss": 0.2545,
|
| 1290 |
+
"step": 1830
|
| 1291 |
+
},
|
| 1292 |
+
{
|
| 1293 |
+
"epoch": 2.944,
|
| 1294 |
+
"grad_norm": 1.7016791883476743,
|
| 1295 |
+
"learning_rate": 8.595058425640013e-08,
|
| 1296 |
+
"loss": 0.2596,
|
| 1297 |
+
"step": 1840
|
| 1298 |
+
},
|
| 1299 |
+
{
|
| 1300 |
+
"epoch": 2.96,
|
| 1301 |
+
"grad_norm": 1.6335011497462837,
|
| 1302 |
+
"learning_rate": 4.385849505708084e-08,
|
| 1303 |
+
"loss": 0.2584,
|
| 1304 |
+
"step": 1850
|
| 1305 |
+
},
|
| 1306 |
+
{
|
| 1307 |
+
"epoch": 2.976,
|
| 1308 |
+
"grad_norm": 1.7816152560274576,
|
| 1309 |
+
"learning_rate": 1.5790535835003008e-08,
|
| 1310 |
+
"loss": 0.2752,
|
| 1311 |
+
"step": 1860
|
| 1312 |
+
},
|
| 1313 |
+
{
|
| 1314 |
+
"epoch": 2.992,
|
| 1315 |
+
"grad_norm": 1.579627845758394,
|
| 1316 |
+
"learning_rate": 1.7545860759693445e-09,
|
| 1317 |
+
"loss": 0.2778,
|
| 1318 |
+
"step": 1870
|
| 1319 |
+
},
|
| 1320 |
+
{
|
| 1321 |
+
"epoch": 3.0,
|
| 1322 |
+
"step": 1875,
|
| 1323 |
+
"total_flos": 299543991877632.0,
|
| 1324 |
+
"train_loss": 0.3980048195521037,
|
| 1325 |
+
"train_runtime": 15877.4798,
|
| 1326 |
+
"train_samples_per_second": 0.945,
|
| 1327 |
+
"train_steps_per_second": 0.118
|
| 1328 |
+
}
|
| 1329 |
+
],
|
| 1330 |
+
"logging_steps": 10,
|
| 1331 |
+
"max_steps": 1875,
|
| 1332 |
+
"num_input_tokens_seen": 0,
|
| 1333 |
+
"num_train_epochs": 3,
|
| 1334 |
+
"save_steps": 500,
|
| 1335 |
+
"stateful_callbacks": {
|
| 1336 |
+
"TrainerControl": {
|
| 1337 |
+
"args": {
|
| 1338 |
+
"should_epoch_stop": false,
|
| 1339 |
+
"should_evaluate": false,
|
| 1340 |
+
"should_log": false,
|
| 1341 |
+
"should_save": true,
|
| 1342 |
+
"should_training_stop": true
|
| 1343 |
+
},
|
| 1344 |
+
"attributes": {}
|
| 1345 |
+
}
|
| 1346 |
+
},
|
| 1347 |
+
"total_flos": 299543991877632.0,
|
| 1348 |
+
"train_batch_size": 1,
|
| 1349 |
+
"trial_name": null,
|
| 1350 |
+
"trial_params": null
|
| 1351 |
+
}
|
training_loss.png
ADDED
|