End of training
Browse files- README.md +2 -1
- all_results.json +8 -0
- train_results.json +8 -0
- trainer_state.json +1680 -0
- training_loss.png +0 -0
README.md
CHANGED
|
@@ -4,6 +4,7 @@ license: apache-2.0
|
|
| 4 |
base_model: Qwen/Qwen2.5-7B-Instruct
|
| 5 |
tags:
|
| 6 |
- llama-factory
|
|
|
|
| 7 |
- generated_from_trainer
|
| 8 |
model-index:
|
| 9 |
- name: multiple_samples_none_numina_aime_adjusted_samples
|
|
@@ -15,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 15 |
|
| 16 |
# multiple_samples_none_numina_aime_adjusted_samples
|
| 17 |
|
| 18 |
-
This model is a fine-tuned version of [Qwen/Qwen2.5-7B-Instruct](https://huggingface.co/Qwen/Qwen2.5-7B-Instruct) on
|
| 19 |
|
| 20 |
## Model description
|
| 21 |
|
|
|
|
| 4 |
base_model: Qwen/Qwen2.5-7B-Instruct
|
| 5 |
tags:
|
| 6 |
- llama-factory
|
| 7 |
+
- full
|
| 8 |
- generated_from_trainer
|
| 9 |
model-index:
|
| 10 |
- name: multiple_samples_none_numina_aime_adjusted_samples
|
|
|
|
| 16 |
|
| 17 |
# multiple_samples_none_numina_aime_adjusted_samples
|
| 18 |
|
| 19 |
+
This model is a fine-tuned version of [Qwen/Qwen2.5-7B-Instruct](https://huggingface.co/Qwen/Qwen2.5-7B-Instruct) on the mlfoundations-dev/multiple_samples_none_numina_aime_adjusted_samples dataset.
|
| 20 |
|
| 21 |
## Model description
|
| 22 |
|
all_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 2.9936034115138592,
|
| 3 |
+
"total_flos": 327743678562304.0,
|
| 4 |
+
"train_loss": 0.7533083028263516,
|
| 5 |
+
"train_runtime": 9147.0121,
|
| 6 |
+
"train_samples_per_second": 2.46,
|
| 7 |
+
"train_steps_per_second": 0.026
|
| 8 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 2.9936034115138592,
|
| 3 |
+
"total_flos": 327743678562304.0,
|
| 4 |
+
"train_loss": 0.7533083028263516,
|
| 5 |
+
"train_runtime": 9147.0121,
|
| 6 |
+
"train_samples_per_second": 2.46,
|
| 7 |
+
"train_steps_per_second": 0.026
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1680 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 2.9936034115138592,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 234,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.01279317697228145,
|
| 13 |
+
"grad_norm": 6.832852840423584,
|
| 14 |
+
"learning_rate": 4.1666666666666667e-07,
|
| 15 |
+
"loss": 1.0868,
|
| 16 |
+
"step": 1
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.0255863539445629,
|
| 20 |
+
"grad_norm": 6.904089450836182,
|
| 21 |
+
"learning_rate": 8.333333333333333e-07,
|
| 22 |
+
"loss": 1.1044,
|
| 23 |
+
"step": 2
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.03837953091684435,
|
| 27 |
+
"grad_norm": 6.991506576538086,
|
| 28 |
+
"learning_rate": 1.25e-06,
|
| 29 |
+
"loss": 1.1401,
|
| 30 |
+
"step": 3
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.0511727078891258,
|
| 34 |
+
"grad_norm": 6.811994552612305,
|
| 35 |
+
"learning_rate": 1.6666666666666667e-06,
|
| 36 |
+
"loss": 1.1168,
|
| 37 |
+
"step": 4
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.06396588486140725,
|
| 41 |
+
"grad_norm": 6.493495464324951,
|
| 42 |
+
"learning_rate": 2.0833333333333334e-06,
|
| 43 |
+
"loss": 1.0934,
|
| 44 |
+
"step": 5
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.0767590618336887,
|
| 48 |
+
"grad_norm": 5.279635429382324,
|
| 49 |
+
"learning_rate": 2.5e-06,
|
| 50 |
+
"loss": 1.081,
|
| 51 |
+
"step": 6
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.08955223880597014,
|
| 55 |
+
"grad_norm": 4.791995525360107,
|
| 56 |
+
"learning_rate": 2.916666666666667e-06,
|
| 57 |
+
"loss": 1.0356,
|
| 58 |
+
"step": 7
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.1023454157782516,
|
| 62 |
+
"grad_norm": 2.9714953899383545,
|
| 63 |
+
"learning_rate": 3.3333333333333333e-06,
|
| 64 |
+
"loss": 0.9912,
|
| 65 |
+
"step": 8
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.11513859275053305,
|
| 69 |
+
"grad_norm": 2.6682045459747314,
|
| 70 |
+
"learning_rate": 3.7500000000000005e-06,
|
| 71 |
+
"loss": 0.947,
|
| 72 |
+
"step": 9
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.1279317697228145,
|
| 76 |
+
"grad_norm": 2.5980541706085205,
|
| 77 |
+
"learning_rate": 4.166666666666667e-06,
|
| 78 |
+
"loss": 0.9328,
|
| 79 |
+
"step": 10
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.14072494669509594,
|
| 83 |
+
"grad_norm": 4.253756999969482,
|
| 84 |
+
"learning_rate": 4.583333333333333e-06,
|
| 85 |
+
"loss": 0.9871,
|
| 86 |
+
"step": 11
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.1535181236673774,
|
| 90 |
+
"grad_norm": 4.215671062469482,
|
| 91 |
+
"learning_rate": 5e-06,
|
| 92 |
+
"loss": 0.9252,
|
| 93 |
+
"step": 12
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.16631130063965885,
|
| 97 |
+
"grad_norm": 4.390238285064697,
|
| 98 |
+
"learning_rate": 5.416666666666667e-06,
|
| 99 |
+
"loss": 0.9241,
|
| 100 |
+
"step": 13
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.1791044776119403,
|
| 104 |
+
"grad_norm": 3.742227077484131,
|
| 105 |
+
"learning_rate": 5.833333333333334e-06,
|
| 106 |
+
"loss": 0.9443,
|
| 107 |
+
"step": 14
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.19189765458422176,
|
| 111 |
+
"grad_norm": 3.979121446609497,
|
| 112 |
+
"learning_rate": 6.25e-06,
|
| 113 |
+
"loss": 0.946,
|
| 114 |
+
"step": 15
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.2046908315565032,
|
| 118 |
+
"grad_norm": 2.8892107009887695,
|
| 119 |
+
"learning_rate": 6.666666666666667e-06,
|
| 120 |
+
"loss": 0.8909,
|
| 121 |
+
"step": 16
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.21748400852878466,
|
| 125 |
+
"grad_norm": 2.102924346923828,
|
| 126 |
+
"learning_rate": 7.083333333333335e-06,
|
| 127 |
+
"loss": 0.8488,
|
| 128 |
+
"step": 17
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.2302771855010661,
|
| 132 |
+
"grad_norm": 2.0470516681671143,
|
| 133 |
+
"learning_rate": 7.500000000000001e-06,
|
| 134 |
+
"loss": 0.9146,
|
| 135 |
+
"step": 18
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.24307036247334754,
|
| 139 |
+
"grad_norm": 1.9300282001495361,
|
| 140 |
+
"learning_rate": 7.916666666666667e-06,
|
| 141 |
+
"loss": 0.8769,
|
| 142 |
+
"step": 19
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.255863539445629,
|
| 146 |
+
"grad_norm": 1.6612228155136108,
|
| 147 |
+
"learning_rate": 8.333333333333334e-06,
|
| 148 |
+
"loss": 0.8631,
|
| 149 |
+
"step": 20
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.26865671641791045,
|
| 153 |
+
"grad_norm": 1.4431664943695068,
|
| 154 |
+
"learning_rate": 8.750000000000001e-06,
|
| 155 |
+
"loss": 0.8269,
|
| 156 |
+
"step": 21
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.2814498933901919,
|
| 160 |
+
"grad_norm": 1.2628264427185059,
|
| 161 |
+
"learning_rate": 9.166666666666666e-06,
|
| 162 |
+
"loss": 0.8353,
|
| 163 |
+
"step": 22
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.2942430703624733,
|
| 167 |
+
"grad_norm": 1.1078248023986816,
|
| 168 |
+
"learning_rate": 9.583333333333335e-06,
|
| 169 |
+
"loss": 0.8076,
|
| 170 |
+
"step": 23
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.3070362473347548,
|
| 174 |
+
"grad_norm": 1.111648440361023,
|
| 175 |
+
"learning_rate": 1e-05,
|
| 176 |
+
"loss": 0.9005,
|
| 177 |
+
"step": 24
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.31982942430703626,
|
| 181 |
+
"grad_norm": 0.8681544065475464,
|
| 182 |
+
"learning_rate": 9.999440509051367e-06,
|
| 183 |
+
"loss": 0.8255,
|
| 184 |
+
"step": 25
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.3326226012793177,
|
| 188 |
+
"grad_norm": 1.1207367181777954,
|
| 189 |
+
"learning_rate": 9.997762161417517e-06,
|
| 190 |
+
"loss": 0.8363,
|
| 191 |
+
"step": 26
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.34541577825159914,
|
| 195 |
+
"grad_norm": 0.9773739576339722,
|
| 196 |
+
"learning_rate": 9.994965332706574e-06,
|
| 197 |
+
"loss": 0.7949,
|
| 198 |
+
"step": 27
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.3582089552238806,
|
| 202 |
+
"grad_norm": 0.7835165858268738,
|
| 203 |
+
"learning_rate": 9.991050648838676e-06,
|
| 204 |
+
"loss": 0.7652,
|
| 205 |
+
"step": 28
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.37100213219616207,
|
| 209 |
+
"grad_norm": 1.0131946802139282,
|
| 210 |
+
"learning_rate": 9.986018985905901e-06,
|
| 211 |
+
"loss": 0.8445,
|
| 212 |
+
"step": 29
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.3837953091684435,
|
| 216 |
+
"grad_norm": 0.8833023905754089,
|
| 217 |
+
"learning_rate": 9.979871469976197e-06,
|
| 218 |
+
"loss": 0.7808,
|
| 219 |
+
"step": 30
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.39658848614072495,
|
| 223 |
+
"grad_norm": 0.8186467289924622,
|
| 224 |
+
"learning_rate": 9.972609476841368e-06,
|
| 225 |
+
"loss": 0.8133,
|
| 226 |
+
"step": 31
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.4093816631130064,
|
| 230 |
+
"grad_norm": 0.8074617385864258,
|
| 231 |
+
"learning_rate": 9.964234631709188e-06,
|
| 232 |
+
"loss": 0.7957,
|
| 233 |
+
"step": 32
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.42217484008528783,
|
| 237 |
+
"grad_norm": 0.8177934288978577,
|
| 238 |
+
"learning_rate": 9.954748808839675e-06,
|
| 239 |
+
"loss": 0.8108,
|
| 240 |
+
"step": 33
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.4349680170575693,
|
| 244 |
+
"grad_norm": 0.5792050957679749,
|
| 245 |
+
"learning_rate": 9.944154131125643e-06,
|
| 246 |
+
"loss": 0.7906,
|
| 247 |
+
"step": 34
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.44776119402985076,
|
| 251 |
+
"grad_norm": 0.7301996946334839,
|
| 252 |
+
"learning_rate": 9.932452969617607e-06,
|
| 253 |
+
"loss": 0.8191,
|
| 254 |
+
"step": 35
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.4605543710021322,
|
| 258 |
+
"grad_norm": 0.6449064612388611,
|
| 259 |
+
"learning_rate": 9.91964794299315e-06,
|
| 260 |
+
"loss": 0.7929,
|
| 261 |
+
"step": 36
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.47334754797441364,
|
| 265 |
+
"grad_norm": 0.6409065127372742,
|
| 266 |
+
"learning_rate": 9.905741916970863e-06,
|
| 267 |
+
"loss": 0.853,
|
| 268 |
+
"step": 37
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.4861407249466951,
|
| 272 |
+
"grad_norm": 0.5154856443405151,
|
| 273 |
+
"learning_rate": 9.890738003669029e-06,
|
| 274 |
+
"loss": 0.7733,
|
| 275 |
+
"step": 38
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.4989339019189765,
|
| 279 |
+
"grad_norm": 0.6274257302284241,
|
| 280 |
+
"learning_rate": 9.874639560909118e-06,
|
| 281 |
+
"loss": 0.8247,
|
| 282 |
+
"step": 39
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.511727078891258,
|
| 286 |
+
"grad_norm": 0.636122465133667,
|
| 287 |
+
"learning_rate": 9.857450191464337e-06,
|
| 288 |
+
"loss": 0.8102,
|
| 289 |
+
"step": 40
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.5245202558635395,
|
| 293 |
+
"grad_norm": 0.5821788311004639,
|
| 294 |
+
"learning_rate": 9.839173742253334e-06,
|
| 295 |
+
"loss": 0.7774,
|
| 296 |
+
"step": 41
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.5373134328358209,
|
| 300 |
+
"grad_norm": 0.5190237760543823,
|
| 301 |
+
"learning_rate": 9.819814303479268e-06,
|
| 302 |
+
"loss": 0.766,
|
| 303 |
+
"step": 42
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.5501066098081023,
|
| 307 |
+
"grad_norm": 0.49539390206336975,
|
| 308 |
+
"learning_rate": 9.799376207714446e-06,
|
| 309 |
+
"loss": 0.7937,
|
| 310 |
+
"step": 43
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.5628997867803838,
|
| 314 |
+
"grad_norm": 0.5138888359069824,
|
| 315 |
+
"learning_rate": 9.777864028930705e-06,
|
| 316 |
+
"loss": 0.7861,
|
| 317 |
+
"step": 44
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.5756929637526652,
|
| 321 |
+
"grad_norm": 0.49708378314971924,
|
| 322 |
+
"learning_rate": 9.755282581475769e-06,
|
| 323 |
+
"loss": 0.7356,
|
| 324 |
+
"step": 45
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.5884861407249466,
|
| 328 |
+
"grad_norm": 0.5589980483055115,
|
| 329 |
+
"learning_rate": 9.731636918995821e-06,
|
| 330 |
+
"loss": 0.7157,
|
| 331 |
+
"step": 46
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.6012793176972282,
|
| 335 |
+
"grad_norm": 0.5140479207038879,
|
| 336 |
+
"learning_rate": 9.706932333304518e-06,
|
| 337 |
+
"loss": 0.7599,
|
| 338 |
+
"step": 47
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.6140724946695096,
|
| 342 |
+
"grad_norm": 0.5195008516311646,
|
| 343 |
+
"learning_rate": 9.681174353198687e-06,
|
| 344 |
+
"loss": 0.7896,
|
| 345 |
+
"step": 48
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.6268656716417911,
|
| 349 |
+
"grad_norm": 0.510276734828949,
|
| 350 |
+
"learning_rate": 9.654368743221022e-06,
|
| 351 |
+
"loss": 0.7775,
|
| 352 |
+
"step": 49
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.6396588486140725,
|
| 356 |
+
"grad_norm": 0.6014343500137329,
|
| 357 |
+
"learning_rate": 9.626521502369984e-06,
|
| 358 |
+
"loss": 0.7615,
|
| 359 |
+
"step": 50
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.652452025586354,
|
| 363 |
+
"grad_norm": 0.5976372361183167,
|
| 364 |
+
"learning_rate": 9.597638862757255e-06,
|
| 365 |
+
"loss": 0.8036,
|
| 366 |
+
"step": 51
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.6652452025586354,
|
| 370 |
+
"grad_norm": 0.6337112188339233,
|
| 371 |
+
"learning_rate": 9.567727288213005e-06,
|
| 372 |
+
"loss": 0.8269,
|
| 373 |
+
"step": 52
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.6780383795309168,
|
| 377 |
+
"grad_norm": 0.4970976710319519,
|
| 378 |
+
"learning_rate": 9.536793472839325e-06,
|
| 379 |
+
"loss": 0.7883,
|
| 380 |
+
"step": 53
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.6908315565031983,
|
| 384 |
+
"grad_norm": 0.5633640289306641,
|
| 385 |
+
"learning_rate": 9.504844339512096e-06,
|
| 386 |
+
"loss": 0.7726,
|
| 387 |
+
"step": 54
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.7036247334754797,
|
| 391 |
+
"grad_norm": 0.6632335186004639,
|
| 392 |
+
"learning_rate": 9.471887038331686e-06,
|
| 393 |
+
"loss": 0.7631,
|
| 394 |
+
"step": 55
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.7164179104477612,
|
| 398 |
+
"grad_norm": 0.6567376852035522,
|
| 399 |
+
"learning_rate": 9.437928945022772e-06,
|
| 400 |
+
"loss": 0.7916,
|
| 401 |
+
"step": 56
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.7292110874200426,
|
| 405 |
+
"grad_norm": 0.4805288016796112,
|
| 406 |
+
"learning_rate": 9.40297765928369e-06,
|
| 407 |
+
"loss": 0.7652,
|
| 408 |
+
"step": 57
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.7420042643923241,
|
| 412 |
+
"grad_norm": 0.5745772123336792,
|
| 413 |
+
"learning_rate": 9.36704100308565e-06,
|
| 414 |
+
"loss": 0.7946,
|
| 415 |
+
"step": 58
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.7547974413646056,
|
| 419 |
+
"grad_norm": 0.5194308161735535,
|
| 420 |
+
"learning_rate": 9.330127018922195e-06,
|
| 421 |
+
"loss": 0.7118,
|
| 422 |
+
"step": 59
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.767590618336887,
|
| 426 |
+
"grad_norm": 0.4913547933101654,
|
| 427 |
+
"learning_rate": 9.292243968009332e-06,
|
| 428 |
+
"loss": 0.7412,
|
| 429 |
+
"step": 60
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.7803837953091685,
|
| 433 |
+
"grad_norm": 0.5241764187812805,
|
| 434 |
+
"learning_rate": 9.253400328436699e-06,
|
| 435 |
+
"loss": 0.7634,
|
| 436 |
+
"step": 61
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.7931769722814499,
|
| 440 |
+
"grad_norm": 0.5345350503921509,
|
| 441 |
+
"learning_rate": 9.213604793270196e-06,
|
| 442 |
+
"loss": 0.7353,
|
| 443 |
+
"step": 62
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.8059701492537313,
|
| 447 |
+
"grad_norm": 0.5227858424186707,
|
| 448 |
+
"learning_rate": 9.172866268606514e-06,
|
| 449 |
+
"loss": 0.7697,
|
| 450 |
+
"step": 63
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.8187633262260128,
|
| 454 |
+
"grad_norm": 0.4864361584186554,
|
| 455 |
+
"learning_rate": 9.131193871579975e-06,
|
| 456 |
+
"loss": 0.8112,
|
| 457 |
+
"step": 64
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.8315565031982942,
|
| 461 |
+
"grad_norm": 0.5334176421165466,
|
| 462 |
+
"learning_rate": 9.088596928322158e-06,
|
| 463 |
+
"loss": 0.731,
|
| 464 |
+
"step": 65
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.8443496801705757,
|
| 468 |
+
"grad_norm": 0.4496009945869446,
|
| 469 |
+
"learning_rate": 9.045084971874738e-06,
|
| 470 |
+
"loss": 0.7206,
|
| 471 |
+
"step": 66
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.8571428571428571,
|
| 475 |
+
"grad_norm": 0.4755212068557739,
|
| 476 |
+
"learning_rate": 9.000667740056033e-06,
|
| 477 |
+
"loss": 0.7843,
|
| 478 |
+
"step": 67
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.8699360341151386,
|
| 482 |
+
"grad_norm": 0.45297691226005554,
|
| 483 |
+
"learning_rate": 8.955355173281709e-06,
|
| 484 |
+
"loss": 0.76,
|
| 485 |
+
"step": 68
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.8827292110874201,
|
| 489 |
+
"grad_norm": 0.4970425069332123,
|
| 490 |
+
"learning_rate": 8.90915741234015e-06,
|
| 491 |
+
"loss": 0.7674,
|
| 492 |
+
"step": 69
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 0.8955223880597015,
|
| 496 |
+
"grad_norm": 0.5083153247833252,
|
| 497 |
+
"learning_rate": 8.862084796122998e-06,
|
| 498 |
+
"loss": 0.7656,
|
| 499 |
+
"step": 70
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.908315565031983,
|
| 503 |
+
"grad_norm": 0.45638445019721985,
|
| 504 |
+
"learning_rate": 8.814147859311333e-06,
|
| 505 |
+
"loss": 0.7303,
|
| 506 |
+
"step": 71
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.9211087420042644,
|
| 510 |
+
"grad_norm": 0.4436575472354889,
|
| 511 |
+
"learning_rate": 8.765357330018056e-06,
|
| 512 |
+
"loss": 0.7295,
|
| 513 |
+
"step": 72
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.9339019189765458,
|
| 517 |
+
"grad_norm": 0.5214686393737793,
|
| 518 |
+
"learning_rate": 8.715724127386971e-06,
|
| 519 |
+
"loss": 0.7075,
|
| 520 |
+
"step": 73
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.9466950959488273,
|
| 524 |
+
"grad_norm": 0.48304834961891174,
|
| 525 |
+
"learning_rate": 8.665259359149132e-06,
|
| 526 |
+
"loss": 0.7279,
|
| 527 |
+
"step": 74
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.9594882729211087,
|
| 531 |
+
"grad_norm": 0.5089297294616699,
|
| 532 |
+
"learning_rate": 8.613974319136959e-06,
|
| 533 |
+
"loss": 0.7424,
|
| 534 |
+
"step": 75
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.9722814498933902,
|
| 538 |
+
"grad_norm": 0.49402785301208496,
|
| 539 |
+
"learning_rate": 8.561880484756726e-06,
|
| 540 |
+
"loss": 0.7957,
|
| 541 |
+
"step": 76
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.9850746268656716,
|
| 545 |
+
"grad_norm": 0.5247993469238281,
|
| 546 |
+
"learning_rate": 8.508989514419959e-06,
|
| 547 |
+
"loss": 0.7539,
|
| 548 |
+
"step": 77
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.997867803837953,
|
| 552 |
+
"grad_norm": 0.4710896611213684,
|
| 553 |
+
"learning_rate": 8.455313244934324e-06,
|
| 554 |
+
"loss": 0.7057,
|
| 555 |
+
"step": 78
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 1.0106609808102345,
|
| 559 |
+
"grad_norm": 1.7053124904632568,
|
| 560 |
+
"learning_rate": 8.400863688854598e-06,
|
| 561 |
+
"loss": 1.2963,
|
| 562 |
+
"step": 79
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 1.023454157782516,
|
| 566 |
+
"grad_norm": 0.5480591654777527,
|
| 567 |
+
"learning_rate": 8.345653031794292e-06,
|
| 568 |
+
"loss": 0.7255,
|
| 569 |
+
"step": 80
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 1.0362473347547974,
|
| 573 |
+
"grad_norm": 0.5014137029647827,
|
| 574 |
+
"learning_rate": 8.289693629698564e-06,
|
| 575 |
+
"loss": 0.7831,
|
| 576 |
+
"step": 81
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 1.049040511727079,
|
| 580 |
+
"grad_norm": 0.4693015515804291,
|
| 581 |
+
"learning_rate": 8.232998006078998e-06,
|
| 582 |
+
"loss": 0.7265,
|
| 583 |
+
"step": 82
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 1.0618336886993602,
|
| 587 |
+
"grad_norm": 0.4948160648345947,
|
| 588 |
+
"learning_rate": 8.175578849210894e-06,
|
| 589 |
+
"loss": 0.7189,
|
| 590 |
+
"step": 83
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 1.0746268656716418,
|
| 594 |
+
"grad_norm": 0.6815115213394165,
|
| 595 |
+
"learning_rate": 8.117449009293668e-06,
|
| 596 |
+
"loss": 0.7964,
|
| 597 |
+
"step": 84
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 1.0874200426439233,
|
| 601 |
+
"grad_norm": 0.44852644205093384,
|
| 602 |
+
"learning_rate": 8.058621495575032e-06,
|
| 603 |
+
"loss": 0.6198,
|
| 604 |
+
"step": 85
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 1.1002132196162047,
|
| 608 |
+
"grad_norm": 0.535753071308136,
|
| 609 |
+
"learning_rate": 7.99910947343957e-06,
|
| 610 |
+
"loss": 0.6642,
|
| 611 |
+
"step": 86
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 1.1130063965884862,
|
| 615 |
+
"grad_norm": 0.46651574969291687,
|
| 616 |
+
"learning_rate": 7.938926261462366e-06,
|
| 617 |
+
"loss": 0.731,
|
| 618 |
+
"step": 87
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 1.1257995735607675,
|
| 622 |
+
"grad_norm": 0.5469245910644531,
|
| 623 |
+
"learning_rate": 7.87808532842837e-06,
|
| 624 |
+
"loss": 0.794,
|
| 625 |
+
"step": 88
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 1.138592750533049,
|
| 629 |
+
"grad_norm": 0.5276331901550293,
|
| 630 |
+
"learning_rate": 7.81660029031811e-06,
|
| 631 |
+
"loss": 0.6953,
|
| 632 |
+
"step": 89
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 1.1513859275053304,
|
| 636 |
+
"grad_norm": 0.42721840739250183,
|
| 637 |
+
"learning_rate": 7.754484907260513e-06,
|
| 638 |
+
"loss": 0.7541,
|
| 639 |
+
"step": 90
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 1.164179104477612,
|
| 643 |
+
"grad_norm": 0.4775303602218628,
|
| 644 |
+
"learning_rate": 7.691753080453413e-06,
|
| 645 |
+
"loss": 0.704,
|
| 646 |
+
"step": 91
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 1.1769722814498933,
|
| 650 |
+
"grad_norm": 0.5940173268318176,
|
| 651 |
+
"learning_rate": 7.628418849052523e-06,
|
| 652 |
+
"loss": 0.778,
|
| 653 |
+
"step": 92
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 1.1897654584221748,
|
| 657 |
+
"grad_norm": 0.4526679217815399,
|
| 658 |
+
"learning_rate": 7.564496387029532e-06,
|
| 659 |
+
"loss": 0.7215,
|
| 660 |
+
"step": 93
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 1.2025586353944564,
|
| 664 |
+
"grad_norm": 0.4627590775489807,
|
| 665 |
+
"learning_rate": 7.500000000000001e-06,
|
| 666 |
+
"loss": 0.7299,
|
| 667 |
+
"step": 94
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 1.2153518123667377,
|
| 671 |
+
"grad_norm": 0.4942200183868408,
|
| 672 |
+
"learning_rate": 7.434944122021837e-06,
|
| 673 |
+
"loss": 0.7224,
|
| 674 |
+
"step": 95
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 1.2281449893390193,
|
| 678 |
+
"grad_norm": 0.428196519613266,
|
| 679 |
+
"learning_rate": 7.369343312364994e-06,
|
| 680 |
+
"loss": 0.7617,
|
| 681 |
+
"step": 96
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 1.2409381663113006,
|
| 685 |
+
"grad_norm": 0.4644979238510132,
|
| 686 |
+
"learning_rate": 7.303212252253163e-06,
|
| 687 |
+
"loss": 0.7441,
|
| 688 |
+
"step": 97
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 1.2537313432835822,
|
| 692 |
+
"grad_norm": 0.5221542119979858,
|
| 693 |
+
"learning_rate": 7.236565741578163e-06,
|
| 694 |
+
"loss": 0.7311,
|
| 695 |
+
"step": 98
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 1.2665245202558635,
|
| 699 |
+
"grad_norm": 0.43763238191604614,
|
| 700 |
+
"learning_rate": 7.169418695587791e-06,
|
| 701 |
+
"loss": 0.7698,
|
| 702 |
+
"step": 99
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 1.279317697228145,
|
| 706 |
+
"grad_norm": 0.4271838366985321,
|
| 707 |
+
"learning_rate": 7.101786141547829e-06,
|
| 708 |
+
"loss": 0.6731,
|
| 709 |
+
"step": 100
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 1.2921108742004264,
|
| 713 |
+
"grad_norm": 0.4979677200317383,
|
| 714 |
+
"learning_rate": 7.033683215379002e-06,
|
| 715 |
+
"loss": 0.7077,
|
| 716 |
+
"step": 101
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 1.304904051172708,
|
| 720 |
+
"grad_norm": 0.4421441853046417,
|
| 721 |
+
"learning_rate": 6.965125158269619e-06,
|
| 722 |
+
"loss": 0.7422,
|
| 723 |
+
"step": 102
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 1.3176972281449895,
|
| 727 |
+
"grad_norm": 0.44597184658050537,
|
| 728 |
+
"learning_rate": 6.896127313264643e-06,
|
| 729 |
+
"loss": 0.7129,
|
| 730 |
+
"step": 103
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 1.3304904051172708,
|
| 734 |
+
"grad_norm": 0.48497170209884644,
|
| 735 |
+
"learning_rate": 6.8267051218319766e-06,
|
| 736 |
+
"loss": 0.7248,
|
| 737 |
+
"step": 104
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 1.3432835820895521,
|
| 741 |
+
"grad_norm": 0.5230187177658081,
|
| 742 |
+
"learning_rate": 6.7568741204067145e-06,
|
| 743 |
+
"loss": 0.6923,
|
| 744 |
+
"step": 105
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 1.3560767590618337,
|
| 748 |
+
"grad_norm": 0.47476544976234436,
|
| 749 |
+
"learning_rate": 6.686649936914151e-06,
|
| 750 |
+
"loss": 0.732,
|
| 751 |
+
"step": 106
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 1.3688699360341152,
|
| 755 |
+
"grad_norm": 0.40445131063461304,
|
| 756 |
+
"learning_rate": 6.616048287272301e-06,
|
| 757 |
+
"loss": 0.7672,
|
| 758 |
+
"step": 107
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 1.3816631130063965,
|
| 762 |
+
"grad_norm": 0.4726698100566864,
|
| 763 |
+
"learning_rate": 6.545084971874738e-06,
|
| 764 |
+
"loss": 0.7898,
|
| 765 |
+
"step": 108
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 1.394456289978678,
|
| 769 |
+
"grad_norm": 0.4922250211238861,
|
| 770 |
+
"learning_rate": 6.473775872054522e-06,
|
| 771 |
+
"loss": 0.7054,
|
| 772 |
+
"step": 109
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 1.4072494669509594,
|
| 776 |
+
"grad_norm": 0.4099796712398529,
|
| 777 |
+
"learning_rate": 6.402136946530014e-06,
|
| 778 |
+
"loss": 0.6099,
|
| 779 |
+
"step": 110
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 1.420042643923241,
|
| 783 |
+
"grad_norm": 0.5015040040016174,
|
| 784 |
+
"learning_rate": 6.330184227833376e-06,
|
| 785 |
+
"loss": 0.7295,
|
| 786 |
+
"step": 111
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 1.4328358208955223,
|
| 790 |
+
"grad_norm": 0.43680623173713684,
|
| 791 |
+
"learning_rate": 6.257933818722544e-06,
|
| 792 |
+
"loss": 0.7267,
|
| 793 |
+
"step": 112
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 1.4456289978678039,
|
| 797 |
+
"grad_norm": 0.4174647927284241,
|
| 798 |
+
"learning_rate": 6.185401888577488e-06,
|
| 799 |
+
"loss": 0.6789,
|
| 800 |
+
"step": 113
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 1.4584221748400852,
|
| 804 |
+
"grad_norm": 0.45874616503715515,
|
| 805 |
+
"learning_rate": 6.112604669781572e-06,
|
| 806 |
+
"loss": 0.8178,
|
| 807 |
+
"step": 114
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 1.4712153518123667,
|
| 811 |
+
"grad_norm": 0.39787718653678894,
|
| 812 |
+
"learning_rate": 6.039558454088796e-06,
|
| 813 |
+
"loss": 0.6645,
|
| 814 |
+
"step": 115
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 1.4840085287846483,
|
| 818 |
+
"grad_norm": 0.6014237403869629,
|
| 819 |
+
"learning_rate": 5.9662795889777666e-06,
|
| 820 |
+
"loss": 0.7598,
|
| 821 |
+
"step": 116
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 1.4968017057569296,
|
| 825 |
+
"grad_norm": 0.4093945026397705,
|
| 826 |
+
"learning_rate": 5.892784473993184e-06,
|
| 827 |
+
"loss": 0.6873,
|
| 828 |
+
"step": 117
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 1.509594882729211,
|
| 832 |
+
"grad_norm": 0.48048579692840576,
|
| 833 |
+
"learning_rate": 5.819089557075689e-06,
|
| 834 |
+
"loss": 0.7689,
|
| 835 |
+
"step": 118
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 1.5223880597014925,
|
| 839 |
+
"grad_norm": 0.45258212089538574,
|
| 840 |
+
"learning_rate": 5.745211330880872e-06,
|
| 841 |
+
"loss": 0.6531,
|
| 842 |
+
"step": 119
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 1.535181236673774,
|
| 846 |
+
"grad_norm": 0.4489460289478302,
|
| 847 |
+
"learning_rate": 5.671166329088278e-06,
|
| 848 |
+
"loss": 0.7181,
|
| 849 |
+
"step": 120
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 1.5479744136460556,
|
| 853 |
+
"grad_norm": 0.39668354392051697,
|
| 854 |
+
"learning_rate": 5.596971122701221e-06,
|
| 855 |
+
"loss": 0.6797,
|
| 856 |
+
"step": 121
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 1.560767590618337,
|
| 860 |
+
"grad_norm": 0.3912845849990845,
|
| 861 |
+
"learning_rate": 5.522642316338268e-06,
|
| 862 |
+
"loss": 0.736,
|
| 863 |
+
"step": 122
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 1.5735607675906182,
|
| 867 |
+
"grad_norm": 0.45017972588539124,
|
| 868 |
+
"learning_rate": 5.448196544517168e-06,
|
| 869 |
+
"loss": 0.7289,
|
| 870 |
+
"step": 123
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 1.5863539445628998,
|
| 874 |
+
"grad_norm": 0.4025014042854309,
|
| 875 |
+
"learning_rate": 5.373650467932122e-06,
|
| 876 |
+
"loss": 0.7216,
|
| 877 |
+
"step": 124
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 1.5991471215351813,
|
| 881 |
+
"grad_norm": 0.4008171558380127,
|
| 882 |
+
"learning_rate": 5.299020769725172e-06,
|
| 883 |
+
"loss": 0.6882,
|
| 884 |
+
"step": 125
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 1.6119402985074627,
|
| 888 |
+
"grad_norm": 0.3734460175037384,
|
| 889 |
+
"learning_rate": 5.224324151752575e-06,
|
| 890 |
+
"loss": 0.7057,
|
| 891 |
+
"step": 126
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 1.624733475479744,
|
| 895 |
+
"grad_norm": 0.41330069303512573,
|
| 896 |
+
"learning_rate": 5.1495773308469935e-06,
|
| 897 |
+
"loss": 0.6429,
|
| 898 |
+
"step": 127
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 1.6375266524520256,
|
| 902 |
+
"grad_norm": 0.409628689289093,
|
| 903 |
+
"learning_rate": 5.074797035076319e-06,
|
| 904 |
+
"loss": 0.7052,
|
| 905 |
+
"step": 128
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 1.650319829424307,
|
| 909 |
+
"grad_norm": 0.3930547535419464,
|
| 910 |
+
"learning_rate": 5e-06,
|
| 911 |
+
"loss": 0.773,
|
| 912 |
+
"step": 129
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 1.6631130063965884,
|
| 916 |
+
"grad_norm": 0.47582224011421204,
|
| 917 |
+
"learning_rate": 4.9252029649236835e-06,
|
| 918 |
+
"loss": 0.7179,
|
| 919 |
+
"step": 130
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 1.67590618336887,
|
| 923 |
+
"grad_norm": 0.4209410548210144,
|
| 924 |
+
"learning_rate": 4.850422669153009e-06,
|
| 925 |
+
"loss": 0.6379,
|
| 926 |
+
"step": 131
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 1.6886993603411513,
|
| 930 |
+
"grad_norm": 0.4784112870693207,
|
| 931 |
+
"learning_rate": 4.775675848247427e-06,
|
| 932 |
+
"loss": 0.7232,
|
| 933 |
+
"step": 132
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 1.7014925373134329,
|
| 937 |
+
"grad_norm": 0.44215840101242065,
|
| 938 |
+
"learning_rate": 4.700979230274829e-06,
|
| 939 |
+
"loss": 0.663,
|
| 940 |
+
"step": 133
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 1.7142857142857144,
|
| 944 |
+
"grad_norm": 0.5071489810943604,
|
| 945 |
+
"learning_rate": 4.626349532067879e-06,
|
| 946 |
+
"loss": 0.7215,
|
| 947 |
+
"step": 134
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 1.7270788912579957,
|
| 951 |
+
"grad_norm": 0.47784915566444397,
|
| 952 |
+
"learning_rate": 4.551803455482833e-06,
|
| 953 |
+
"loss": 0.788,
|
| 954 |
+
"step": 135
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 1.739872068230277,
|
| 958 |
+
"grad_norm": 0.5307613611221313,
|
| 959 |
+
"learning_rate": 4.477357683661734e-06,
|
| 960 |
+
"loss": 0.745,
|
| 961 |
+
"step": 136
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 1.7526652452025586,
|
| 965 |
+
"grad_norm": 0.455413281917572,
|
| 966 |
+
"learning_rate": 4.4030288772987795e-06,
|
| 967 |
+
"loss": 0.7129,
|
| 968 |
+
"step": 137
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 1.7654584221748402,
|
| 972 |
+
"grad_norm": 0.4110308289527893,
|
| 973 |
+
"learning_rate": 4.3288336709117246e-06,
|
| 974 |
+
"loss": 0.7328,
|
| 975 |
+
"step": 138
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 1.7782515991471215,
|
| 979 |
+
"grad_norm": 0.5132991075515747,
|
| 980 |
+
"learning_rate": 4.254788669119127e-06,
|
| 981 |
+
"loss": 0.7738,
|
| 982 |
+
"step": 139
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 1.7910447761194028,
|
| 986 |
+
"grad_norm": 0.4838980436325073,
|
| 987 |
+
"learning_rate": 4.180910442924312e-06,
|
| 988 |
+
"loss": 0.6317,
|
| 989 |
+
"step": 140
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 1.8038379530916844,
|
| 993 |
+
"grad_norm": 0.4499025046825409,
|
| 994 |
+
"learning_rate": 4.107215526006818e-06,
|
| 995 |
+
"loss": 0.7288,
|
| 996 |
+
"step": 141
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 1.816631130063966,
|
| 1000 |
+
"grad_norm": 0.47396278381347656,
|
| 1001 |
+
"learning_rate": 4.033720411022235e-06,
|
| 1002 |
+
"loss": 0.776,
|
| 1003 |
+
"step": 142
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 1.8294243070362475,
|
| 1007 |
+
"grad_norm": 0.3979196548461914,
|
| 1008 |
+
"learning_rate": 3.960441545911205e-06,
|
| 1009 |
+
"loss": 0.683,
|
| 1010 |
+
"step": 143
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 1.8422174840085288,
|
| 1014 |
+
"grad_norm": 0.43219274282455444,
|
| 1015 |
+
"learning_rate": 3.887395330218429e-06,
|
| 1016 |
+
"loss": 0.7194,
|
| 1017 |
+
"step": 144
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 1.8550106609808101,
|
| 1021 |
+
"grad_norm": 0.4215470254421234,
|
| 1022 |
+
"learning_rate": 3.8145981114225135e-06,
|
| 1023 |
+
"loss": 0.7744,
|
| 1024 |
+
"step": 145
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 1.8678038379530917,
|
| 1028 |
+
"grad_norm": 0.4141583740711212,
|
| 1029 |
+
"learning_rate": 3.7420661812774577e-06,
|
| 1030 |
+
"loss": 0.7245,
|
| 1031 |
+
"step": 146
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 1.8805970149253732,
|
| 1035 |
+
"grad_norm": 0.3775157332420349,
|
| 1036 |
+
"learning_rate": 3.669815772166625e-06,
|
| 1037 |
+
"loss": 0.6878,
|
| 1038 |
+
"step": 147
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 1.8933901918976546,
|
| 1042 |
+
"grad_norm": 0.4012461304664612,
|
| 1043 |
+
"learning_rate": 3.5978630534699873e-06,
|
| 1044 |
+
"loss": 0.7315,
|
| 1045 |
+
"step": 148
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 1.906183368869936,
|
| 1049 |
+
"grad_norm": 0.3893410861492157,
|
| 1050 |
+
"learning_rate": 3.526224127945479e-06,
|
| 1051 |
+
"loss": 0.6943,
|
| 1052 |
+
"step": 149
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 1.9189765458422174,
|
| 1056 |
+
"grad_norm": 0.4317570626735687,
|
| 1057 |
+
"learning_rate": 3.4549150281252635e-06,
|
| 1058 |
+
"loss": 0.6904,
|
| 1059 |
+
"step": 150
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 1.931769722814499,
|
| 1063 |
+
"grad_norm": 0.4641903340816498,
|
| 1064 |
+
"learning_rate": 3.383951712727701e-06,
|
| 1065 |
+
"loss": 0.7495,
|
| 1066 |
+
"step": 151
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 1.9445628997867805,
|
| 1070 |
+
"grad_norm": 0.4203890264034271,
|
| 1071 |
+
"learning_rate": 3.3133500630858507e-06,
|
| 1072 |
+
"loss": 0.7436,
|
| 1073 |
+
"step": 152
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 1.9573560767590619,
|
| 1077 |
+
"grad_norm": 0.39781543612480164,
|
| 1078 |
+
"learning_rate": 3.2431258795932863e-06,
|
| 1079 |
+
"loss": 0.7042,
|
| 1080 |
+
"step": 153
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 1.9701492537313432,
|
| 1084 |
+
"grad_norm": 0.4704498052597046,
|
| 1085 |
+
"learning_rate": 3.173294878168025e-06,
|
| 1086 |
+
"loss": 0.7379,
|
| 1087 |
+
"step": 154
|
| 1088 |
+
},
|
| 1089 |
+
{
|
| 1090 |
+
"epoch": 1.9829424307036247,
|
| 1091 |
+
"grad_norm": 0.39690351486206055,
|
| 1092 |
+
"learning_rate": 3.1038726867353587e-06,
|
| 1093 |
+
"loss": 0.6986,
|
| 1094 |
+
"step": 155
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 1.9957356076759063,
|
| 1098 |
+
"grad_norm": 0.46861645579338074,
|
| 1099 |
+
"learning_rate": 3.0348748417303826e-06,
|
| 1100 |
+
"loss": 0.7317,
|
| 1101 |
+
"step": 156
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 2.008528784648188,
|
| 1105 |
+
"grad_norm": 1.1607420444488525,
|
| 1106 |
+
"learning_rate": 2.966316784621e-06,
|
| 1107 |
+
"loss": 1.2703,
|
| 1108 |
+
"step": 157
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"epoch": 2.021321961620469,
|
| 1112 |
+
"grad_norm": 0.38464683294296265,
|
| 1113 |
+
"learning_rate": 2.8982138584521734e-06,
|
| 1114 |
+
"loss": 0.7064,
|
| 1115 |
+
"step": 158
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 2.0341151385927505,
|
| 1119 |
+
"grad_norm": 0.5022476315498352,
|
| 1120 |
+
"learning_rate": 2.83058130441221e-06,
|
| 1121 |
+
"loss": 0.7061,
|
| 1122 |
+
"step": 159
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 2.046908315565032,
|
| 1126 |
+
"grad_norm": 0.5013148188591003,
|
| 1127 |
+
"learning_rate": 2.7634342584218364e-06,
|
| 1128 |
+
"loss": 0.7014,
|
| 1129 |
+
"step": 160
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 2.0597014925373136,
|
| 1133 |
+
"grad_norm": 0.47904056310653687,
|
| 1134 |
+
"learning_rate": 2.6967877477468394e-06,
|
| 1135 |
+
"loss": 0.6972,
|
| 1136 |
+
"step": 161
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 2.0724946695095947,
|
| 1140 |
+
"grad_norm": 0.41209492087364197,
|
| 1141 |
+
"learning_rate": 2.6306566876350072e-06,
|
| 1142 |
+
"loss": 0.6418,
|
| 1143 |
+
"step": 162
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 2.0852878464818763,
|
| 1147 |
+
"grad_norm": 0.46157318353652954,
|
| 1148 |
+
"learning_rate": 2.5650558779781635e-06,
|
| 1149 |
+
"loss": 0.7516,
|
| 1150 |
+
"step": 163
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 2.098081023454158,
|
| 1154 |
+
"grad_norm": 0.42628005146980286,
|
| 1155 |
+
"learning_rate": 2.5000000000000015e-06,
|
| 1156 |
+
"loss": 0.7129,
|
| 1157 |
+
"step": 164
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 2.1108742004264394,
|
| 1161 |
+
"grad_norm": 0.45490726828575134,
|
| 1162 |
+
"learning_rate": 2.43550361297047e-06,
|
| 1163 |
+
"loss": 0.7031,
|
| 1164 |
+
"step": 165
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 2.1236673773987205,
|
| 1168 |
+
"grad_norm": 0.4108670651912689,
|
| 1169 |
+
"learning_rate": 2.371581150947476e-06,
|
| 1170 |
+
"loss": 0.7059,
|
| 1171 |
+
"step": 166
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 2.136460554371002,
|
| 1175 |
+
"grad_norm": 0.4316941797733307,
|
| 1176 |
+
"learning_rate": 2.3082469195465893e-06,
|
| 1177 |
+
"loss": 0.6921,
|
| 1178 |
+
"step": 167
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 2.1492537313432836,
|
| 1182 |
+
"grad_norm": 0.4376332759857178,
|
| 1183 |
+
"learning_rate": 2.245515092739488e-06,
|
| 1184 |
+
"loss": 0.6459,
|
| 1185 |
+
"step": 168
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 2.162046908315565,
|
| 1189 |
+
"grad_norm": 0.4335668981075287,
|
| 1190 |
+
"learning_rate": 2.1833997096818897e-06,
|
| 1191 |
+
"loss": 0.6815,
|
| 1192 |
+
"step": 169
|
| 1193 |
+
},
|
| 1194 |
+
{
|
| 1195 |
+
"epoch": 2.1748400852878467,
|
| 1196 |
+
"grad_norm": 0.37335702776908875,
|
| 1197 |
+
"learning_rate": 2.1219146715716332e-06,
|
| 1198 |
+
"loss": 0.6917,
|
| 1199 |
+
"step": 170
|
| 1200 |
+
},
|
| 1201 |
+
{
|
| 1202 |
+
"epoch": 2.1876332622601278,
|
| 1203 |
+
"grad_norm": 0.4028381109237671,
|
| 1204 |
+
"learning_rate": 2.061073738537635e-06,
|
| 1205 |
+
"loss": 0.6415,
|
| 1206 |
+
"step": 171
|
| 1207 |
+
},
|
| 1208 |
+
{
|
| 1209 |
+
"epoch": 2.2004264392324093,
|
| 1210 |
+
"grad_norm": 0.4265342056751251,
|
| 1211 |
+
"learning_rate": 2.0008905265604316e-06,
|
| 1212 |
+
"loss": 0.6476,
|
| 1213 |
+
"step": 172
|
| 1214 |
+
},
|
| 1215 |
+
{
|
| 1216 |
+
"epoch": 2.213219616204691,
|
| 1217 |
+
"grad_norm": 0.4345915615558624,
|
| 1218 |
+
"learning_rate": 1.941378504424968e-06,
|
| 1219 |
+
"loss": 0.8098,
|
| 1220 |
+
"step": 173
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 2.2260127931769724,
|
| 1224 |
+
"grad_norm": 0.35505157709121704,
|
| 1225 |
+
"learning_rate": 1.8825509907063328e-06,
|
| 1226 |
+
"loss": 0.6996,
|
| 1227 |
+
"step": 174
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"epoch": 2.2388059701492535,
|
| 1231 |
+
"grad_norm": 0.3520065248012543,
|
| 1232 |
+
"learning_rate": 1.8244211507891064e-06,
|
| 1233 |
+
"loss": 0.6946,
|
| 1234 |
+
"step": 175
|
| 1235 |
+
},
|
| 1236 |
+
{
|
| 1237 |
+
"epoch": 2.251599147121535,
|
| 1238 |
+
"grad_norm": 0.3407677114009857,
|
| 1239 |
+
"learning_rate": 1.7670019939210025e-06,
|
| 1240 |
+
"loss": 0.7051,
|
| 1241 |
+
"step": 176
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 2.2643923240938166,
|
| 1245 |
+
"grad_norm": 0.39579153060913086,
|
| 1246 |
+
"learning_rate": 1.7103063703014372e-06,
|
| 1247 |
+
"loss": 0.6582,
|
| 1248 |
+
"step": 177
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 2.277185501066098,
|
| 1252 |
+
"grad_norm": 0.3585205078125,
|
| 1253 |
+
"learning_rate": 1.6543469682057105e-06,
|
| 1254 |
+
"loss": 0.6591,
|
| 1255 |
+
"step": 178
|
| 1256 |
+
},
|
| 1257 |
+
{
|
| 1258 |
+
"epoch": 2.2899786780383797,
|
| 1259 |
+
"grad_norm": 0.3757244646549225,
|
| 1260 |
+
"learning_rate": 1.5991363111454023e-06,
|
| 1261 |
+
"loss": 0.6544,
|
| 1262 |
+
"step": 179
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 2.302771855010661,
|
| 1266 |
+
"grad_norm": 0.34080037474632263,
|
| 1267 |
+
"learning_rate": 1.544686755065677e-06,
|
| 1268 |
+
"loss": 0.6695,
|
| 1269 |
+
"step": 180
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 2.3155650319829424,
|
| 1273 |
+
"grad_norm": 0.36547356843948364,
|
| 1274 |
+
"learning_rate": 1.4910104855800429e-06,
|
| 1275 |
+
"loss": 0.6983,
|
| 1276 |
+
"step": 181
|
| 1277 |
+
},
|
| 1278 |
+
{
|
| 1279 |
+
"epoch": 2.328358208955224,
|
| 1280 |
+
"grad_norm": 0.3294934630393982,
|
| 1281 |
+
"learning_rate": 1.438119515243277e-06,
|
| 1282 |
+
"loss": 0.6253,
|
| 1283 |
+
"step": 182
|
| 1284 |
+
},
|
| 1285 |
+
{
|
| 1286 |
+
"epoch": 2.3411513859275055,
|
| 1287 |
+
"grad_norm": 0.3605087399482727,
|
| 1288 |
+
"learning_rate": 1.3860256808630429e-06,
|
| 1289 |
+
"loss": 0.6788,
|
| 1290 |
+
"step": 183
|
| 1291 |
+
},
|
| 1292 |
+
{
|
| 1293 |
+
"epoch": 2.3539445628997866,
|
| 1294 |
+
"grad_norm": 0.39677152037620544,
|
| 1295 |
+
"learning_rate": 1.3347406408508695e-06,
|
| 1296 |
+
"loss": 0.7431,
|
| 1297 |
+
"step": 184
|
| 1298 |
+
},
|
| 1299 |
+
{
|
| 1300 |
+
"epoch": 2.366737739872068,
|
| 1301 |
+
"grad_norm": 0.3269940912723541,
|
| 1302 |
+
"learning_rate": 1.2842758726130283e-06,
|
| 1303 |
+
"loss": 0.6587,
|
| 1304 |
+
"step": 185
|
| 1305 |
+
},
|
| 1306 |
+
{
|
| 1307 |
+
"epoch": 2.3795309168443497,
|
| 1308 |
+
"grad_norm": 0.3385223150253296,
|
| 1309 |
+
"learning_rate": 1.234642669981946e-06,
|
| 1310 |
+
"loss": 0.7,
|
| 1311 |
+
"step": 186
|
| 1312 |
+
},
|
| 1313 |
+
{
|
| 1314 |
+
"epoch": 2.3923240938166312,
|
| 1315 |
+
"grad_norm": 0.35292384028434753,
|
| 1316 |
+
"learning_rate": 1.1858521406886674e-06,
|
| 1317 |
+
"loss": 0.7007,
|
| 1318 |
+
"step": 187
|
| 1319 |
+
},
|
| 1320 |
+
{
|
| 1321 |
+
"epoch": 2.405117270788913,
|
| 1322 |
+
"grad_norm": 0.34651538729667664,
|
| 1323 |
+
"learning_rate": 1.137915203877003e-06,
|
| 1324 |
+
"loss": 0.7469,
|
| 1325 |
+
"step": 188
|
| 1326 |
+
},
|
| 1327 |
+
{
|
| 1328 |
+
"epoch": 2.417910447761194,
|
| 1329 |
+
"grad_norm": 0.32731908559799194,
|
| 1330 |
+
"learning_rate": 1.0908425876598512e-06,
|
| 1331 |
+
"loss": 0.6608,
|
| 1332 |
+
"step": 189
|
| 1333 |
+
},
|
| 1334 |
+
{
|
| 1335 |
+
"epoch": 2.4307036247334755,
|
| 1336 |
+
"grad_norm": 0.3515162765979767,
|
| 1337 |
+
"learning_rate": 1.044644826718295e-06,
|
| 1338 |
+
"loss": 0.7566,
|
| 1339 |
+
"step": 190
|
| 1340 |
+
},
|
| 1341 |
+
{
|
| 1342 |
+
"epoch": 2.443496801705757,
|
| 1343 |
+
"grad_norm": 0.31811511516571045,
|
| 1344 |
+
"learning_rate": 9.993322599439692e-07,
|
| 1345 |
+
"loss": 0.6596,
|
| 1346 |
+
"step": 191
|
| 1347 |
+
},
|
| 1348 |
+
{
|
| 1349 |
+
"epoch": 2.4562899786780386,
|
| 1350 |
+
"grad_norm": 0.3392126262187958,
|
| 1351 |
+
"learning_rate": 9.549150281252633e-07,
|
| 1352 |
+
"loss": 0.7325,
|
| 1353 |
+
"step": 192
|
| 1354 |
+
},
|
| 1355 |
+
{
|
| 1356 |
+
"epoch": 2.4690831556503197,
|
| 1357 |
+
"grad_norm": 0.33756223320961,
|
| 1358 |
+
"learning_rate": 9.114030716778433e-07,
|
| 1359 |
+
"loss": 0.6592,
|
| 1360 |
+
"step": 193
|
| 1361 |
+
},
|
| 1362 |
+
{
|
| 1363 |
+
"epoch": 2.481876332622601,
|
| 1364 |
+
"grad_norm": 0.34133580327033997,
|
| 1365 |
+
"learning_rate": 8.688061284200266e-07,
|
| 1366 |
+
"loss": 0.7221,
|
| 1367 |
+
"step": 194
|
| 1368 |
+
},
|
| 1369 |
+
{
|
| 1370 |
+
"epoch": 2.4946695095948828,
|
| 1371 |
+
"grad_norm": 0.3212495446205139,
|
| 1372 |
+
"learning_rate": 8.271337313934869e-07,
|
| 1373 |
+
"loss": 0.6698,
|
| 1374 |
+
"step": 195
|
| 1375 |
+
},
|
| 1376 |
+
{
|
| 1377 |
+
"epoch": 2.5074626865671643,
|
| 1378 |
+
"grad_norm": 0.3041667640209198,
|
| 1379 |
+
"learning_rate": 7.863952067298042e-07,
|
| 1380 |
+
"loss": 0.691,
|
| 1381 |
+
"step": 196
|
| 1382 |
+
},
|
| 1383 |
+
{
|
| 1384 |
+
"epoch": 2.520255863539446,
|
| 1385 |
+
"grad_norm": 0.35703015327453613,
|
| 1386 |
+
"learning_rate": 7.465996715633028e-07,
|
| 1387 |
+
"loss": 0.7766,
|
| 1388 |
+
"step": 197
|
| 1389 |
+
},
|
| 1390 |
+
{
|
| 1391 |
+
"epoch": 2.533049040511727,
|
| 1392 |
+
"grad_norm": 0.3571200966835022,
|
| 1393 |
+
"learning_rate": 7.077560319906696e-07,
|
| 1394 |
+
"loss": 0.7563,
|
| 1395 |
+
"step": 198
|
| 1396 |
+
},
|
| 1397 |
+
{
|
| 1398 |
+
"epoch": 2.5458422174840085,
|
| 1399 |
+
"grad_norm": 0.33537036180496216,
|
| 1400 |
+
"learning_rate": 6.698729810778065e-07,
|
| 1401 |
+
"loss": 0.7013,
|
| 1402 |
+
"step": 199
|
| 1403 |
+
},
|
| 1404 |
+
{
|
| 1405 |
+
"epoch": 2.55863539445629,
|
| 1406 |
+
"grad_norm": 0.3055112063884735,
|
| 1407 |
+
"learning_rate": 6.329589969143518e-07,
|
| 1408 |
+
"loss": 0.6536,
|
| 1409 |
+
"step": 200
|
| 1410 |
+
},
|
| 1411 |
+
{
|
| 1412 |
+
"epoch": 2.571428571428571,
|
| 1413 |
+
"grad_norm": 0.3426555097103119,
|
| 1414 |
+
"learning_rate": 5.9702234071631e-07,
|
| 1415 |
+
"loss": 0.7274,
|
| 1416 |
+
"step": 201
|
| 1417 |
+
},
|
| 1418 |
+
{
|
| 1419 |
+
"epoch": 2.5842217484008527,
|
| 1420 |
+
"grad_norm": 0.3432694673538208,
|
| 1421 |
+
"learning_rate": 5.620710549772295e-07,
|
| 1422 |
+
"loss": 0.6523,
|
| 1423 |
+
"step": 202
|
| 1424 |
+
},
|
| 1425 |
+
{
|
| 1426 |
+
"epoch": 2.5970149253731343,
|
| 1427 |
+
"grad_norm": 0.36352601647377014,
|
| 1428 |
+
"learning_rate": 5.281129616683167e-07,
|
| 1429 |
+
"loss": 0.7254,
|
| 1430 |
+
"step": 203
|
| 1431 |
+
},
|
| 1432 |
+
{
|
| 1433 |
+
"epoch": 2.609808102345416,
|
| 1434 |
+
"grad_norm": 0.3131335973739624,
|
| 1435 |
+
"learning_rate": 4.951556604879049e-07,
|
| 1436 |
+
"loss": 0.6832,
|
| 1437 |
+
"step": 204
|
| 1438 |
+
},
|
| 1439 |
+
{
|
| 1440 |
+
"epoch": 2.6226012793176974,
|
| 1441 |
+
"grad_norm": 0.3547823429107666,
|
| 1442 |
+
"learning_rate": 4.632065271606756e-07,
|
| 1443 |
+
"loss": 0.7226,
|
| 1444 |
+
"step": 205
|
| 1445 |
+
},
|
| 1446 |
+
{
|
| 1447 |
+
"epoch": 2.635394456289979,
|
| 1448 |
+
"grad_norm": 0.2958380877971649,
|
| 1449 |
+
"learning_rate": 4.322727117869951e-07,
|
| 1450 |
+
"loss": 0.6507,
|
| 1451 |
+
"step": 206
|
| 1452 |
+
},
|
| 1453 |
+
{
|
| 1454 |
+
"epoch": 2.64818763326226,
|
| 1455 |
+
"grad_norm": 0.2793337404727936,
|
| 1456 |
+
"learning_rate": 4.0236113724274716e-07,
|
| 1457 |
+
"loss": 0.6371,
|
| 1458 |
+
"step": 207
|
| 1459 |
+
},
|
| 1460 |
+
{
|
| 1461 |
+
"epoch": 2.6609808102345416,
|
| 1462 |
+
"grad_norm": 0.33492887020111084,
|
| 1463 |
+
"learning_rate": 3.734784976300165e-07,
|
| 1464 |
+
"loss": 0.7792,
|
| 1465 |
+
"step": 208
|
| 1466 |
+
},
|
| 1467 |
+
{
|
| 1468 |
+
"epoch": 2.673773987206823,
|
| 1469 |
+
"grad_norm": 0.3088694214820862,
|
| 1470 |
+
"learning_rate": 3.4563125677897936e-07,
|
| 1471 |
+
"loss": 0.5887,
|
| 1472 |
+
"step": 209
|
| 1473 |
+
},
|
| 1474 |
+
{
|
| 1475 |
+
"epoch": 2.6865671641791042,
|
| 1476 |
+
"grad_norm": 0.3331063389778137,
|
| 1477 |
+
"learning_rate": 3.18825646801314e-07,
|
| 1478 |
+
"loss": 0.6793,
|
| 1479 |
+
"step": 210
|
| 1480 |
+
},
|
| 1481 |
+
{
|
| 1482 |
+
"epoch": 2.699360341151386,
|
| 1483 |
+
"grad_norm": 0.3037247359752655,
|
| 1484 |
+
"learning_rate": 2.930676666954846e-07,
|
| 1485 |
+
"loss": 0.6553,
|
| 1486 |
+
"step": 211
|
| 1487 |
+
},
|
| 1488 |
+
{
|
| 1489 |
+
"epoch": 2.7121535181236673,
|
| 1490 |
+
"grad_norm": 0.3542158305644989,
|
| 1491 |
+
"learning_rate": 2.6836308100417874e-07,
|
| 1492 |
+
"loss": 0.7101,
|
| 1493 |
+
"step": 212
|
| 1494 |
+
},
|
| 1495 |
+
{
|
| 1496 |
+
"epoch": 2.724946695095949,
|
| 1497 |
+
"grad_norm": 0.3504466116428375,
|
| 1498 |
+
"learning_rate": 2.447174185242324e-07,
|
| 1499 |
+
"loss": 0.7189,
|
| 1500 |
+
"step": 213
|
| 1501 |
+
},
|
| 1502 |
+
{
|
| 1503 |
+
"epoch": 2.7377398720682304,
|
| 1504 |
+
"grad_norm": 0.30136266350746155,
|
| 1505 |
+
"learning_rate": 2.2213597106929608e-07,
|
| 1506 |
+
"loss": 0.6076,
|
| 1507 |
+
"step": 214
|
| 1508 |
+
},
|
| 1509 |
+
{
|
| 1510 |
+
"epoch": 2.750533049040512,
|
| 1511 |
+
"grad_norm": 0.3250851631164551,
|
| 1512 |
+
"learning_rate": 2.006237922855553e-07,
|
| 1513 |
+
"loss": 0.7334,
|
| 1514 |
+
"step": 215
|
| 1515 |
+
},
|
| 1516 |
+
{
|
| 1517 |
+
"epoch": 2.763326226012793,
|
| 1518 |
+
"grad_norm": 0.3266373574733734,
|
| 1519 |
+
"learning_rate": 1.801856965207338e-07,
|
| 1520 |
+
"loss": 0.7144,
|
| 1521 |
+
"step": 216
|
| 1522 |
+
},
|
| 1523 |
+
{
|
| 1524 |
+
"epoch": 2.7761194029850746,
|
| 1525 |
+
"grad_norm": 0.2980578541755676,
|
| 1526 |
+
"learning_rate": 1.6082625774666793e-07,
|
| 1527 |
+
"loss": 0.6139,
|
| 1528 |
+
"step": 217
|
| 1529 |
+
},
|
| 1530 |
+
{
|
| 1531 |
+
"epoch": 2.788912579957356,
|
| 1532 |
+
"grad_norm": 0.31971806287765503,
|
| 1533 |
+
"learning_rate": 1.4254980853566248e-07,
|
| 1534 |
+
"loss": 0.7111,
|
| 1535 |
+
"step": 218
|
| 1536 |
+
},
|
| 1537 |
+
{
|
| 1538 |
+
"epoch": 2.8017057569296373,
|
| 1539 |
+
"grad_norm": 0.30652180314064026,
|
| 1540 |
+
"learning_rate": 1.253604390908819e-07,
|
| 1541 |
+
"loss": 0.6719,
|
| 1542 |
+
"step": 219
|
| 1543 |
+
},
|
| 1544 |
+
{
|
| 1545 |
+
"epoch": 2.814498933901919,
|
| 1546 |
+
"grad_norm": 0.32493188977241516,
|
| 1547 |
+
"learning_rate": 1.0926199633097156e-07,
|
| 1548 |
+
"loss": 0.7834,
|
| 1549 |
+
"step": 220
|
| 1550 |
+
},
|
| 1551 |
+
{
|
| 1552 |
+
"epoch": 2.8272921108742004,
|
| 1553 |
+
"grad_norm": 0.30289193987846375,
|
| 1554 |
+
"learning_rate": 9.42580830291373e-08,
|
| 1555 |
+
"loss": 0.6486,
|
| 1556 |
+
"step": 221
|
| 1557 |
+
},
|
| 1558 |
+
{
|
| 1559 |
+
"epoch": 2.840085287846482,
|
| 1560 |
+
"grad_norm": 0.39349403977394104,
|
| 1561 |
+
"learning_rate": 8.035205700685167e-08,
|
| 1562 |
+
"loss": 0.6523,
|
| 1563 |
+
"step": 222
|
| 1564 |
+
},
|
| 1565 |
+
{
|
| 1566 |
+
"epoch": 2.8528784648187635,
|
| 1567 |
+
"grad_norm": 0.32352349162101746,
|
| 1568 |
+
"learning_rate": 6.75470303823933e-08,
|
| 1569 |
+
"loss": 0.708,
|
| 1570 |
+
"step": 223
|
| 1571 |
+
},
|
| 1572 |
+
{
|
| 1573 |
+
"epoch": 2.8656716417910446,
|
| 1574 |
+
"grad_norm": 0.31589657068252563,
|
| 1575 |
+
"learning_rate": 5.584586887435739e-08,
|
| 1576 |
+
"loss": 0.7025,
|
| 1577 |
+
"step": 224
|
| 1578 |
+
},
|
| 1579 |
+
{
|
| 1580 |
+
"epoch": 2.878464818763326,
|
| 1581 |
+
"grad_norm": 0.3233773708343506,
|
| 1582 |
+
"learning_rate": 4.52511911603265e-08,
|
| 1583 |
+
"loss": 0.7003,
|
| 1584 |
+
"step": 225
|
| 1585 |
+
},
|
| 1586 |
+
{
|
| 1587 |
+
"epoch": 2.8912579957356077,
|
| 1588 |
+
"grad_norm": 0.3037286400794983,
|
| 1589 |
+
"learning_rate": 3.576536829081323e-08,
|
| 1590 |
+
"loss": 0.7252,
|
| 1591 |
+
"step": 226
|
| 1592 |
+
},
|
| 1593 |
+
{
|
| 1594 |
+
"epoch": 2.9040511727078893,
|
| 1595 |
+
"grad_norm": 0.3222990334033966,
|
| 1596 |
+
"learning_rate": 2.7390523158633552e-08,
|
| 1597 |
+
"loss": 0.7079,
|
| 1598 |
+
"step": 227
|
| 1599 |
+
},
|
| 1600 |
+
{
|
| 1601 |
+
"epoch": 2.9168443496801704,
|
| 1602 |
+
"grad_norm": 0.30910524725914,
|
| 1603 |
+
"learning_rate": 2.012853002380466e-08,
|
| 1604 |
+
"loss": 0.6701,
|
| 1605 |
+
"step": 228
|
| 1606 |
+
},
|
| 1607 |
+
{
|
| 1608 |
+
"epoch": 2.929637526652452,
|
| 1609 |
+
"grad_norm": 0.29240769147872925,
|
| 1610 |
+
"learning_rate": 1.3981014094099354e-08,
|
| 1611 |
+
"loss": 0.6388,
|
| 1612 |
+
"step": 229
|
| 1613 |
+
},
|
| 1614 |
+
{
|
| 1615 |
+
"epoch": 2.9424307036247335,
|
| 1616 |
+
"grad_norm": 0.34161651134490967,
|
| 1617 |
+
"learning_rate": 8.949351161324227e-09,
|
| 1618 |
+
"loss": 0.7613,
|
| 1619 |
+
"step": 230
|
| 1620 |
+
},
|
| 1621 |
+
{
|
| 1622 |
+
"epoch": 2.955223880597015,
|
| 1623 |
+
"grad_norm": 0.299943745136261,
|
| 1624 |
+
"learning_rate": 5.034667293427053e-09,
|
| 1625 |
+
"loss": 0.6471,
|
| 1626 |
+
"step": 231
|
| 1627 |
+
},
|
| 1628 |
+
{
|
| 1629 |
+
"epoch": 2.9680170575692966,
|
| 1630 |
+
"grad_norm": 0.3515676259994507,
|
| 1631 |
+
"learning_rate": 2.237838582483387e-09,
|
| 1632 |
+
"loss": 0.7412,
|
| 1633 |
+
"step": 232
|
| 1634 |
+
},
|
| 1635 |
+
{
|
| 1636 |
+
"epoch": 2.9808102345415777,
|
| 1637 |
+
"grad_norm": 0.3143077790737152,
|
| 1638 |
+
"learning_rate": 5.594909486328348e-10,
|
| 1639 |
+
"loss": 0.6572,
|
| 1640 |
+
"step": 233
|
| 1641 |
+
},
|
| 1642 |
+
{
|
| 1643 |
+
"epoch": 2.9936034115138592,
|
| 1644 |
+
"grad_norm": 0.3188575506210327,
|
| 1645 |
+
"learning_rate": 0.0,
|
| 1646 |
+
"loss": 0.7006,
|
| 1647 |
+
"step": 234
|
| 1648 |
+
},
|
| 1649 |
+
{
|
| 1650 |
+
"epoch": 2.9936034115138592,
|
| 1651 |
+
"step": 234,
|
| 1652 |
+
"total_flos": 327743678562304.0,
|
| 1653 |
+
"train_loss": 0.7533083028263516,
|
| 1654 |
+
"train_runtime": 9147.0121,
|
| 1655 |
+
"train_samples_per_second": 2.46,
|
| 1656 |
+
"train_steps_per_second": 0.026
|
| 1657 |
+
}
|
| 1658 |
+
],
|
| 1659 |
+
"logging_steps": 1.0,
|
| 1660 |
+
"max_steps": 234,
|
| 1661 |
+
"num_input_tokens_seen": 0,
|
| 1662 |
+
"num_train_epochs": 3,
|
| 1663 |
+
"save_steps": 500,
|
| 1664 |
+
"stateful_callbacks": {
|
| 1665 |
+
"TrainerControl": {
|
| 1666 |
+
"args": {
|
| 1667 |
+
"should_epoch_stop": false,
|
| 1668 |
+
"should_evaluate": false,
|
| 1669 |
+
"should_log": false,
|
| 1670 |
+
"should_save": true,
|
| 1671 |
+
"should_training_stop": true
|
| 1672 |
+
},
|
| 1673 |
+
"attributes": {}
|
| 1674 |
+
}
|
| 1675 |
+
},
|
| 1676 |
+
"total_flos": 327743678562304.0,
|
| 1677 |
+
"train_batch_size": 1,
|
| 1678 |
+
"trial_name": null,
|
| 1679 |
+
"trial_params": null
|
| 1680 |
+
}
|
training_loss.png
ADDED
|