End of training
Browse files- README.md +2 -1
- all_results.json +8 -0
- train_results.json +8 -0
- trainer_state.json +1218 -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: seed_math_multiple_samples_scale_up_scaredy_cat_test
|
|
@@ -15,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 15 |
|
| 16 |
# seed_math_multiple_samples_scale_up_scaredy_cat_test
|
| 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: seed_math_multiple_samples_scale_up_scaredy_cat_test
|
|
|
|
| 16 |
|
| 17 |
# seed_math_multiple_samples_scale_up_scaredy_cat_test
|
| 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/seed_math_multiple_samples_scale_up_scaredy_cat_test dataset.
|
| 20 |
|
| 21 |
## Model description
|
| 22 |
|
all_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 2.986666666666667,
|
| 3 |
+
"total_flos": 139896582144000.0,
|
| 4 |
+
"train_loss": 0.7518468662386849,
|
| 5 |
+
"train_runtime": 7945.17,
|
| 6 |
+
"train_samples_per_second": 2.039,
|
| 7 |
+
"train_steps_per_second": 0.021
|
| 8 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 2.986666666666667,
|
| 3 |
+
"total_flos": 139896582144000.0,
|
| 4 |
+
"train_loss": 0.7518468662386849,
|
| 5 |
+
"train_runtime": 7945.17,
|
| 6 |
+
"train_samples_per_second": 2.039,
|
| 7 |
+
"train_steps_per_second": 0.021
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1218 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 2.986666666666667,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 168,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.017777777777777778,
|
| 13 |
+
"grad_norm": 6.802438735961914,
|
| 14 |
+
"learning_rate": 5.882352941176471e-07,
|
| 15 |
+
"loss": 1.1138,
|
| 16 |
+
"step": 1
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.035555555555555556,
|
| 20 |
+
"grad_norm": 6.787415981292725,
|
| 21 |
+
"learning_rate": 1.1764705882352942e-06,
|
| 22 |
+
"loss": 1.0806,
|
| 23 |
+
"step": 2
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.05333333333333334,
|
| 27 |
+
"grad_norm": 6.674739360809326,
|
| 28 |
+
"learning_rate": 1.7647058823529414e-06,
|
| 29 |
+
"loss": 1.0469,
|
| 30 |
+
"step": 3
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.07111111111111111,
|
| 34 |
+
"grad_norm": 6.719709873199463,
|
| 35 |
+
"learning_rate": 2.3529411764705885e-06,
|
| 36 |
+
"loss": 1.0525,
|
| 37 |
+
"step": 4
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.08888888888888889,
|
| 41 |
+
"grad_norm": 6.4609293937683105,
|
| 42 |
+
"learning_rate": 2.9411764705882355e-06,
|
| 43 |
+
"loss": 1.1458,
|
| 44 |
+
"step": 5
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.10666666666666667,
|
| 48 |
+
"grad_norm": 5.147739887237549,
|
| 49 |
+
"learning_rate": 3.529411764705883e-06,
|
| 50 |
+
"loss": 1.0836,
|
| 51 |
+
"step": 6
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.12444444444444444,
|
| 55 |
+
"grad_norm": 3.1252267360687256,
|
| 56 |
+
"learning_rate": 4.11764705882353e-06,
|
| 57 |
+
"loss": 0.9522,
|
| 58 |
+
"step": 7
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.14222222222222222,
|
| 62 |
+
"grad_norm": 2.836021661758423,
|
| 63 |
+
"learning_rate": 4.705882352941177e-06,
|
| 64 |
+
"loss": 0.9805,
|
| 65 |
+
"step": 8
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.16,
|
| 69 |
+
"grad_norm": 4.235110759735107,
|
| 70 |
+
"learning_rate": 5.294117647058824e-06,
|
| 71 |
+
"loss": 0.9539,
|
| 72 |
+
"step": 9
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.17777777777777778,
|
| 76 |
+
"grad_norm": 4.741976261138916,
|
| 77 |
+
"learning_rate": 5.882352941176471e-06,
|
| 78 |
+
"loss": 0.9584,
|
| 79 |
+
"step": 10
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.19555555555555557,
|
| 83 |
+
"grad_norm": 4.572456359863281,
|
| 84 |
+
"learning_rate": 6.470588235294119e-06,
|
| 85 |
+
"loss": 0.9436,
|
| 86 |
+
"step": 11
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.21333333333333335,
|
| 90 |
+
"grad_norm": 3.5882203578948975,
|
| 91 |
+
"learning_rate": 7.058823529411766e-06,
|
| 92 |
+
"loss": 0.9193,
|
| 93 |
+
"step": 12
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.2311111111111111,
|
| 97 |
+
"grad_norm": 3.677826404571533,
|
| 98 |
+
"learning_rate": 7.647058823529411e-06,
|
| 99 |
+
"loss": 0.9732,
|
| 100 |
+
"step": 13
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.24888888888888888,
|
| 104 |
+
"grad_norm": 2.6755166053771973,
|
| 105 |
+
"learning_rate": 8.23529411764706e-06,
|
| 106 |
+
"loss": 0.8779,
|
| 107 |
+
"step": 14
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.26666666666666666,
|
| 111 |
+
"grad_norm": 2.1610584259033203,
|
| 112 |
+
"learning_rate": 8.823529411764707e-06,
|
| 113 |
+
"loss": 0.8581,
|
| 114 |
+
"step": 15
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.28444444444444444,
|
| 118 |
+
"grad_norm": 1.9563807249069214,
|
| 119 |
+
"learning_rate": 9.411764705882354e-06,
|
| 120 |
+
"loss": 0.8518,
|
| 121 |
+
"step": 16
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.3022222222222222,
|
| 125 |
+
"grad_norm": 2.030823230743408,
|
| 126 |
+
"learning_rate": 1e-05,
|
| 127 |
+
"loss": 0.7811,
|
| 128 |
+
"step": 17
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.32,
|
| 132 |
+
"grad_norm": 1.7580914497375488,
|
| 133 |
+
"learning_rate": 9.998917893031615e-06,
|
| 134 |
+
"loss": 0.8713,
|
| 135 |
+
"step": 18
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.3377777777777778,
|
| 139 |
+
"grad_norm": 1.7509064674377441,
|
| 140 |
+
"learning_rate": 9.995672040508656e-06,
|
| 141 |
+
"loss": 0.85,
|
| 142 |
+
"step": 19
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.35555555555555557,
|
| 146 |
+
"grad_norm": 1.547536015510559,
|
| 147 |
+
"learning_rate": 9.990263847374976e-06,
|
| 148 |
+
"loss": 0.8085,
|
| 149 |
+
"step": 20
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.37333333333333335,
|
| 153 |
+
"grad_norm": 1.1073737144470215,
|
| 154 |
+
"learning_rate": 9.982695654527966e-06,
|
| 155 |
+
"loss": 0.7902,
|
| 156 |
+
"step": 21
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.39111111111111113,
|
| 160 |
+
"grad_norm": 0.9871188998222351,
|
| 161 |
+
"learning_rate": 9.972970737805312e-06,
|
| 162 |
+
"loss": 0.8036,
|
| 163 |
+
"step": 22
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.4088888888888889,
|
| 167 |
+
"grad_norm": 1.1322752237319946,
|
| 168 |
+
"learning_rate": 9.961093306567076e-06,
|
| 169 |
+
"loss": 0.8417,
|
| 170 |
+
"step": 23
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.4266666666666667,
|
| 174 |
+
"grad_norm": 0.8192468881607056,
|
| 175 |
+
"learning_rate": 9.947068501873702e-06,
|
| 176 |
+
"loss": 0.794,
|
| 177 |
+
"step": 24
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.4444444444444444,
|
| 181 |
+
"grad_norm": 0.8386516571044922,
|
| 182 |
+
"learning_rate": 9.930902394260746e-06,
|
| 183 |
+
"loss": 0.767,
|
| 184 |
+
"step": 25
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.4622222222222222,
|
| 188 |
+
"grad_norm": 0.959214448928833,
|
| 189 |
+
"learning_rate": 9.912601981111287e-06,
|
| 190 |
+
"loss": 0.8015,
|
| 191 |
+
"step": 26
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.48,
|
| 195 |
+
"grad_norm": 0.7772365808486938,
|
| 196 |
+
"learning_rate": 9.892175183627161e-06,
|
| 197 |
+
"loss": 0.764,
|
| 198 |
+
"step": 27
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.49777777777777776,
|
| 202 |
+
"grad_norm": 0.831992506980896,
|
| 203 |
+
"learning_rate": 9.869630843400331e-06,
|
| 204 |
+
"loss": 0.8278,
|
| 205 |
+
"step": 28
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.5155555555555555,
|
| 209 |
+
"grad_norm": 0.7600440979003906,
|
| 210 |
+
"learning_rate": 9.844978718585855e-06,
|
| 211 |
+
"loss": 0.7788,
|
| 212 |
+
"step": 29
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.5333333333333333,
|
| 216 |
+
"grad_norm": 0.7171022891998291,
|
| 217 |
+
"learning_rate": 9.81822947967816e-06,
|
| 218 |
+
"loss": 0.7713,
|
| 219 |
+
"step": 30
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.5511111111111111,
|
| 223 |
+
"grad_norm": 0.8401284217834473,
|
| 224 |
+
"learning_rate": 9.789394704892364e-06,
|
| 225 |
+
"loss": 0.8337,
|
| 226 |
+
"step": 31
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.5688888888888889,
|
| 230 |
+
"grad_norm": 0.6657397150993347,
|
| 231 |
+
"learning_rate": 9.758486875152766e-06,
|
| 232 |
+
"loss": 0.8159,
|
| 233 |
+
"step": 32
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.5866666666666667,
|
| 237 |
+
"grad_norm": 0.7162612676620483,
|
| 238 |
+
"learning_rate": 9.725519368690539e-06,
|
| 239 |
+
"loss": 0.8484,
|
| 240 |
+
"step": 33
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.6044444444444445,
|
| 244 |
+
"grad_norm": 0.6413375735282898,
|
| 245 |
+
"learning_rate": 9.690506455253073e-06,
|
| 246 |
+
"loss": 0.7798,
|
| 247 |
+
"step": 34
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.6222222222222222,
|
| 251 |
+
"grad_norm": 0.641582190990448,
|
| 252 |
+
"learning_rate": 9.65346328992741e-06,
|
| 253 |
+
"loss": 0.8282,
|
| 254 |
+
"step": 35
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.64,
|
| 258 |
+
"grad_norm": 0.6429569721221924,
|
| 259 |
+
"learning_rate": 9.614405906580486e-06,
|
| 260 |
+
"loss": 0.7737,
|
| 261 |
+
"step": 36
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.6577777777777778,
|
| 265 |
+
"grad_norm": 0.5829933285713196,
|
| 266 |
+
"learning_rate": 9.573351210918976e-06,
|
| 267 |
+
"loss": 0.8027,
|
| 268 |
+
"step": 37
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.6755555555555556,
|
| 272 |
+
"grad_norm": 0.5490368008613586,
|
| 273 |
+
"learning_rate": 9.53031697317178e-06,
|
| 274 |
+
"loss": 0.7706,
|
| 275 |
+
"step": 38
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.6933333333333334,
|
| 279 |
+
"grad_norm": 0.5687077045440674,
|
| 280 |
+
"learning_rate": 9.485321820398321e-06,
|
| 281 |
+
"loss": 0.8054,
|
| 282 |
+
"step": 39
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.7111111111111111,
|
| 286 |
+
"grad_norm": 0.5813000798225403,
|
| 287 |
+
"learning_rate": 9.43838522842594e-06,
|
| 288 |
+
"loss": 0.7565,
|
| 289 |
+
"step": 40
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.7288888888888889,
|
| 293 |
+
"grad_norm": 0.558438777923584,
|
| 294 |
+
"learning_rate": 9.389527513419935e-06,
|
| 295 |
+
"loss": 0.7622,
|
| 296 |
+
"step": 41
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.7466666666666667,
|
| 300 |
+
"grad_norm": 0.6196303367614746,
|
| 301 |
+
"learning_rate": 9.338769823089853e-06,
|
| 302 |
+
"loss": 0.7578,
|
| 303 |
+
"step": 42
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.7644444444444445,
|
| 307 |
+
"grad_norm": 0.658560037612915,
|
| 308 |
+
"learning_rate": 9.286134127535859e-06,
|
| 309 |
+
"loss": 0.7589,
|
| 310 |
+
"step": 43
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.7822222222222223,
|
| 314 |
+
"grad_norm": 0.6177524328231812,
|
| 315 |
+
"learning_rate": 9.231643209739128e-06,
|
| 316 |
+
"loss": 0.7243,
|
| 317 |
+
"step": 44
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.8,
|
| 321 |
+
"grad_norm": 0.5707345604896545,
|
| 322 |
+
"learning_rate": 9.175320655700407e-06,
|
| 323 |
+
"loss": 0.7538,
|
| 324 |
+
"step": 45
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.8177777777777778,
|
| 328 |
+
"grad_norm": 0.5330144762992859,
|
| 329 |
+
"learning_rate": 9.117190844230971e-06,
|
| 330 |
+
"loss": 0.7676,
|
| 331 |
+
"step": 46
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.8355555555555556,
|
| 335 |
+
"grad_norm": 0.6449193358421326,
|
| 336 |
+
"learning_rate": 9.057278936400453e-06,
|
| 337 |
+
"loss": 0.7789,
|
| 338 |
+
"step": 47
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.8533333333333334,
|
| 342 |
+
"grad_norm": 0.5758365988731384,
|
| 343 |
+
"learning_rate": 8.99561086464603e-06,
|
| 344 |
+
"loss": 0.7404,
|
| 345 |
+
"step": 48
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.8711111111111111,
|
| 349 |
+
"grad_norm": 0.5622390508651733,
|
| 350 |
+
"learning_rate": 8.932213321547769e-06,
|
| 351 |
+
"loss": 0.7493,
|
| 352 |
+
"step": 49
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.8888888888888888,
|
| 356 |
+
"grad_norm": 0.6213387846946716,
|
| 357 |
+
"learning_rate": 8.86711374827494e-06,
|
| 358 |
+
"loss": 0.7431,
|
| 359 |
+
"step": 50
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.9066666666666666,
|
| 363 |
+
"grad_norm": 0.5265310406684875,
|
| 364 |
+
"learning_rate": 8.800340322708291e-06,
|
| 365 |
+
"loss": 0.7748,
|
| 366 |
+
"step": 51
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.9244444444444444,
|
| 370 |
+
"grad_norm": 0.6248610019683838,
|
| 371 |
+
"learning_rate": 8.73192194724347e-06,
|
| 372 |
+
"loss": 0.756,
|
| 373 |
+
"step": 52
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.9422222222222222,
|
| 377 |
+
"grad_norm": 0.5746708512306213,
|
| 378 |
+
"learning_rate": 8.661888236280813e-06,
|
| 379 |
+
"loss": 0.7826,
|
| 380 |
+
"step": 53
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.96,
|
| 384 |
+
"grad_norm": 0.6057078242301941,
|
| 385 |
+
"learning_rate": 8.590269503406986e-06,
|
| 386 |
+
"loss": 0.7477,
|
| 387 |
+
"step": 54
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.9777777777777777,
|
| 391 |
+
"grad_norm": 0.6079862117767334,
|
| 392 |
+
"learning_rate": 8.517096748273951e-06,
|
| 393 |
+
"loss": 0.7411,
|
| 394 |
+
"step": 55
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.9955555555555555,
|
| 398 |
+
"grad_norm": 0.6125847697257996,
|
| 399 |
+
"learning_rate": 8.442401643181e-06,
|
| 400 |
+
"loss": 0.7707,
|
| 401 |
+
"step": 56
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 1.0133333333333334,
|
| 405 |
+
"grad_norm": 1.0918083190917969,
|
| 406 |
+
"learning_rate": 8.366216519365623e-06,
|
| 407 |
+
"loss": 1.2307,
|
| 408 |
+
"step": 57
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 1.031111111111111,
|
| 412 |
+
"grad_norm": 0.6814342141151428,
|
| 413 |
+
"learning_rate": 8.288574353009164e-06,
|
| 414 |
+
"loss": 0.6984,
|
| 415 |
+
"step": 58
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 1.048888888888889,
|
| 419 |
+
"grad_norm": 0.7980371713638306,
|
| 420 |
+
"learning_rate": 8.20950875096333e-06,
|
| 421 |
+
"loss": 0.8081,
|
| 422 |
+
"step": 59
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 1.0666666666666667,
|
| 426 |
+
"grad_norm": 0.7088991403579712,
|
| 427 |
+
"learning_rate": 8.129053936203688e-06,
|
| 428 |
+
"loss": 0.7388,
|
| 429 |
+
"step": 60
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 1.0844444444444445,
|
| 433 |
+
"grad_norm": 0.6209843754768372,
|
| 434 |
+
"learning_rate": 8.04724473301652e-06,
|
| 435 |
+
"loss": 0.8232,
|
| 436 |
+
"step": 61
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 1.1022222222222222,
|
| 440 |
+
"grad_norm": 0.5992409586906433,
|
| 441 |
+
"learning_rate": 7.964116551925365e-06,
|
| 442 |
+
"loss": 0.6424,
|
| 443 |
+
"step": 62
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 1.12,
|
| 447 |
+
"grad_norm": 0.8191760778427124,
|
| 448 |
+
"learning_rate": 7.879705374363831e-06,
|
| 449 |
+
"loss": 0.743,
|
| 450 |
+
"step": 63
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 1.1377777777777778,
|
| 454 |
+
"grad_norm": 0.5577734112739563,
|
| 455 |
+
"learning_rate": 7.794047737101298e-06,
|
| 456 |
+
"loss": 0.6329,
|
| 457 |
+
"step": 64
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 1.1555555555555554,
|
| 461 |
+
"grad_norm": 0.590572714805603,
|
| 462 |
+
"learning_rate": 7.707180716428237e-06,
|
| 463 |
+
"loss": 0.7648,
|
| 464 |
+
"step": 65
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 1.1733333333333333,
|
| 468 |
+
"grad_norm": 0.7350046634674072,
|
| 469 |
+
"learning_rate": 7.619141912108008e-06,
|
| 470 |
+
"loss": 0.9037,
|
| 471 |
+
"step": 66
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 1.1911111111111112,
|
| 475 |
+
"grad_norm": 0.5404512286186218,
|
| 476 |
+
"learning_rate": 7.529969431102063e-06,
|
| 477 |
+
"loss": 0.7325,
|
| 478 |
+
"step": 67
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 1.208888888888889,
|
| 482 |
+
"grad_norm": 0.6300895810127258,
|
| 483 |
+
"learning_rate": 7.4397018710756415e-06,
|
| 484 |
+
"loss": 0.6517,
|
| 485 |
+
"step": 68
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 1.2266666666666666,
|
| 489 |
+
"grad_norm": 0.5601981282234192,
|
| 490 |
+
"learning_rate": 7.34837830369103e-06,
|
| 491 |
+
"loss": 0.6863,
|
| 492 |
+
"step": 69
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 1.2444444444444445,
|
| 496 |
+
"grad_norm": 0.6357449293136597,
|
| 497 |
+
"learning_rate": 7.2560382576956875e-06,
|
| 498 |
+
"loss": 0.7334,
|
| 499 |
+
"step": 70
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 1.2622222222222224,
|
| 503 |
+
"grad_norm": 0.5976100564002991,
|
| 504 |
+
"learning_rate": 7.162721701812506e-06,
|
| 505 |
+
"loss": 0.6825,
|
| 506 |
+
"step": 71
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 1.28,
|
| 510 |
+
"grad_norm": 0.5376277565956116,
|
| 511 |
+
"learning_rate": 7.068469027439642e-06,
|
| 512 |
+
"loss": 0.7138,
|
| 513 |
+
"step": 72
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 1.2977777777777777,
|
| 517 |
+
"grad_norm": 0.5158504843711853,
|
| 518 |
+
"learning_rate": 6.9733210311673826e-06,
|
| 519 |
+
"loss": 0.7004,
|
| 520 |
+
"step": 73
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 1.3155555555555556,
|
| 524 |
+
"grad_norm": 0.6006451845169067,
|
| 525 |
+
"learning_rate": 6.8773188971196515e-06,
|
| 526 |
+
"loss": 0.6926,
|
| 527 |
+
"step": 74
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 1.3333333333333333,
|
| 531 |
+
"grad_norm": 0.5877394080162048,
|
| 532 |
+
"learning_rate": 6.780504179127735e-06,
|
| 533 |
+
"loss": 0.7251,
|
| 534 |
+
"step": 75
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 1.3511111111111112,
|
| 538 |
+
"grad_norm": 0.47379350662231445,
|
| 539 |
+
"learning_rate": 6.682918782744033e-06,
|
| 540 |
+
"loss": 0.6315,
|
| 541 |
+
"step": 76
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 1.3688888888888888,
|
| 545 |
+
"grad_norm": 0.6031640768051147,
|
| 546 |
+
"learning_rate": 6.584604947103515e-06,
|
| 547 |
+
"loss": 0.8246,
|
| 548 |
+
"step": 77
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 1.3866666666666667,
|
| 552 |
+
"grad_norm": 0.5239250659942627,
|
| 553 |
+
"learning_rate": 6.4856052266408375e-06,
|
| 554 |
+
"loss": 0.6918,
|
| 555 |
+
"step": 78
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 1.4044444444444444,
|
| 559 |
+
"grad_norm": 0.5310869216918945,
|
| 560 |
+
"learning_rate": 6.385962472670953e-06,
|
| 561 |
+
"loss": 0.7386,
|
| 562 |
+
"step": 79
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 1.4222222222222223,
|
| 566 |
+
"grad_norm": 0.49420294165611267,
|
| 567 |
+
"learning_rate": 6.28571981484123e-06,
|
| 568 |
+
"loss": 0.6715,
|
| 569 |
+
"step": 80
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 1.44,
|
| 573 |
+
"grad_norm": 0.5260173082351685,
|
| 574 |
+
"learning_rate": 6.184920642463095e-06,
|
| 575 |
+
"loss": 0.662,
|
| 576 |
+
"step": 81
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 1.4577777777777778,
|
| 580 |
+
"grad_norm": 0.5475410223007202,
|
| 581 |
+
"learning_rate": 6.083608585731283e-06,
|
| 582 |
+
"loss": 0.7322,
|
| 583 |
+
"step": 82
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 1.4755555555555555,
|
| 587 |
+
"grad_norm": 0.5325695276260376,
|
| 588 |
+
"learning_rate": 5.9818274968388225e-06,
|
| 589 |
+
"loss": 0.6793,
|
| 590 |
+
"step": 83
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 1.4933333333333334,
|
| 594 |
+
"grad_norm": 0.49327635765075684,
|
| 595 |
+
"learning_rate": 5.879621430995927e-06,
|
| 596 |
+
"loss": 0.6983,
|
| 597 |
+
"step": 84
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 1.511111111111111,
|
| 601 |
+
"grad_norm": 0.5579730272293091,
|
| 602 |
+
"learning_rate": 5.777034627361025e-06,
|
| 603 |
+
"loss": 0.7993,
|
| 604 |
+
"step": 85
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 1.528888888888889,
|
| 608 |
+
"grad_norm": 0.5502652525901794,
|
| 609 |
+
"learning_rate": 5.674111489892144e-06,
|
| 610 |
+
"loss": 0.6955,
|
| 611 |
+
"step": 86
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 1.5466666666666666,
|
| 615 |
+
"grad_norm": 0.5440042018890381,
|
| 616 |
+
"learning_rate": 5.570896568126994e-06,
|
| 617 |
+
"loss": 0.694,
|
| 618 |
+
"step": 87
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 1.5644444444444443,
|
| 622 |
+
"grad_norm": 0.473174124956131,
|
| 623 |
+
"learning_rate": 5.4674345379e-06,
|
| 624 |
+
"loss": 0.6934,
|
| 625 |
+
"step": 88
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 1.5822222222222222,
|
| 629 |
+
"grad_norm": 0.492448627948761,
|
| 630 |
+
"learning_rate": 5.36377018200472e-06,
|
| 631 |
+
"loss": 0.7066,
|
| 632 |
+
"step": 89
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 1.6,
|
| 636 |
+
"grad_norm": 0.5301864147186279,
|
| 637 |
+
"learning_rate": 5.259948370809902e-06,
|
| 638 |
+
"loss": 0.608,
|
| 639 |
+
"step": 90
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 1.6177777777777778,
|
| 643 |
+
"grad_norm": 0.5350763201713562,
|
| 644 |
+
"learning_rate": 5.156014042837696e-06,
|
| 645 |
+
"loss": 0.762,
|
| 646 |
+
"step": 91
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 1.6355555555555554,
|
| 650 |
+
"grad_norm": 0.43970951437950134,
|
| 651 |
+
"learning_rate": 5.052012185312322e-06,
|
| 652 |
+
"loss": 0.6945,
|
| 653 |
+
"step": 92
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 1.6533333333333333,
|
| 657 |
+
"grad_norm": 0.5044556856155396,
|
| 658 |
+
"learning_rate": 4.94798781468768e-06,
|
| 659 |
+
"loss": 0.6902,
|
| 660 |
+
"step": 93
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 1.6711111111111112,
|
| 664 |
+
"grad_norm": 0.5784631967544556,
|
| 665 |
+
"learning_rate": 4.843985957162304e-06,
|
| 666 |
+
"loss": 0.8185,
|
| 667 |
+
"step": 94
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 1.6888888888888889,
|
| 671 |
+
"grad_norm": 0.470632404088974,
|
| 672 |
+
"learning_rate": 4.740051629190099e-06,
|
| 673 |
+
"loss": 0.6987,
|
| 674 |
+
"step": 95
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 1.7066666666666666,
|
| 678 |
+
"grad_norm": 0.43977099657058716,
|
| 679 |
+
"learning_rate": 4.636229817995281e-06,
|
| 680 |
+
"loss": 0.759,
|
| 681 |
+
"step": 96
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 1.7244444444444444,
|
| 685 |
+
"grad_norm": 0.4898114502429962,
|
| 686 |
+
"learning_rate": 4.532565462099999e-06,
|
| 687 |
+
"loss": 0.7005,
|
| 688 |
+
"step": 97
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 1.7422222222222223,
|
| 692 |
+
"grad_norm": 0.45880624651908875,
|
| 693 |
+
"learning_rate": 4.429103431873009e-06,
|
| 694 |
+
"loss": 0.6753,
|
| 695 |
+
"step": 98
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 1.76,
|
| 699 |
+
"grad_norm": 0.44095566868782043,
|
| 700 |
+
"learning_rate": 4.3258885101078565e-06,
|
| 701 |
+
"loss": 0.6962,
|
| 702 |
+
"step": 99
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 1.7777777777777777,
|
| 706 |
+
"grad_norm": 0.4869973063468933,
|
| 707 |
+
"learning_rate": 4.2229653726389765e-06,
|
| 708 |
+
"loss": 0.7709,
|
| 709 |
+
"step": 100
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 1.7955555555555556,
|
| 713 |
+
"grad_norm": 0.3903906047344208,
|
| 714 |
+
"learning_rate": 4.120378569004074e-06,
|
| 715 |
+
"loss": 0.6639,
|
| 716 |
+
"step": 101
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 1.8133333333333335,
|
| 720 |
+
"grad_norm": 0.42414912581443787,
|
| 721 |
+
"learning_rate": 4.018172503161179e-06,
|
| 722 |
+
"loss": 0.679,
|
| 723 |
+
"step": 102
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 1.8311111111111111,
|
| 727 |
+
"grad_norm": 0.5198123455047607,
|
| 728 |
+
"learning_rate": 3.9163914142687185e-06,
|
| 729 |
+
"loss": 0.7619,
|
| 730 |
+
"step": 103
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 1.8488888888888888,
|
| 734 |
+
"grad_norm": 0.4870392084121704,
|
| 735 |
+
"learning_rate": 3.815079357536907e-06,
|
| 736 |
+
"loss": 0.5876,
|
| 737 |
+
"step": 104
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 1.8666666666666667,
|
| 741 |
+
"grad_norm": 0.48693349957466125,
|
| 742 |
+
"learning_rate": 3.714280185158771e-06,
|
| 743 |
+
"loss": 0.7726,
|
| 744 |
+
"step": 105
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 1.8844444444444446,
|
| 748 |
+
"grad_norm": 0.4303606450557709,
|
| 749 |
+
"learning_rate": 3.614037527329048e-06,
|
| 750 |
+
"loss": 0.7155,
|
| 751 |
+
"step": 106
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 1.9022222222222223,
|
| 755 |
+
"grad_norm": 0.5268415808677673,
|
| 756 |
+
"learning_rate": 3.5143947733591633e-06,
|
| 757 |
+
"loss": 0.7175,
|
| 758 |
+
"step": 107
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 1.92,
|
| 762 |
+
"grad_norm": 0.4765200614929199,
|
| 763 |
+
"learning_rate": 3.4153950528964867e-06,
|
| 764 |
+
"loss": 0.6943,
|
| 765 |
+
"step": 108
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 1.9377777777777778,
|
| 769 |
+
"grad_norm": 0.42151421308517456,
|
| 770 |
+
"learning_rate": 3.3170812172559695e-06,
|
| 771 |
+
"loss": 0.7555,
|
| 772 |
+
"step": 109
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 1.9555555555555557,
|
| 776 |
+
"grad_norm": 0.426076740026474,
|
| 777 |
+
"learning_rate": 3.2194958208722656e-06,
|
| 778 |
+
"loss": 0.6852,
|
| 779 |
+
"step": 110
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 1.9733333333333334,
|
| 783 |
+
"grad_norm": 0.5378457307815552,
|
| 784 |
+
"learning_rate": 3.1226811028803514e-06,
|
| 785 |
+
"loss": 0.7171,
|
| 786 |
+
"step": 111
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 1.991111111111111,
|
| 790 |
+
"grad_norm": 0.39721420407295227,
|
| 791 |
+
"learning_rate": 3.0266789688326187e-06,
|
| 792 |
+
"loss": 0.6122,
|
| 793 |
+
"step": 112
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 2.008888888888889,
|
| 797 |
+
"grad_norm": 0.9009656310081482,
|
| 798 |
+
"learning_rate": 2.9315309725603596e-06,
|
| 799 |
+
"loss": 1.3572,
|
| 800 |
+
"step": 113
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 2.026666666666667,
|
| 804 |
+
"grad_norm": 0.4720599353313446,
|
| 805 |
+
"learning_rate": 2.8372782981874964e-06,
|
| 806 |
+
"loss": 0.7745,
|
| 807 |
+
"step": 114
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 2.0444444444444443,
|
| 811 |
+
"grad_norm": 0.3901655972003937,
|
| 812 |
+
"learning_rate": 2.7439617423043146e-06,
|
| 813 |
+
"loss": 0.6196,
|
| 814 |
+
"step": 115
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 2.062222222222222,
|
| 818 |
+
"grad_norm": 0.4585684537887573,
|
| 819 |
+
"learning_rate": 2.6516216963089698e-06,
|
| 820 |
+
"loss": 0.5975,
|
| 821 |
+
"step": 116
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 2.08,
|
| 825 |
+
"grad_norm": 0.44775474071502686,
|
| 826 |
+
"learning_rate": 2.560298128924358e-06,
|
| 827 |
+
"loss": 0.7238,
|
| 828 |
+
"step": 117
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 2.097777777777778,
|
| 832 |
+
"grad_norm": 0.45278117060661316,
|
| 833 |
+
"learning_rate": 2.470030568897938e-06,
|
| 834 |
+
"loss": 0.7086,
|
| 835 |
+
"step": 118
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 2.1155555555555554,
|
| 839 |
+
"grad_norm": 0.48289376497268677,
|
| 840 |
+
"learning_rate": 2.3808580878919948e-06,
|
| 841 |
+
"loss": 0.7369,
|
| 842 |
+
"step": 119
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 2.1333333333333333,
|
| 846 |
+
"grad_norm": 0.4182297885417938,
|
| 847 |
+
"learning_rate": 2.2928192835717642e-06,
|
| 848 |
+
"loss": 0.6709,
|
| 849 |
+
"step": 120
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 2.151111111111111,
|
| 853 |
+
"grad_norm": 0.41355010867118835,
|
| 854 |
+
"learning_rate": 2.205952262898704e-06,
|
| 855 |
+
"loss": 0.6351,
|
| 856 |
+
"step": 121
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 2.168888888888889,
|
| 860 |
+
"grad_norm": 0.38360127806663513,
|
| 861 |
+
"learning_rate": 2.120294625636171e-06,
|
| 862 |
+
"loss": 0.6603,
|
| 863 |
+
"step": 122
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 2.1866666666666665,
|
| 867 |
+
"grad_norm": 0.4717971980571747,
|
| 868 |
+
"learning_rate": 2.0358834480746363e-06,
|
| 869 |
+
"loss": 0.7047,
|
| 870 |
+
"step": 123
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 2.2044444444444444,
|
| 874 |
+
"grad_norm": 0.44043272733688354,
|
| 875 |
+
"learning_rate": 1.9527552669834797e-06,
|
| 876 |
+
"loss": 0.6574,
|
| 877 |
+
"step": 124
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 2.2222222222222223,
|
| 881 |
+
"grad_norm": 0.439424604177475,
|
| 882 |
+
"learning_rate": 1.8709460637963123e-06,
|
| 883 |
+
"loss": 0.7172,
|
| 884 |
+
"step": 125
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 2.24,
|
| 888 |
+
"grad_norm": 0.4320985674858093,
|
| 889 |
+
"learning_rate": 1.7904912490366723e-06,
|
| 890 |
+
"loss": 0.6212,
|
| 891 |
+
"step": 126
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 2.2577777777777777,
|
| 895 |
+
"grad_norm": 0.4921020567417145,
|
| 896 |
+
"learning_rate": 1.711425646990838e-06,
|
| 897 |
+
"loss": 0.7009,
|
| 898 |
+
"step": 127
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 2.2755555555555556,
|
| 902 |
+
"grad_norm": 0.43402716517448425,
|
| 903 |
+
"learning_rate": 1.6337834806343783e-06,
|
| 904 |
+
"loss": 0.6447,
|
| 905 |
+
"step": 128
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 2.2933333333333334,
|
| 909 |
+
"grad_norm": 0.417825847864151,
|
| 910 |
+
"learning_rate": 1.557598356819e-06,
|
| 911 |
+
"loss": 0.6013,
|
| 912 |
+
"step": 129
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 2.311111111111111,
|
| 916 |
+
"grad_norm": 0.4234044849872589,
|
| 917 |
+
"learning_rate": 1.482903251726049e-06,
|
| 918 |
+
"loss": 0.6542,
|
| 919 |
+
"step": 130
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 2.328888888888889,
|
| 923 |
+
"grad_norm": 0.41442492604255676,
|
| 924 |
+
"learning_rate": 1.409730496593016e-06,
|
| 925 |
+
"loss": 0.6156,
|
| 926 |
+
"step": 131
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 2.3466666666666667,
|
| 930 |
+
"grad_norm": 0.44905272126197815,
|
| 931 |
+
"learning_rate": 1.3381117637191887e-06,
|
| 932 |
+
"loss": 0.8226,
|
| 933 |
+
"step": 132
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 2.3644444444444446,
|
| 937 |
+
"grad_norm": 0.4065934717655182,
|
| 938 |
+
"learning_rate": 1.2680780527565313e-06,
|
| 939 |
+
"loss": 0.6674,
|
| 940 |
+
"step": 133
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 2.3822222222222225,
|
| 944 |
+
"grad_norm": 0.44520464539527893,
|
| 945 |
+
"learning_rate": 1.1996596772917091e-06,
|
| 946 |
+
"loss": 0.7292,
|
| 947 |
+
"step": 134
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 2.4,
|
| 951 |
+
"grad_norm": 0.3922524154186249,
|
| 952 |
+
"learning_rate": 1.132886251725061e-06,
|
| 953 |
+
"loss": 0.6922,
|
| 954 |
+
"step": 135
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 2.417777777777778,
|
| 958 |
+
"grad_norm": 0.40516623854637146,
|
| 959 |
+
"learning_rate": 1.0677866784522317e-06,
|
| 960 |
+
"loss": 0.7064,
|
| 961 |
+
"step": 136
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 2.4355555555555557,
|
| 965 |
+
"grad_norm": 0.3747960031032562,
|
| 966 |
+
"learning_rate": 1.004389135353972e-06,
|
| 967 |
+
"loss": 0.6197,
|
| 968 |
+
"step": 137
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 2.453333333333333,
|
| 972 |
+
"grad_norm": 0.4241096079349518,
|
| 973 |
+
"learning_rate": 9.427210635995482e-07,
|
| 974 |
+
"loss": 0.6535,
|
| 975 |
+
"step": 138
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 2.471111111111111,
|
| 979 |
+
"grad_norm": 0.38984397053718567,
|
| 980 |
+
"learning_rate": 8.828091557690288e-07,
|
| 981 |
+
"loss": 0.7117,
|
| 982 |
+
"step": 139
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 2.488888888888889,
|
| 986 |
+
"grad_norm": 0.3895849585533142,
|
| 987 |
+
"learning_rate": 8.246793442995954e-07,
|
| 988 |
+
"loss": 0.7144,
|
| 989 |
+
"step": 140
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 2.506666666666667,
|
| 993 |
+
"grad_norm": 0.3922812342643738,
|
| 994 |
+
"learning_rate": 7.68356790260873e-07,
|
| 995 |
+
"loss": 0.6957,
|
| 996 |
+
"step": 141
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 2.5244444444444447,
|
| 1000 |
+
"grad_norm": 0.4137726128101349,
|
| 1001 |
+
"learning_rate": 7.138658724641417e-07,
|
| 1002 |
+
"loss": 0.7647,
|
| 1003 |
+
"step": 142
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 2.542222222222222,
|
| 1007 |
+
"grad_norm": 0.41041967272758484,
|
| 1008 |
+
"learning_rate": 6.612301769101464e-07,
|
| 1009 |
+
"loss": 0.7211,
|
| 1010 |
+
"step": 143
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 2.56,
|
| 1014 |
+
"grad_norm": 0.3760773837566376,
|
| 1015 |
+
"learning_rate": 6.104724865800665e-07,
|
| 1016 |
+
"loss": 0.635,
|
| 1017 |
+
"step": 144
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 2.5777777777777775,
|
| 1021 |
+
"grad_norm": 0.38913705945014954,
|
| 1022 |
+
"learning_rate": 5.616147715740611e-07,
|
| 1023 |
+
"loss": 0.5809,
|
| 1024 |
+
"step": 145
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 2.5955555555555554,
|
| 1028 |
+
"grad_norm": 0.47968509793281555,
|
| 1029 |
+
"learning_rate": 5.146781796016798e-07,
|
| 1030 |
+
"loss": 0.8847,
|
| 1031 |
+
"step": 146
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 2.6133333333333333,
|
| 1035 |
+
"grad_norm": 0.36607035994529724,
|
| 1036 |
+
"learning_rate": 4.696830268282204e-07,
|
| 1037 |
+
"loss": 0.6341,
|
| 1038 |
+
"step": 147
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 2.631111111111111,
|
| 1042 |
+
"grad_norm": 0.37374281883239746,
|
| 1043 |
+
"learning_rate": 4.2664878908102556e-07,
|
| 1044 |
+
"loss": 0.7013,
|
| 1045 |
+
"step": 148
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 2.648888888888889,
|
| 1049 |
+
"grad_norm": 0.3905697464942932,
|
| 1050 |
+
"learning_rate": 3.855940934195146e-07,
|
| 1051 |
+
"loss": 0.7076,
|
| 1052 |
+
"step": 149
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 2.6666666666666665,
|
| 1056 |
+
"grad_norm": 0.38409143686294556,
|
| 1057 |
+
"learning_rate": 3.4653671007259084e-07,
|
| 1058 |
+
"loss": 0.7147,
|
| 1059 |
+
"step": 150
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 2.6844444444444444,
|
| 1063 |
+
"grad_norm": 0.36509791016578674,
|
| 1064 |
+
"learning_rate": 3.0949354474692937e-07,
|
| 1065 |
+
"loss": 0.6029,
|
| 1066 |
+
"step": 151
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 2.7022222222222223,
|
| 1070 |
+
"grad_norm": 0.4057907164096832,
|
| 1071 |
+
"learning_rate": 2.7448063130946224e-07,
|
| 1072 |
+
"loss": 0.7018,
|
| 1073 |
+
"step": 152
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 2.7199999999999998,
|
| 1077 |
+
"grad_norm": 0.4523860216140747,
|
| 1078 |
+
"learning_rate": 2.4151312484723465e-07,
|
| 1079 |
+
"loss": 0.7996,
|
| 1080 |
+
"step": 153
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 2.7377777777777776,
|
| 1084 |
+
"grad_norm": 0.37582913041114807,
|
| 1085 |
+
"learning_rate": 2.106052951076365e-07,
|
| 1086 |
+
"loss": 0.5584,
|
| 1087 |
+
"step": 154
|
| 1088 |
+
},
|
| 1089 |
+
{
|
| 1090 |
+
"epoch": 2.7555555555555555,
|
| 1091 |
+
"grad_norm": 0.4033922255039215,
|
| 1092 |
+
"learning_rate": 1.8177052032184285e-07,
|
| 1093 |
+
"loss": 0.626,
|
| 1094 |
+
"step": 155
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 2.7733333333333334,
|
| 1098 |
+
"grad_norm": 0.39120516180992126,
|
| 1099 |
+
"learning_rate": 1.5502128141414496e-07,
|
| 1100 |
+
"loss": 0.6039,
|
| 1101 |
+
"step": 156
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 2.7911111111111113,
|
| 1105 |
+
"grad_norm": 0.3827860951423645,
|
| 1106 |
+
"learning_rate": 1.303691565996712e-07,
|
| 1107 |
+
"loss": 0.7568,
|
| 1108 |
+
"step": 157
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"epoch": 2.8088888888888888,
|
| 1112 |
+
"grad_norm": 0.41667839884757996,
|
| 1113 |
+
"learning_rate": 1.0782481637284014e-07,
|
| 1114 |
+
"loss": 0.7259,
|
| 1115 |
+
"step": 158
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 2.8266666666666667,
|
| 1119 |
+
"grad_norm": 0.3329305052757263,
|
| 1120 |
+
"learning_rate": 8.739801888871468e-08,
|
| 1121 |
+
"loss": 0.5574,
|
| 1122 |
+
"step": 159
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 2.8444444444444446,
|
| 1126 |
+
"grad_norm": 0.39114266633987427,
|
| 1127 |
+
"learning_rate": 6.909760573925561e-08,
|
| 1128 |
+
"loss": 0.7079,
|
| 1129 |
+
"step": 160
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 2.862222222222222,
|
| 1133 |
+
"grad_norm": 0.39278408885002136,
|
| 1134 |
+
"learning_rate": 5.2931498126298495e-08,
|
| 1135 |
+
"loss": 0.6241,
|
| 1136 |
+
"step": 161
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 2.88,
|
| 1140 |
+
"grad_norm": 0.3883131146430969,
|
| 1141 |
+
"learning_rate": 3.890669343292464e-08,
|
| 1142 |
+
"loss": 0.7255,
|
| 1143 |
+
"step": 162
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 2.897777777777778,
|
| 1147 |
+
"grad_norm": 0.3658883273601532,
|
| 1148 |
+
"learning_rate": 2.702926219468882e-08,
|
| 1149 |
+
"loss": 0.6244,
|
| 1150 |
+
"step": 163
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 2.9155555555555557,
|
| 1154 |
+
"grad_norm": 0.3795766532421112,
|
| 1155 |
+
"learning_rate": 1.7304345472035634e-08,
|
| 1156 |
+
"loss": 0.7233,
|
| 1157 |
+
"step": 164
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 2.9333333333333336,
|
| 1161 |
+
"grad_norm": 0.3794161081314087,
|
| 1162 |
+
"learning_rate": 9.73615262502503e-09,
|
| 1163 |
+
"loss": 0.5895,
|
| 1164 |
+
"step": 165
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 2.951111111111111,
|
| 1168 |
+
"grad_norm": 0.41713038086891174,
|
| 1169 |
+
"learning_rate": 4.327959491344791e-09,
|
| 1170 |
+
"loss": 0.7434,
|
| 1171 |
+
"step": 166
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 2.968888888888889,
|
| 1175 |
+
"grad_norm": 0.36857718229293823,
|
| 1176 |
+
"learning_rate": 1.082106968385288e-09,
|
| 1177 |
+
"loss": 0.6239,
|
| 1178 |
+
"step": 167
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 2.986666666666667,
|
| 1182 |
+
"grad_norm": 0.38884976506233215,
|
| 1183 |
+
"learning_rate": 0.0,
|
| 1184 |
+
"loss": 0.6308,
|
| 1185 |
+
"step": 168
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 2.986666666666667,
|
| 1189 |
+
"step": 168,
|
| 1190 |
+
"total_flos": 139896582144000.0,
|
| 1191 |
+
"train_loss": 0.7518468662386849,
|
| 1192 |
+
"train_runtime": 7945.17,
|
| 1193 |
+
"train_samples_per_second": 2.039,
|
| 1194 |
+
"train_steps_per_second": 0.021
|
| 1195 |
+
}
|
| 1196 |
+
],
|
| 1197 |
+
"logging_steps": 1.0,
|
| 1198 |
+
"max_steps": 168,
|
| 1199 |
+
"num_input_tokens_seen": 0,
|
| 1200 |
+
"num_train_epochs": 3,
|
| 1201 |
+
"save_steps": 500,
|
| 1202 |
+
"stateful_callbacks": {
|
| 1203 |
+
"TrainerControl": {
|
| 1204 |
+
"args": {
|
| 1205 |
+
"should_epoch_stop": false,
|
| 1206 |
+
"should_evaluate": false,
|
| 1207 |
+
"should_log": false,
|
| 1208 |
+
"should_save": true,
|
| 1209 |
+
"should_training_stop": true
|
| 1210 |
+
},
|
| 1211 |
+
"attributes": {}
|
| 1212 |
+
}
|
| 1213 |
+
},
|
| 1214 |
+
"total_flos": 139896582144000.0,
|
| 1215 |
+
"train_batch_size": 1,
|
| 1216 |
+
"trial_name": null,
|
| 1217 |
+
"trial_params": null
|
| 1218 |
+
}
|
training_loss.png
ADDED
|