End of training
Browse files- README.md +3 -2
- all_results.json +12 -0
- eval_results.json +7 -0
- train_results.json +8 -0
- trainer_state.json +2894 -0
- training_eval_loss.png +0 -0
- training_loss.png +0 -0
README.md
CHANGED
|
@@ -4,6 +4,7 @@ license: mit
|
|
| 4 |
base_model: klyang/MentaLLaMA-chat-7B-hf
|
| 5 |
tags:
|
| 6 |
- llama-factory
|
|
|
|
| 7 |
- generated_from_trainer
|
| 8 |
model-index:
|
| 9 |
- name: MentaLLaMA-chat-7B-PsyCourse-fold3
|
|
@@ -15,9 +16,9 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 15 |
|
| 16 |
# MentaLLaMA-chat-7B-PsyCourse-fold3
|
| 17 |
|
| 18 |
-
This model is a fine-tuned version of [klyang/MentaLLaMA-chat-7B-hf](https://huggingface.co/klyang/MentaLLaMA-chat-7B-hf) on
|
| 19 |
It achieves the following results on the evaluation set:
|
| 20 |
-
- Loss: 0.
|
| 21 |
|
| 22 |
## Model description
|
| 23 |
|
|
|
|
| 4 |
base_model: klyang/MentaLLaMA-chat-7B-hf
|
| 5 |
tags:
|
| 6 |
- llama-factory
|
| 7 |
+
- lora
|
| 8 |
- generated_from_trainer
|
| 9 |
model-index:
|
| 10 |
- name: MentaLLaMA-chat-7B-PsyCourse-fold3
|
|
|
|
| 16 |
|
| 17 |
# MentaLLaMA-chat-7B-PsyCourse-fold3
|
| 18 |
|
| 19 |
+
This model is a fine-tuned version of [klyang/MentaLLaMA-chat-7B-hf](https://huggingface.co/klyang/MentaLLaMA-chat-7B-hf) on the course-train-fold3 dataset.
|
| 20 |
It achieves the following results on the evaluation set:
|
| 21 |
+
- Loss: 0.0302
|
| 22 |
|
| 23 |
## Model description
|
| 24 |
|
all_results.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 4.9976479443033215,
|
| 3 |
+
"eval_loss": 0.030188467353582382,
|
| 4 |
+
"eval_runtime": 180.1979,
|
| 5 |
+
"eval_samples_per_second": 6.554,
|
| 6 |
+
"eval_steps_per_second": 6.554,
|
| 7 |
+
"total_flos": 8.981559806648648e+17,
|
| 8 |
+
"train_loss": 0.048681382634620886,
|
| 9 |
+
"train_runtime": 37834.6736,
|
| 10 |
+
"train_samples_per_second": 1.405,
|
| 11 |
+
"train_steps_per_second": 0.088
|
| 12 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 4.9976479443033215,
|
| 3 |
+
"eval_loss": 0.030188467353582382,
|
| 4 |
+
"eval_runtime": 180.1979,
|
| 5 |
+
"eval_samples_per_second": 6.554,
|
| 6 |
+
"eval_steps_per_second": 6.554
|
| 7 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 4.9976479443033215,
|
| 3 |
+
"total_flos": 8.981559806648648e+17,
|
| 4 |
+
"train_loss": 0.048681382634620886,
|
| 5 |
+
"train_runtime": 37834.6736,
|
| 6 |
+
"train_samples_per_second": 1.405,
|
| 7 |
+
"train_steps_per_second": 0.088
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,2894 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.030188467353582382,
|
| 3 |
+
"best_model_checkpoint": "saves/psy-course/MentaLLaMA-chat-7B/train/fold3/checkpoint-1950",
|
| 4 |
+
"epoch": 4.9976479443033215,
|
| 5 |
+
"eval_steps": 50,
|
| 6 |
+
"global_step": 3320,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.015053156458744943,
|
| 13 |
+
"grad_norm": 1.822564721107483,
|
| 14 |
+
"learning_rate": 3.0120481927710846e-06,
|
| 15 |
+
"loss": 1.7049,
|
| 16 |
+
"step": 10
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.030106312917489886,
|
| 20 |
+
"grad_norm": 2.226895570755005,
|
| 21 |
+
"learning_rate": 6.024096385542169e-06,
|
| 22 |
+
"loss": 1.6672,
|
| 23 |
+
"step": 20
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.04515946937623483,
|
| 27 |
+
"grad_norm": 2.538222551345825,
|
| 28 |
+
"learning_rate": 9.036144578313253e-06,
|
| 29 |
+
"loss": 1.615,
|
| 30 |
+
"step": 30
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.06021262583497977,
|
| 34 |
+
"grad_norm": 3.4118905067443848,
|
| 35 |
+
"learning_rate": 1.2048192771084338e-05,
|
| 36 |
+
"loss": 1.4092,
|
| 37 |
+
"step": 40
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.07526578229372471,
|
| 41 |
+
"grad_norm": 0.845978856086731,
|
| 42 |
+
"learning_rate": 1.5060240963855424e-05,
|
| 43 |
+
"loss": 0.8176,
|
| 44 |
+
"step": 50
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.07526578229372471,
|
| 48 |
+
"eval_loss": 0.6253208518028259,
|
| 49 |
+
"eval_runtime": 178.2787,
|
| 50 |
+
"eval_samples_per_second": 6.624,
|
| 51 |
+
"eval_steps_per_second": 6.624,
|
| 52 |
+
"step": 50
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.09031893875246966,
|
| 56 |
+
"grad_norm": 0.6405627727508545,
|
| 57 |
+
"learning_rate": 1.8072289156626505e-05,
|
| 58 |
+
"loss": 0.5568,
|
| 59 |
+
"step": 60
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.1053720952112146,
|
| 63 |
+
"grad_norm": 0.9309800267219543,
|
| 64 |
+
"learning_rate": 2.1084337349397593e-05,
|
| 65 |
+
"loss": 0.4914,
|
| 66 |
+
"step": 70
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.12042525166995954,
|
| 70 |
+
"grad_norm": 0.7481735348701477,
|
| 71 |
+
"learning_rate": 2.4096385542168677e-05,
|
| 72 |
+
"loss": 0.3461,
|
| 73 |
+
"step": 80
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.1354784081287045,
|
| 77 |
+
"grad_norm": 0.840279757976532,
|
| 78 |
+
"learning_rate": 2.7108433734939758e-05,
|
| 79 |
+
"loss": 0.173,
|
| 80 |
+
"step": 90
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.15053156458744943,
|
| 84 |
+
"grad_norm": 0.6460654735565186,
|
| 85 |
+
"learning_rate": 3.012048192771085e-05,
|
| 86 |
+
"loss": 0.149,
|
| 87 |
+
"step": 100
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.15053156458744943,
|
| 91 |
+
"eval_loss": 0.12084457278251648,
|
| 92 |
+
"eval_runtime": 178.2793,
|
| 93 |
+
"eval_samples_per_second": 6.624,
|
| 94 |
+
"eval_steps_per_second": 6.624,
|
| 95 |
+
"step": 100
|
| 96 |
+
},
|
| 97 |
+
{
|
| 98 |
+
"epoch": 0.16558472104619437,
|
| 99 |
+
"grad_norm": 0.8036267757415771,
|
| 100 |
+
"learning_rate": 3.313253012048193e-05,
|
| 101 |
+
"loss": 0.1049,
|
| 102 |
+
"step": 110
|
| 103 |
+
},
|
| 104 |
+
{
|
| 105 |
+
"epoch": 0.18063787750493931,
|
| 106 |
+
"grad_norm": 0.4858762323856354,
|
| 107 |
+
"learning_rate": 3.614457831325301e-05,
|
| 108 |
+
"loss": 0.1118,
|
| 109 |
+
"step": 120
|
| 110 |
+
},
|
| 111 |
+
{
|
| 112 |
+
"epoch": 0.19569103396368426,
|
| 113 |
+
"grad_norm": 0.5866274833679199,
|
| 114 |
+
"learning_rate": 3.91566265060241e-05,
|
| 115 |
+
"loss": 0.091,
|
| 116 |
+
"step": 130
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 0.2107441904224292,
|
| 120 |
+
"grad_norm": 0.6353292465209961,
|
| 121 |
+
"learning_rate": 4.2168674698795186e-05,
|
| 122 |
+
"loss": 0.0815,
|
| 123 |
+
"step": 140
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"epoch": 0.22579734688117414,
|
| 127 |
+
"grad_norm": 0.8361664414405823,
|
| 128 |
+
"learning_rate": 4.5180722891566266e-05,
|
| 129 |
+
"loss": 0.0858,
|
| 130 |
+
"step": 150
|
| 131 |
+
},
|
| 132 |
+
{
|
| 133 |
+
"epoch": 0.22579734688117414,
|
| 134 |
+
"eval_loss": 0.0719056949019432,
|
| 135 |
+
"eval_runtime": 178.4648,
|
| 136 |
+
"eval_samples_per_second": 6.618,
|
| 137 |
+
"eval_steps_per_second": 6.618,
|
| 138 |
+
"step": 150
|
| 139 |
+
},
|
| 140 |
+
{
|
| 141 |
+
"epoch": 0.2408505033399191,
|
| 142 |
+
"grad_norm": 0.5231442451477051,
|
| 143 |
+
"learning_rate": 4.8192771084337354e-05,
|
| 144 |
+
"loss": 0.076,
|
| 145 |
+
"step": 160
|
| 146 |
+
},
|
| 147 |
+
{
|
| 148 |
+
"epoch": 0.25590365979866403,
|
| 149 |
+
"grad_norm": 0.7356946468353271,
|
| 150 |
+
"learning_rate": 5.120481927710844e-05,
|
| 151 |
+
"loss": 0.0844,
|
| 152 |
+
"step": 170
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 0.270956816257409,
|
| 156 |
+
"grad_norm": 0.7905665636062622,
|
| 157 |
+
"learning_rate": 5.4216867469879516e-05,
|
| 158 |
+
"loss": 0.0734,
|
| 159 |
+
"step": 180
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 0.2860099727161539,
|
| 163 |
+
"grad_norm": 0.9920386075973511,
|
| 164 |
+
"learning_rate": 5.72289156626506e-05,
|
| 165 |
+
"loss": 0.0687,
|
| 166 |
+
"step": 190
|
| 167 |
+
},
|
| 168 |
+
{
|
| 169 |
+
"epoch": 0.30106312917489886,
|
| 170 |
+
"grad_norm": 0.8998428583145142,
|
| 171 |
+
"learning_rate": 6.02409638554217e-05,
|
| 172 |
+
"loss": 0.0728,
|
| 173 |
+
"step": 200
|
| 174 |
+
},
|
| 175 |
+
{
|
| 176 |
+
"epoch": 0.30106312917489886,
|
| 177 |
+
"eval_loss": 0.056333187967538834,
|
| 178 |
+
"eval_runtime": 178.4366,
|
| 179 |
+
"eval_samples_per_second": 6.619,
|
| 180 |
+
"eval_steps_per_second": 6.619,
|
| 181 |
+
"step": 200
|
| 182 |
+
},
|
| 183 |
+
{
|
| 184 |
+
"epoch": 0.3161162856336438,
|
| 185 |
+
"grad_norm": 0.517510175704956,
|
| 186 |
+
"learning_rate": 6.325301204819278e-05,
|
| 187 |
+
"loss": 0.0601,
|
| 188 |
+
"step": 210
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 0.33116944209238874,
|
| 192 |
+
"grad_norm": 0.5527799725532532,
|
| 193 |
+
"learning_rate": 6.626506024096386e-05,
|
| 194 |
+
"loss": 0.0636,
|
| 195 |
+
"step": 220
|
| 196 |
+
},
|
| 197 |
+
{
|
| 198 |
+
"epoch": 0.3462225985511337,
|
| 199 |
+
"grad_norm": 0.5699729919433594,
|
| 200 |
+
"learning_rate": 6.927710843373494e-05,
|
| 201 |
+
"loss": 0.0742,
|
| 202 |
+
"step": 230
|
| 203 |
+
},
|
| 204 |
+
{
|
| 205 |
+
"epoch": 0.36127575500987863,
|
| 206 |
+
"grad_norm": 0.4265763461589813,
|
| 207 |
+
"learning_rate": 7.228915662650602e-05,
|
| 208 |
+
"loss": 0.0485,
|
| 209 |
+
"step": 240
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 0.3763289114686236,
|
| 213 |
+
"grad_norm": 0.8758322596549988,
|
| 214 |
+
"learning_rate": 7.530120481927712e-05,
|
| 215 |
+
"loss": 0.0574,
|
| 216 |
+
"step": 250
|
| 217 |
+
},
|
| 218 |
+
{
|
| 219 |
+
"epoch": 0.3763289114686236,
|
| 220 |
+
"eval_loss": 0.04959677904844284,
|
| 221 |
+
"eval_runtime": 178.4826,
|
| 222 |
+
"eval_samples_per_second": 6.617,
|
| 223 |
+
"eval_steps_per_second": 6.617,
|
| 224 |
+
"step": 250
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 0.3913820679273685,
|
| 228 |
+
"grad_norm": 0.4359073340892792,
|
| 229 |
+
"learning_rate": 7.83132530120482e-05,
|
| 230 |
+
"loss": 0.0393,
|
| 231 |
+
"step": 260
|
| 232 |
+
},
|
| 233 |
+
{
|
| 234 |
+
"epoch": 0.40643522438611346,
|
| 235 |
+
"grad_norm": 0.4522893726825714,
|
| 236 |
+
"learning_rate": 8.132530120481928e-05,
|
| 237 |
+
"loss": 0.0538,
|
| 238 |
+
"step": 270
|
| 239 |
+
},
|
| 240 |
+
{
|
| 241 |
+
"epoch": 0.4214883808448584,
|
| 242 |
+
"grad_norm": 0.531626284122467,
|
| 243 |
+
"learning_rate": 8.433734939759037e-05,
|
| 244 |
+
"loss": 0.0481,
|
| 245 |
+
"step": 280
|
| 246 |
+
},
|
| 247 |
+
{
|
| 248 |
+
"epoch": 0.43654153730360334,
|
| 249 |
+
"grad_norm": 0.26284611225128174,
|
| 250 |
+
"learning_rate": 8.734939759036145e-05,
|
| 251 |
+
"loss": 0.049,
|
| 252 |
+
"step": 290
|
| 253 |
+
},
|
| 254 |
+
{
|
| 255 |
+
"epoch": 0.4515946937623483,
|
| 256 |
+
"grad_norm": 0.639433741569519,
|
| 257 |
+
"learning_rate": 9.036144578313253e-05,
|
| 258 |
+
"loss": 0.0603,
|
| 259 |
+
"step": 300
|
| 260 |
+
},
|
| 261 |
+
{
|
| 262 |
+
"epoch": 0.4515946937623483,
|
| 263 |
+
"eval_loss": 0.05344177782535553,
|
| 264 |
+
"eval_runtime": 178.3037,
|
| 265 |
+
"eval_samples_per_second": 6.624,
|
| 266 |
+
"eval_steps_per_second": 6.624,
|
| 267 |
+
"step": 300
|
| 268 |
+
},
|
| 269 |
+
{
|
| 270 |
+
"epoch": 0.46664785022109323,
|
| 271 |
+
"grad_norm": 0.23814865946769714,
|
| 272 |
+
"learning_rate": 9.337349397590361e-05,
|
| 273 |
+
"loss": 0.0443,
|
| 274 |
+
"step": 310
|
| 275 |
+
},
|
| 276 |
+
{
|
| 277 |
+
"epoch": 0.4817010066798382,
|
| 278 |
+
"grad_norm": 0.8485608100891113,
|
| 279 |
+
"learning_rate": 9.638554216867471e-05,
|
| 280 |
+
"loss": 0.0691,
|
| 281 |
+
"step": 320
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 0.4967541631385831,
|
| 285 |
+
"grad_norm": 0.7342369556427002,
|
| 286 |
+
"learning_rate": 9.939759036144579e-05,
|
| 287 |
+
"loss": 0.0535,
|
| 288 |
+
"step": 330
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"epoch": 0.5118073195973281,
|
| 292 |
+
"grad_norm": 0.5011388063430786,
|
| 293 |
+
"learning_rate": 9.999823129264712e-05,
|
| 294 |
+
"loss": 0.038,
|
| 295 |
+
"step": 340
|
| 296 |
+
},
|
| 297 |
+
{
|
| 298 |
+
"epoch": 0.526860476056073,
|
| 299 |
+
"grad_norm": 0.2543235421180725,
|
| 300 |
+
"learning_rate": 9.999104613348688e-05,
|
| 301 |
+
"loss": 0.0432,
|
| 302 |
+
"step": 350
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 0.526860476056073,
|
| 306 |
+
"eval_loss": 0.043458491563797,
|
| 307 |
+
"eval_runtime": 178.3108,
|
| 308 |
+
"eval_samples_per_second": 6.623,
|
| 309 |
+
"eval_steps_per_second": 6.623,
|
| 310 |
+
"step": 350
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.541913632514818,
|
| 314 |
+
"grad_norm": 0.19088613986968994,
|
| 315 |
+
"learning_rate": 9.997833477197385e-05,
|
| 316 |
+
"loss": 0.054,
|
| 317 |
+
"step": 360
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.5569667889735629,
|
| 321 |
+
"grad_norm": 0.4376489818096161,
|
| 322 |
+
"learning_rate": 9.996009861327077e-05,
|
| 323 |
+
"loss": 0.043,
|
| 324 |
+
"step": 370
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.5720199454323078,
|
| 328 |
+
"grad_norm": 0.33153238892555237,
|
| 329 |
+
"learning_rate": 9.993633967327269e-05,
|
| 330 |
+
"loss": 0.0445,
|
| 331 |
+
"step": 380
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.5870731018910528,
|
| 335 |
+
"grad_norm": 0.5583683252334595,
|
| 336 |
+
"learning_rate": 9.990706057838416e-05,
|
| 337 |
+
"loss": 0.04,
|
| 338 |
+
"step": 390
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.6021262583497977,
|
| 342 |
+
"grad_norm": 0.4868285655975342,
|
| 343 |
+
"learning_rate": 9.987226456522884e-05,
|
| 344 |
+
"loss": 0.0506,
|
| 345 |
+
"step": 400
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.6021262583497977,
|
| 349 |
+
"eval_loss": 0.04465359449386597,
|
| 350 |
+
"eval_runtime": 178.2721,
|
| 351 |
+
"eval_samples_per_second": 6.625,
|
| 352 |
+
"eval_steps_per_second": 6.625,
|
| 353 |
+
"step": 400
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.6171794148085427,
|
| 357 |
+
"grad_norm": 0.4056116044521332,
|
| 358 |
+
"learning_rate": 9.983195548029173e-05,
|
| 359 |
+
"loss": 0.0334,
|
| 360 |
+
"step": 410
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 0.6322325712672876,
|
| 364 |
+
"grad_norm": 0.5368084907531738,
|
| 365 |
+
"learning_rate": 9.9786137779494e-05,
|
| 366 |
+
"loss": 0.0537,
|
| 367 |
+
"step": 420
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 0.6472857277260325,
|
| 371 |
+
"grad_norm": 0.2836042046546936,
|
| 372 |
+
"learning_rate": 9.973481652770038e-05,
|
| 373 |
+
"loss": 0.0403,
|
| 374 |
+
"step": 430
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.6623388841847775,
|
| 378 |
+
"grad_norm": 0.2168349325656891,
|
| 379 |
+
"learning_rate": 9.967799739815925e-05,
|
| 380 |
+
"loss": 0.0322,
|
| 381 |
+
"step": 440
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 0.6773920406435224,
|
| 385 |
+
"grad_norm": 0.5306774973869324,
|
| 386 |
+
"learning_rate": 9.961568667187556e-05,
|
| 387 |
+
"loss": 0.0418,
|
| 388 |
+
"step": 450
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 0.6773920406435224,
|
| 392 |
+
"eval_loss": 0.042699169367551804,
|
| 393 |
+
"eval_runtime": 178.1761,
|
| 394 |
+
"eval_samples_per_second": 6.628,
|
| 395 |
+
"eval_steps_per_second": 6.628,
|
| 396 |
+
"step": 450
|
| 397 |
+
},
|
| 398 |
+
{
|
| 399 |
+
"epoch": 0.6924451971022674,
|
| 400 |
+
"grad_norm": 0.5590943098068237,
|
| 401 |
+
"learning_rate": 9.954789123691642e-05,
|
| 402 |
+
"loss": 0.0423,
|
| 403 |
+
"step": 460
|
| 404 |
+
},
|
| 405 |
+
{
|
| 406 |
+
"epoch": 0.7074983535610123,
|
| 407 |
+
"grad_norm": 0.46472442150115967,
|
| 408 |
+
"learning_rate": 9.947461858764978e-05,
|
| 409 |
+
"loss": 0.0383,
|
| 410 |
+
"step": 470
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 0.7225515100197573,
|
| 414 |
+
"grad_norm": 0.307923823595047,
|
| 415 |
+
"learning_rate": 9.939587682391586e-05,
|
| 416 |
+
"loss": 0.0392,
|
| 417 |
+
"step": 480
|
| 418 |
+
},
|
| 419 |
+
{
|
| 420 |
+
"epoch": 0.7376046664785022,
|
| 421 |
+
"grad_norm": 0.3933861255645752,
|
| 422 |
+
"learning_rate": 9.931167465013182e-05,
|
| 423 |
+
"loss": 0.0519,
|
| 424 |
+
"step": 490
|
| 425 |
+
},
|
| 426 |
+
{
|
| 427 |
+
"epoch": 0.7526578229372471,
|
| 428 |
+
"grad_norm": 0.31609928607940674,
|
| 429 |
+
"learning_rate": 9.922202137432955e-05,
|
| 430 |
+
"loss": 0.0319,
|
| 431 |
+
"step": 500
|
| 432 |
+
},
|
| 433 |
+
{
|
| 434 |
+
"epoch": 0.7526578229372471,
|
| 435 |
+
"eval_loss": 0.03804313763976097,
|
| 436 |
+
"eval_runtime": 178.3743,
|
| 437 |
+
"eval_samples_per_second": 6.621,
|
| 438 |
+
"eval_steps_per_second": 6.621,
|
| 439 |
+
"step": 500
|
| 440 |
+
},
|
| 441 |
+
{
|
| 442 |
+
"epoch": 0.7677109793959921,
|
| 443 |
+
"grad_norm": 0.2750873863697052,
|
| 444 |
+
"learning_rate": 9.912692690712665e-05,
|
| 445 |
+
"loss": 0.0356,
|
| 446 |
+
"step": 510
|
| 447 |
+
},
|
| 448 |
+
{
|
| 449 |
+
"epoch": 0.782764135854737,
|
| 450 |
+
"grad_norm": 0.4082382023334503,
|
| 451 |
+
"learning_rate": 9.902640176063103e-05,
|
| 452 |
+
"loss": 0.0426,
|
| 453 |
+
"step": 520
|
| 454 |
+
},
|
| 455 |
+
{
|
| 456 |
+
"epoch": 0.797817292313482,
|
| 457 |
+
"grad_norm": 0.2213507443666458,
|
| 458 |
+
"learning_rate": 9.892045704727864e-05,
|
| 459 |
+
"loss": 0.0579,
|
| 460 |
+
"step": 530
|
| 461 |
+
},
|
| 462 |
+
{
|
| 463 |
+
"epoch": 0.8128704487722269,
|
| 464 |
+
"grad_norm": 0.24513815343379974,
|
| 465 |
+
"learning_rate": 9.880910447860527e-05,
|
| 466 |
+
"loss": 0.0512,
|
| 467 |
+
"step": 540
|
| 468 |
+
},
|
| 469 |
+
{
|
| 470 |
+
"epoch": 0.8279236052309719,
|
| 471 |
+
"grad_norm": 0.33264800906181335,
|
| 472 |
+
"learning_rate": 9.869235636395177e-05,
|
| 473 |
+
"loss": 0.0409,
|
| 474 |
+
"step": 550
|
| 475 |
+
},
|
| 476 |
+
{
|
| 477 |
+
"epoch": 0.8279236052309719,
|
| 478 |
+
"eval_loss": 0.03934917971491814,
|
| 479 |
+
"eval_runtime": 178.3094,
|
| 480 |
+
"eval_samples_per_second": 6.623,
|
| 481 |
+
"eval_steps_per_second": 6.623,
|
| 482 |
+
"step": 550
|
| 483 |
+
},
|
| 484 |
+
{
|
| 485 |
+
"epoch": 0.8429767616897168,
|
| 486 |
+
"grad_norm": 0.4701898694038391,
|
| 487 |
+
"learning_rate": 9.857022560910338e-05,
|
| 488 |
+
"loss": 0.0513,
|
| 489 |
+
"step": 560
|
| 490 |
+
},
|
| 491 |
+
{
|
| 492 |
+
"epoch": 0.8580299181484617,
|
| 493 |
+
"grad_norm": 0.36034145951271057,
|
| 494 |
+
"learning_rate": 9.844272571486311e-05,
|
| 495 |
+
"loss": 0.0375,
|
| 496 |
+
"step": 570
|
| 497 |
+
},
|
| 498 |
+
{
|
| 499 |
+
"epoch": 0.8730830746072067,
|
| 500 |
+
"grad_norm": 0.1605140119791031,
|
| 501 |
+
"learning_rate": 9.830987077555924e-05,
|
| 502 |
+
"loss": 0.0471,
|
| 503 |
+
"step": 580
|
| 504 |
+
},
|
| 505 |
+
{
|
| 506 |
+
"epoch": 0.8881362310659516,
|
| 507 |
+
"grad_norm": 0.3974078893661499,
|
| 508 |
+
"learning_rate": 9.817167547748729e-05,
|
| 509 |
+
"loss": 0.0588,
|
| 510 |
+
"step": 590
|
| 511 |
+
},
|
| 512 |
+
{
|
| 513 |
+
"epoch": 0.9031893875246966,
|
| 514 |
+
"grad_norm": 0.21184875071048737,
|
| 515 |
+
"learning_rate": 9.802815509728662e-05,
|
| 516 |
+
"loss": 0.0289,
|
| 517 |
+
"step": 600
|
| 518 |
+
},
|
| 519 |
+
{
|
| 520 |
+
"epoch": 0.9031893875246966,
|
| 521 |
+
"eval_loss": 0.0372873954474926,
|
| 522 |
+
"eval_runtime": 178.4025,
|
| 523 |
+
"eval_samples_per_second": 6.62,
|
| 524 |
+
"eval_steps_per_second": 6.62,
|
| 525 |
+
"step": 600
|
| 526 |
+
},
|
| 527 |
+
{
|
| 528 |
+
"epoch": 0.9182425439834415,
|
| 529 |
+
"grad_norm": 0.3172430992126465,
|
| 530 |
+
"learning_rate": 9.787932550025158e-05,
|
| 531 |
+
"loss": 0.0351,
|
| 532 |
+
"step": 610
|
| 533 |
+
},
|
| 534 |
+
{
|
| 535 |
+
"epoch": 0.9332957004421865,
|
| 536 |
+
"grad_norm": 0.25452977418899536,
|
| 537 |
+
"learning_rate": 9.772520313857775e-05,
|
| 538 |
+
"loss": 0.0354,
|
| 539 |
+
"step": 620
|
| 540 |
+
},
|
| 541 |
+
{
|
| 542 |
+
"epoch": 0.9483488569009314,
|
| 543 |
+
"grad_norm": 0.31180253624916077,
|
| 544 |
+
"learning_rate": 9.756580504954334e-05,
|
| 545 |
+
"loss": 0.0408,
|
| 546 |
+
"step": 630
|
| 547 |
+
},
|
| 548 |
+
{
|
| 549 |
+
"epoch": 0.9634020133596763,
|
| 550 |
+
"grad_norm": 0.2877283990383148,
|
| 551 |
+
"learning_rate": 9.740114885362562e-05,
|
| 552 |
+
"loss": 0.0357,
|
| 553 |
+
"step": 640
|
| 554 |
+
},
|
| 555 |
+
{
|
| 556 |
+
"epoch": 0.9784551698184213,
|
| 557 |
+
"grad_norm": 0.31497249007225037,
|
| 558 |
+
"learning_rate": 9.723125275255325e-05,
|
| 559 |
+
"loss": 0.0301,
|
| 560 |
+
"step": 650
|
| 561 |
+
},
|
| 562 |
+
{
|
| 563 |
+
"epoch": 0.9784551698184213,
|
| 564 |
+
"eval_loss": 0.03777872771024704,
|
| 565 |
+
"eval_runtime": 178.3928,
|
| 566 |
+
"eval_samples_per_second": 6.62,
|
| 567 |
+
"eval_steps_per_second": 6.62,
|
| 568 |
+
"step": 650
|
| 569 |
+
},
|
| 570 |
+
{
|
| 571 |
+
"epoch": 0.9935083262771662,
|
| 572 |
+
"grad_norm": 0.18348178267478943,
|
| 573 |
+
"learning_rate": 9.705613552729415e-05,
|
| 574 |
+
"loss": 0.0259,
|
| 575 |
+
"step": 660
|
| 576 |
+
},
|
| 577 |
+
{
|
| 578 |
+
"epoch": 1.0085614827359113,
|
| 579 |
+
"grad_norm": 0.15709549188613892,
|
| 580 |
+
"learning_rate": 9.68758165359794e-05,
|
| 581 |
+
"loss": 0.0335,
|
| 582 |
+
"step": 670
|
| 583 |
+
},
|
| 584 |
+
{
|
| 585 |
+
"epoch": 1.0236146391946561,
|
| 586 |
+
"grad_norm": 0.25934895873069763,
|
| 587 |
+
"learning_rate": 9.669031571176322e-05,
|
| 588 |
+
"loss": 0.0353,
|
| 589 |
+
"step": 680
|
| 590 |
+
},
|
| 591 |
+
{
|
| 592 |
+
"epoch": 1.0386677956534012,
|
| 593 |
+
"grad_norm": 0.21780580282211304,
|
| 594 |
+
"learning_rate": 9.64996535606196e-05,
|
| 595 |
+
"loss": 0.0318,
|
| 596 |
+
"step": 690
|
| 597 |
+
},
|
| 598 |
+
{
|
| 599 |
+
"epoch": 1.053720952112146,
|
| 600 |
+
"grad_norm": 0.14777664840221405,
|
| 601 |
+
"learning_rate": 9.630385115907545e-05,
|
| 602 |
+
"loss": 0.0289,
|
| 603 |
+
"step": 700
|
| 604 |
+
},
|
| 605 |
+
{
|
| 606 |
+
"epoch": 1.053720952112146,
|
| 607 |
+
"eval_loss": 0.03666107356548309,
|
| 608 |
+
"eval_runtime": 178.3948,
|
| 609 |
+
"eval_samples_per_second": 6.62,
|
| 610 |
+
"eval_steps_per_second": 6.62,
|
| 611 |
+
"step": 700
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 1.068774108570891,
|
| 615 |
+
"grad_norm": 0.09680185467004776,
|
| 616 |
+
"learning_rate": 9.610293015188067e-05,
|
| 617 |
+
"loss": 0.0305,
|
| 618 |
+
"step": 710
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 1.083827265029636,
|
| 622 |
+
"grad_norm": 0.21849073469638824,
|
| 623 |
+
"learning_rate": 9.589691274961556e-05,
|
| 624 |
+
"loss": 0.0241,
|
| 625 |
+
"step": 720
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 1.098880421488381,
|
| 629 |
+
"grad_norm": 0.23137938976287842,
|
| 630 |
+
"learning_rate": 9.568582172623544e-05,
|
| 631 |
+
"loss": 0.0367,
|
| 632 |
+
"step": 730
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 1.1139335779471258,
|
| 636 |
+
"grad_norm": 0.38276636600494385,
|
| 637 |
+
"learning_rate": 9.546968041655326e-05,
|
| 638 |
+
"loss": 0.0302,
|
| 639 |
+
"step": 740
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 1.1289867344058706,
|
| 643 |
+
"grad_norm": 0.17088480293750763,
|
| 644 |
+
"learning_rate": 9.524851271366001e-05,
|
| 645 |
+
"loss": 0.0283,
|
| 646 |
+
"step": 750
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 1.1289867344058706,
|
| 650 |
+
"eval_loss": 0.03535303846001625,
|
| 651 |
+
"eval_runtime": 178.355,
|
| 652 |
+
"eval_samples_per_second": 6.622,
|
| 653 |
+
"eval_steps_per_second": 6.622,
|
| 654 |
+
"step": 750
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 1.1440398908646157,
|
| 658 |
+
"grad_norm": 0.236710324883461,
|
| 659 |
+
"learning_rate": 9.502234306628355e-05,
|
| 660 |
+
"loss": 0.0295,
|
| 661 |
+
"step": 760
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 1.1590930473233607,
|
| 665 |
+
"grad_norm": 0.23621885478496552,
|
| 666 |
+
"learning_rate": 9.47911964760858e-05,
|
| 667 |
+
"loss": 0.0272,
|
| 668 |
+
"step": 770
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 1.1741462037821055,
|
| 672 |
+
"grad_norm": 0.41253966093063354,
|
| 673 |
+
"learning_rate": 9.455509849489915e-05,
|
| 674 |
+
"loss": 0.0254,
|
| 675 |
+
"step": 780
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 1.1891993602408506,
|
| 679 |
+
"grad_norm": 0.2661738991737366,
|
| 680 |
+
"learning_rate": 9.431407522190175e-05,
|
| 681 |
+
"loss": 0.0321,
|
| 682 |
+
"step": 790
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 1.2042525166995954,
|
| 686 |
+
"grad_norm": 0.26856687664985657,
|
| 687 |
+
"learning_rate": 9.406815330073244e-05,
|
| 688 |
+
"loss": 0.0338,
|
| 689 |
+
"step": 800
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 1.2042525166995954,
|
| 693 |
+
"eval_loss": 0.03521593660116196,
|
| 694 |
+
"eval_runtime": 178.2893,
|
| 695 |
+
"eval_samples_per_second": 6.624,
|
| 696 |
+
"eval_steps_per_second": 6.624,
|
| 697 |
+
"step": 800
|
| 698 |
+
},
|
| 699 |
+
{
|
| 700 |
+
"epoch": 1.2193056731583405,
|
| 701 |
+
"grad_norm": 0.15742696821689606,
|
| 702 |
+
"learning_rate": 9.381735991654546e-05,
|
| 703 |
+
"loss": 0.03,
|
| 704 |
+
"step": 810
|
| 705 |
+
},
|
| 706 |
+
{
|
| 707 |
+
"epoch": 1.2343588296170853,
|
| 708 |
+
"grad_norm": 0.38911351561546326,
|
| 709 |
+
"learning_rate": 9.356172279300528e-05,
|
| 710 |
+
"loss": 0.0352,
|
| 711 |
+
"step": 820
|
| 712 |
+
},
|
| 713 |
+
{
|
| 714 |
+
"epoch": 1.2494119860758304,
|
| 715 |
+
"grad_norm": 0.12868359684944153,
|
| 716 |
+
"learning_rate": 9.330127018922194e-05,
|
| 717 |
+
"loss": 0.0296,
|
| 718 |
+
"step": 830
|
| 719 |
+
},
|
| 720 |
+
{
|
| 721 |
+
"epoch": 1.2644651425345752,
|
| 722 |
+
"grad_norm": 0.1884261518716812,
|
| 723 |
+
"learning_rate": 9.303603089662716e-05,
|
| 724 |
+
"loss": 0.0368,
|
| 725 |
+
"step": 840
|
| 726 |
+
},
|
| 727 |
+
{
|
| 728 |
+
"epoch": 1.2795182989933203,
|
| 729 |
+
"grad_norm": 0.14614073932170868,
|
| 730 |
+
"learning_rate": 9.276603423579164e-05,
|
| 731 |
+
"loss": 0.0286,
|
| 732 |
+
"step": 850
|
| 733 |
+
},
|
| 734 |
+
{
|
| 735 |
+
"epoch": 1.2795182989933203,
|
| 736 |
+
"eval_loss": 0.035069651901721954,
|
| 737 |
+
"eval_runtime": 178.3662,
|
| 738 |
+
"eval_samples_per_second": 6.621,
|
| 739 |
+
"eval_steps_per_second": 6.621,
|
| 740 |
+
"step": 850
|
| 741 |
+
},
|
| 742 |
+
{
|
| 743 |
+
"epoch": 1.294571455452065,
|
| 744 |
+
"grad_norm": 0.33718934655189514,
|
| 745 |
+
"learning_rate": 9.249131005318387e-05,
|
| 746 |
+
"loss": 0.0412,
|
| 747 |
+
"step": 860
|
| 748 |
+
},
|
| 749 |
+
{
|
| 750 |
+
"epoch": 1.30962461191081,
|
| 751 |
+
"grad_norm": 0.1532343029975891,
|
| 752 |
+
"learning_rate": 9.221188871787075e-05,
|
| 753 |
+
"loss": 0.0327,
|
| 754 |
+
"step": 870
|
| 755 |
+
},
|
| 756 |
+
{
|
| 757 |
+
"epoch": 1.324677768369555,
|
| 758 |
+
"grad_norm": 0.16133852303028107,
|
| 759 |
+
"learning_rate": 9.192780111816047e-05,
|
| 760 |
+
"loss": 0.0297,
|
| 761 |
+
"step": 880
|
| 762 |
+
},
|
| 763 |
+
{
|
| 764 |
+
"epoch": 1.3397309248283,
|
| 765 |
+
"grad_norm": 0.19326288998126984,
|
| 766 |
+
"learning_rate": 9.163907865818806e-05,
|
| 767 |
+
"loss": 0.0304,
|
| 768 |
+
"step": 890
|
| 769 |
+
},
|
| 770 |
+
{
|
| 771 |
+
"epoch": 1.3547840812870449,
|
| 772 |
+
"grad_norm": 0.2810550332069397,
|
| 773 |
+
"learning_rate": 9.134575325444376e-05,
|
| 774 |
+
"loss": 0.0396,
|
| 775 |
+
"step": 900
|
| 776 |
+
},
|
| 777 |
+
{
|
| 778 |
+
"epoch": 1.3547840812870449,
|
| 779 |
+
"eval_loss": 0.0357198566198349,
|
| 780 |
+
"eval_runtime": 178.4266,
|
| 781 |
+
"eval_samples_per_second": 6.619,
|
| 782 |
+
"eval_steps_per_second": 6.619,
|
| 783 |
+
"step": 900
|
| 784 |
+
},
|
| 785 |
+
{
|
| 786 |
+
"epoch": 1.3698372377457897,
|
| 787 |
+
"grad_norm": 0.24054096639156342,
|
| 788 |
+
"learning_rate": 9.104785733224496e-05,
|
| 789 |
+
"loss": 0.0297,
|
| 790 |
+
"step": 910
|
| 791 |
+
},
|
| 792 |
+
{
|
| 793 |
+
"epoch": 1.3848903942045347,
|
| 794 |
+
"grad_norm": 0.1945008933544159,
|
| 795 |
+
"learning_rate": 9.07454238221517e-05,
|
| 796 |
+
"loss": 0.0297,
|
| 797 |
+
"step": 920
|
| 798 |
+
},
|
| 799 |
+
{
|
| 800 |
+
"epoch": 1.3999435506632798,
|
| 801 |
+
"grad_norm": 0.22676967084407806,
|
| 802 |
+
"learning_rate": 9.043848615632642e-05,
|
| 803 |
+
"loss": 0.0258,
|
| 804 |
+
"step": 930
|
| 805 |
+
},
|
| 806 |
+
{
|
| 807 |
+
"epoch": 1.4149967071220246,
|
| 808 |
+
"grad_norm": 0.12921719253063202,
|
| 809 |
+
"learning_rate": 9.012707826483823e-05,
|
| 810 |
+
"loss": 0.0331,
|
| 811 |
+
"step": 940
|
| 812 |
+
},
|
| 813 |
+
{
|
| 814 |
+
"epoch": 1.4300498635807695,
|
| 815 |
+
"grad_norm": 0.2984769940376282,
|
| 816 |
+
"learning_rate": 8.98112345719122e-05,
|
| 817 |
+
"loss": 0.0286,
|
| 818 |
+
"step": 950
|
| 819 |
+
},
|
| 820 |
+
{
|
| 821 |
+
"epoch": 1.4300498635807695,
|
| 822 |
+
"eval_loss": 0.0343579463660717,
|
| 823 |
+
"eval_runtime": 178.3834,
|
| 824 |
+
"eval_samples_per_second": 6.621,
|
| 825 |
+
"eval_steps_per_second": 6.621,
|
| 826 |
+
"step": 950
|
| 827 |
+
},
|
| 828 |
+
{
|
| 829 |
+
"epoch": 1.4451030200395145,
|
| 830 |
+
"grad_norm": 0.5328255891799927,
|
| 831 |
+
"learning_rate": 8.949098999212391e-05,
|
| 832 |
+
"loss": 0.0325,
|
| 833 |
+
"step": 960
|
| 834 |
+
},
|
| 835 |
+
{
|
| 836 |
+
"epoch": 1.4601561764982596,
|
| 837 |
+
"grad_norm": 0.22663827240467072,
|
| 838 |
+
"learning_rate": 8.916637992653991e-05,
|
| 839 |
+
"loss": 0.025,
|
| 840 |
+
"step": 970
|
| 841 |
+
},
|
| 842 |
+
{
|
| 843 |
+
"epoch": 1.4752093329570044,
|
| 844 |
+
"grad_norm": 0.31026363372802734,
|
| 845 |
+
"learning_rate": 8.883744025880428e-05,
|
| 846 |
+
"loss": 0.0322,
|
| 847 |
+
"step": 980
|
| 848 |
+
},
|
| 849 |
+
{
|
| 850 |
+
"epoch": 1.4902624894157492,
|
| 851 |
+
"grad_norm": 0.1781584918498993,
|
| 852 |
+
"learning_rate": 8.850420735117202e-05,
|
| 853 |
+
"loss": 0.0239,
|
| 854 |
+
"step": 990
|
| 855 |
+
},
|
| 856 |
+
{
|
| 857 |
+
"epoch": 1.5053156458744943,
|
| 858 |
+
"grad_norm": 0.25284937024116516,
|
| 859 |
+
"learning_rate": 8.816671804048933e-05,
|
| 860 |
+
"loss": 0.0276,
|
| 861 |
+
"step": 1000
|
| 862 |
+
},
|
| 863 |
+
{
|
| 864 |
+
"epoch": 1.5053156458744943,
|
| 865 |
+
"eval_loss": 0.03388373181223869,
|
| 866 |
+
"eval_runtime": 178.2776,
|
| 867 |
+
"eval_samples_per_second": 6.624,
|
| 868 |
+
"eval_steps_per_second": 6.624,
|
| 869 |
+
"step": 1000
|
| 870 |
+
},
|
| 871 |
+
{
|
| 872 |
+
"epoch": 1.5203688023332393,
|
| 873 |
+
"grad_norm": 0.18928562104701996,
|
| 874 |
+
"learning_rate": 8.782500963412156e-05,
|
| 875 |
+
"loss": 0.0317,
|
| 876 |
+
"step": 1010
|
| 877 |
+
},
|
| 878 |
+
{
|
| 879 |
+
"epoch": 1.5354219587919842,
|
| 880 |
+
"grad_norm": 0.24282586574554443,
|
| 881 |
+
"learning_rate": 8.747911990582912e-05,
|
| 882 |
+
"loss": 0.0249,
|
| 883 |
+
"step": 1020
|
| 884 |
+
},
|
| 885 |
+
{
|
| 886 |
+
"epoch": 1.550475115250729,
|
| 887 |
+
"grad_norm": 0.1898348331451416,
|
| 888 |
+
"learning_rate": 8.712908709159183e-05,
|
| 889 |
+
"loss": 0.025,
|
| 890 |
+
"step": 1030
|
| 891 |
+
},
|
| 892 |
+
{
|
| 893 |
+
"epoch": 1.565528271709474,
|
| 894 |
+
"grad_norm": 0.46804437041282654,
|
| 895 |
+
"learning_rate": 8.677494988538211e-05,
|
| 896 |
+
"loss": 0.034,
|
| 897 |
+
"step": 1040
|
| 898 |
+
},
|
| 899 |
+
{
|
| 900 |
+
"epoch": 1.5805814281682191,
|
| 901 |
+
"grad_norm": 0.21851754188537598,
|
| 902 |
+
"learning_rate": 8.641674743488769e-05,
|
| 903 |
+
"loss": 0.0269,
|
| 904 |
+
"step": 1050
|
| 905 |
+
},
|
| 906 |
+
{
|
| 907 |
+
"epoch": 1.5805814281682191,
|
| 908 |
+
"eval_loss": 0.03368181735277176,
|
| 909 |
+
"eval_runtime": 178.3246,
|
| 910 |
+
"eval_samples_per_second": 6.623,
|
| 911 |
+
"eval_steps_per_second": 6.623,
|
| 912 |
+
"step": 1050
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 1.595634584626964,
|
| 916 |
+
"grad_norm": 0.3493017256259918,
|
| 917 |
+
"learning_rate": 8.605451933718397e-05,
|
| 918 |
+
"loss": 0.028,
|
| 919 |
+
"step": 1060
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 1.6106877410857088,
|
| 923 |
+
"grad_norm": 0.3040076792240143,
|
| 924 |
+
"learning_rate": 8.568830563435694e-05,
|
| 925 |
+
"loss": 0.0347,
|
| 926 |
+
"step": 1070
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 1.6257408975444538,
|
| 930 |
+
"grad_norm": 0.2517918646335602,
|
| 931 |
+
"learning_rate": 8.531814680907664e-05,
|
| 932 |
+
"loss": 0.0212,
|
| 933 |
+
"step": 1080
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 1.6407940540031989,
|
| 937 |
+
"grad_norm": 0.28657108545303345,
|
| 938 |
+
"learning_rate": 8.494408378012209e-05,
|
| 939 |
+
"loss": 0.0253,
|
| 940 |
+
"step": 1090
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 1.6558472104619437,
|
| 944 |
+
"grad_norm": 0.12799641489982605,
|
| 945 |
+
"learning_rate": 8.456615789785804e-05,
|
| 946 |
+
"loss": 0.0202,
|
| 947 |
+
"step": 1100
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 1.6558472104619437,
|
| 951 |
+
"eval_loss": 0.033622123301029205,
|
| 952 |
+
"eval_runtime": 178.3151,
|
| 953 |
+
"eval_samples_per_second": 6.623,
|
| 954 |
+
"eval_steps_per_second": 6.623,
|
| 955 |
+
"step": 1100
|
| 956 |
+
},
|
| 957 |
+
{
|
| 958 |
+
"epoch": 1.6709003669206886,
|
| 959 |
+
"grad_norm": 0.220402330160141,
|
| 960 |
+
"learning_rate": 8.418441093966385e-05,
|
| 961 |
+
"loss": 0.0368,
|
| 962 |
+
"step": 1110
|
| 963 |
+
},
|
| 964 |
+
{
|
| 965 |
+
"epoch": 1.6859535233794336,
|
| 966 |
+
"grad_norm": 0.2065258026123047,
|
| 967 |
+
"learning_rate": 8.379888510531535e-05,
|
| 968 |
+
"loss": 0.0253,
|
| 969 |
+
"step": 1120
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 1.7010066798381787,
|
| 973 |
+
"grad_norm": 0.2068171501159668,
|
| 974 |
+
"learning_rate": 8.340962301231981e-05,
|
| 975 |
+
"loss": 0.0214,
|
| 976 |
+
"step": 1130
|
| 977 |
+
},
|
| 978 |
+
{
|
| 979 |
+
"epoch": 1.7160598362969235,
|
| 980 |
+
"grad_norm": 0.12344514578580856,
|
| 981 |
+
"learning_rate": 8.301666769120488e-05,
|
| 982 |
+
"loss": 0.0291,
|
| 983 |
+
"step": 1140
|
| 984 |
+
},
|
| 985 |
+
{
|
| 986 |
+
"epoch": 1.7311129927556683,
|
| 987 |
+
"grad_norm": 0.11504557728767395,
|
| 988 |
+
"learning_rate": 8.262006258076187e-05,
|
| 989 |
+
"loss": 0.0364,
|
| 990 |
+
"step": 1150
|
| 991 |
+
},
|
| 992 |
+
{
|
| 993 |
+
"epoch": 1.7311129927556683,
|
| 994 |
+
"eval_loss": 0.03163878992199898,
|
| 995 |
+
"eval_runtime": 178.2999,
|
| 996 |
+
"eval_samples_per_second": 6.624,
|
| 997 |
+
"eval_steps_per_second": 6.624,
|
| 998 |
+
"step": 1150
|
| 999 |
+
},
|
| 1000 |
+
{
|
| 1001 |
+
"epoch": 1.7461661492144134,
|
| 1002 |
+
"grad_norm": 0.2136172354221344,
|
| 1003 |
+
"learning_rate": 8.221985152324385e-05,
|
| 1004 |
+
"loss": 0.0284,
|
| 1005 |
+
"step": 1160
|
| 1006 |
+
},
|
| 1007 |
+
{
|
| 1008 |
+
"epoch": 1.7612193056731584,
|
| 1009 |
+
"grad_norm": 0.3486640453338623,
|
| 1010 |
+
"learning_rate": 8.18160787595191e-05,
|
| 1011 |
+
"loss": 0.0373,
|
| 1012 |
+
"step": 1170
|
| 1013 |
+
},
|
| 1014 |
+
{
|
| 1015 |
+
"epoch": 1.7762724621319033,
|
| 1016 |
+
"grad_norm": 0.2501416802406311,
|
| 1017 |
+
"learning_rate": 8.14087889241806e-05,
|
| 1018 |
+
"loss": 0.0265,
|
| 1019 |
+
"step": 1180
|
| 1020 |
+
},
|
| 1021 |
+
{
|
| 1022 |
+
"epoch": 1.791325618590648,
|
| 1023 |
+
"grad_norm": 0.21038861572742462,
|
| 1024 |
+
"learning_rate": 8.099802704061195e-05,
|
| 1025 |
+
"loss": 0.0337,
|
| 1026 |
+
"step": 1190
|
| 1027 |
+
},
|
| 1028 |
+
{
|
| 1029 |
+
"epoch": 1.8063787750493931,
|
| 1030 |
+
"grad_norm": 0.3405999541282654,
|
| 1031 |
+
"learning_rate": 8.058383851601027e-05,
|
| 1032 |
+
"loss": 0.0265,
|
| 1033 |
+
"step": 1200
|
| 1034 |
+
},
|
| 1035 |
+
{
|
| 1036 |
+
"epoch": 1.8063787750493931,
|
| 1037 |
+
"eval_loss": 0.032966770231723785,
|
| 1038 |
+
"eval_runtime": 178.2353,
|
| 1039 |
+
"eval_samples_per_second": 6.626,
|
| 1040 |
+
"eval_steps_per_second": 6.626,
|
| 1041 |
+
"step": 1200
|
| 1042 |
+
},
|
| 1043 |
+
{
|
| 1044 |
+
"epoch": 1.8214319315081382,
|
| 1045 |
+
"grad_norm": 0.5199443101882935,
|
| 1046 |
+
"learning_rate": 8.01662691363668e-05,
|
| 1047 |
+
"loss": 0.029,
|
| 1048 |
+
"step": 1210
|
| 1049 |
+
},
|
| 1050 |
+
{
|
| 1051 |
+
"epoch": 1.836485087966883,
|
| 1052 |
+
"grad_norm": 0.35030198097229004,
|
| 1053 |
+
"learning_rate": 7.974536506140547e-05,
|
| 1054 |
+
"loss": 0.0333,
|
| 1055 |
+
"step": 1220
|
| 1056 |
+
},
|
| 1057 |
+
{
|
| 1058 |
+
"epoch": 1.8515382444256279,
|
| 1059 |
+
"grad_norm": 0.27158239483833313,
|
| 1060 |
+
"learning_rate": 7.932117281948021e-05,
|
| 1061 |
+
"loss": 0.0294,
|
| 1062 |
+
"step": 1230
|
| 1063 |
+
},
|
| 1064 |
+
{
|
| 1065 |
+
"epoch": 1.866591400884373,
|
| 1066 |
+
"grad_norm": 0.18654458224773407,
|
| 1067 |
+
"learning_rate": 7.889373930243164e-05,
|
| 1068 |
+
"loss": 0.032,
|
| 1069 |
+
"step": 1240
|
| 1070 |
+
},
|
| 1071 |
+
{
|
| 1072 |
+
"epoch": 1.881644557343118,
|
| 1073 |
+
"grad_norm": 0.13273607194423676,
|
| 1074 |
+
"learning_rate": 7.846311176040331e-05,
|
| 1075 |
+
"loss": 0.0291,
|
| 1076 |
+
"step": 1250
|
| 1077 |
+
},
|
| 1078 |
+
{
|
| 1079 |
+
"epoch": 1.881644557343118,
|
| 1080 |
+
"eval_loss": 0.03071535751223564,
|
| 1081 |
+
"eval_runtime": 178.2583,
|
| 1082 |
+
"eval_samples_per_second": 6.625,
|
| 1083 |
+
"eval_steps_per_second": 6.625,
|
| 1084 |
+
"step": 1250
|
| 1085 |
+
},
|
| 1086 |
+
{
|
| 1087 |
+
"epoch": 1.8966977138018628,
|
| 1088 |
+
"grad_norm": 0.1273525208234787,
|
| 1089 |
+
"learning_rate": 7.802933779661859e-05,
|
| 1090 |
+
"loss": 0.0361,
|
| 1091 |
+
"step": 1260
|
| 1092 |
+
},
|
| 1093 |
+
{
|
| 1094 |
+
"epoch": 1.9117508702606076,
|
| 1095 |
+
"grad_norm": 0.09461209923028946,
|
| 1096 |
+
"learning_rate": 7.759246536211844e-05,
|
| 1097 |
+
"loss": 0.0251,
|
| 1098 |
+
"step": 1270
|
| 1099 |
+
},
|
| 1100 |
+
{
|
| 1101 |
+
"epoch": 1.9268040267193527,
|
| 1102 |
+
"grad_norm": 0.10459864139556885,
|
| 1103 |
+
"learning_rate": 7.715254275046062e-05,
|
| 1104 |
+
"loss": 0.0308,
|
| 1105 |
+
"step": 1280
|
| 1106 |
+
},
|
| 1107 |
+
{
|
| 1108 |
+
"epoch": 1.9418571831780977,
|
| 1109 |
+
"grad_norm": 0.21300184726715088,
|
| 1110 |
+
"learning_rate": 7.670961859238124e-05,
|
| 1111 |
+
"loss": 0.0272,
|
| 1112 |
+
"step": 1290
|
| 1113 |
+
},
|
| 1114 |
+
{
|
| 1115 |
+
"epoch": 1.9569103396368426,
|
| 1116 |
+
"grad_norm": 0.47472086548805237,
|
| 1117 |
+
"learning_rate": 7.626374185041886e-05,
|
| 1118 |
+
"loss": 0.0244,
|
| 1119 |
+
"step": 1300
|
| 1120 |
+
},
|
| 1121 |
+
{
|
| 1122 |
+
"epoch": 1.9569103396368426,
|
| 1123 |
+
"eval_loss": 0.031091291457414627,
|
| 1124 |
+
"eval_runtime": 178.3155,
|
| 1125 |
+
"eval_samples_per_second": 6.623,
|
| 1126 |
+
"eval_steps_per_second": 6.623,
|
| 1127 |
+
"step": 1300
|
| 1128 |
+
},
|
| 1129 |
+
{
|
| 1130 |
+
"epoch": 1.9719634960955874,
|
| 1131 |
+
"grad_norm": 0.29737257957458496,
|
| 1132 |
+
"learning_rate": 7.581496181350203e-05,
|
| 1133 |
+
"loss": 0.0255,
|
| 1134 |
+
"step": 1310
|
| 1135 |
+
},
|
| 1136 |
+
{
|
| 1137 |
+
"epoch": 1.9870166525543325,
|
| 1138 |
+
"grad_norm": 0.18417982757091522,
|
| 1139 |
+
"learning_rate": 7.536332809150067e-05,
|
| 1140 |
+
"loss": 0.0258,
|
| 1141 |
+
"step": 1320
|
| 1142 |
+
},
|
| 1143 |
+
{
|
| 1144 |
+
"epoch": 2.0020698090130775,
|
| 1145 |
+
"grad_norm": 0.27828750014305115,
|
| 1146 |
+
"learning_rate": 7.490889060974201e-05,
|
| 1147 |
+
"loss": 0.0272,
|
| 1148 |
+
"step": 1330
|
| 1149 |
+
},
|
| 1150 |
+
{
|
| 1151 |
+
"epoch": 2.0171229654718226,
|
| 1152 |
+
"grad_norm": 0.194544717669487,
|
| 1153 |
+
"learning_rate": 7.445169960349167e-05,
|
| 1154 |
+
"loss": 0.015,
|
| 1155 |
+
"step": 1340
|
| 1156 |
+
},
|
| 1157 |
+
{
|
| 1158 |
+
"epoch": 2.032176121930567,
|
| 1159 |
+
"grad_norm": 0.21474245190620422,
|
| 1160 |
+
"learning_rate": 7.399180561240044e-05,
|
| 1161 |
+
"loss": 0.0197,
|
| 1162 |
+
"step": 1350
|
| 1163 |
+
},
|
| 1164 |
+
{
|
| 1165 |
+
"epoch": 2.032176121930567,
|
| 1166 |
+
"eval_loss": 0.031300000846385956,
|
| 1167 |
+
"eval_runtime": 178.2318,
|
| 1168 |
+
"eval_samples_per_second": 6.626,
|
| 1169 |
+
"eval_steps_per_second": 6.626,
|
| 1170 |
+
"step": 1350
|
| 1171 |
+
},
|
| 1172 |
+
{
|
| 1173 |
+
"epoch": 2.0472292783893122,
|
| 1174 |
+
"grad_norm": 0.10497045516967773,
|
| 1175 |
+
"learning_rate": 7.352925947491746e-05,
|
| 1176 |
+
"loss": 0.0118,
|
| 1177 |
+
"step": 1360
|
| 1178 |
+
},
|
| 1179 |
+
{
|
| 1180 |
+
"epoch": 2.0622824348480573,
|
| 1181 |
+
"grad_norm": 0.32971543073654175,
|
| 1182 |
+
"learning_rate": 7.306411232267029e-05,
|
| 1183 |
+
"loss": 0.0199,
|
| 1184 |
+
"step": 1370
|
| 1185 |
+
},
|
| 1186 |
+
{
|
| 1187 |
+
"epoch": 2.0773355913068023,
|
| 1188 |
+
"grad_norm": 0.20883502066135406,
|
| 1189 |
+
"learning_rate": 7.259641557481269e-05,
|
| 1190 |
+
"loss": 0.0221,
|
| 1191 |
+
"step": 1380
|
| 1192 |
+
},
|
| 1193 |
+
{
|
| 1194 |
+
"epoch": 2.092388747765547,
|
| 1195 |
+
"grad_norm": 0.12358664721250534,
|
| 1196 |
+
"learning_rate": 7.212622093234049e-05,
|
| 1197 |
+
"loss": 0.0153,
|
| 1198 |
+
"step": 1390
|
| 1199 |
+
},
|
| 1200 |
+
{
|
| 1201 |
+
"epoch": 2.107441904224292,
|
| 1202 |
+
"grad_norm": 0.2762002646923065,
|
| 1203 |
+
"learning_rate": 7.165358037237643e-05,
|
| 1204 |
+
"loss": 0.0223,
|
| 1205 |
+
"step": 1400
|
| 1206 |
+
},
|
| 1207 |
+
{
|
| 1208 |
+
"epoch": 2.107441904224292,
|
| 1209 |
+
"eval_loss": 0.03041531704366207,
|
| 1210 |
+
"eval_runtime": 178.1406,
|
| 1211 |
+
"eval_samples_per_second": 6.63,
|
| 1212 |
+
"eval_steps_per_second": 6.63,
|
| 1213 |
+
"step": 1400
|
| 1214 |
+
},
|
| 1215 |
+
{
|
| 1216 |
+
"epoch": 2.122495060683037,
|
| 1217 |
+
"grad_norm": 0.21662060916423798,
|
| 1218 |
+
"learning_rate": 7.117854614242434e-05,
|
| 1219 |
+
"loss": 0.0208,
|
| 1220 |
+
"step": 1410
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 2.137548217141782,
|
| 1224 |
+
"grad_norm": 0.2806037366390228,
|
| 1225 |
+
"learning_rate": 7.070117075459352e-05,
|
| 1226 |
+
"loss": 0.0203,
|
| 1227 |
+
"step": 1420
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"epoch": 2.1526013736005267,
|
| 1231 |
+
"grad_norm": 0.17611578106880188,
|
| 1232 |
+
"learning_rate": 7.022150697979384e-05,
|
| 1233 |
+
"loss": 0.0226,
|
| 1234 |
+
"step": 1430
|
| 1235 |
+
},
|
| 1236 |
+
{
|
| 1237 |
+
"epoch": 2.167654530059272,
|
| 1238 |
+
"grad_norm": 0.22745737433433533,
|
| 1239 |
+
"learning_rate": 6.973960784190237e-05,
|
| 1240 |
+
"loss": 0.023,
|
| 1241 |
+
"step": 1440
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 2.182707686518017,
|
| 1245 |
+
"grad_norm": 0.2118409425020218,
|
| 1246 |
+
"learning_rate": 6.925552661190166e-05,
|
| 1247 |
+
"loss": 0.0197,
|
| 1248 |
+
"step": 1450
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 2.182707686518017,
|
| 1252 |
+
"eval_loss": 0.03106178157031536,
|
| 1253 |
+
"eval_runtime": 178.1235,
|
| 1254 |
+
"eval_samples_per_second": 6.63,
|
| 1255 |
+
"eval_steps_per_second": 6.63,
|
| 1256 |
+
"step": 1450
|
| 1257 |
+
},
|
| 1258 |
+
{
|
| 1259 |
+
"epoch": 2.197760842976762,
|
| 1260 |
+
"grad_norm": 0.1830424964427948,
|
| 1261 |
+
"learning_rate": 6.876931680199121e-05,
|
| 1262 |
+
"loss": 0.0156,
|
| 1263 |
+
"step": 1460
|
| 1264 |
+
},
|
| 1265 |
+
{
|
| 1266 |
+
"epoch": 2.2128139994355065,
|
| 1267 |
+
"grad_norm": 0.10300840437412262,
|
| 1268 |
+
"learning_rate": 6.828103215967186e-05,
|
| 1269 |
+
"loss": 0.0203,
|
| 1270 |
+
"step": 1470
|
| 1271 |
+
},
|
| 1272 |
+
{
|
| 1273 |
+
"epoch": 2.2278671558942515,
|
| 1274 |
+
"grad_norm": 0.17760959267616272,
|
| 1275 |
+
"learning_rate": 6.779072666180446e-05,
|
| 1276 |
+
"loss": 0.0234,
|
| 1277 |
+
"step": 1480
|
| 1278 |
+
},
|
| 1279 |
+
{
|
| 1280 |
+
"epoch": 2.2429203123529966,
|
| 1281 |
+
"grad_norm": 0.05315260961651802,
|
| 1282 |
+
"learning_rate": 6.729845450864294e-05,
|
| 1283 |
+
"loss": 0.0159,
|
| 1284 |
+
"step": 1490
|
| 1285 |
+
},
|
| 1286 |
+
{
|
| 1287 |
+
"epoch": 2.257973468811741,
|
| 1288 |
+
"grad_norm": 0.21539124846458435,
|
| 1289 |
+
"learning_rate": 6.680427011784292e-05,
|
| 1290 |
+
"loss": 0.0204,
|
| 1291 |
+
"step": 1500
|
| 1292 |
+
},
|
| 1293 |
+
{
|
| 1294 |
+
"epoch": 2.257973468811741,
|
| 1295 |
+
"eval_loss": 0.0323634073138237,
|
| 1296 |
+
"eval_runtime": 178.1357,
|
| 1297 |
+
"eval_samples_per_second": 6.63,
|
| 1298 |
+
"eval_steps_per_second": 6.63,
|
| 1299 |
+
"step": 1500
|
| 1300 |
+
},
|
| 1301 |
+
{
|
| 1302 |
+
"epoch": 2.2730266252704863,
|
| 1303 |
+
"grad_norm": 0.24310541152954102,
|
| 1304 |
+
"learning_rate": 6.630822811844604e-05,
|
| 1305 |
+
"loss": 0.017,
|
| 1306 |
+
"step": 1510
|
| 1307 |
+
},
|
| 1308 |
+
{
|
| 1309 |
+
"epoch": 2.2880797817292313,
|
| 1310 |
+
"grad_norm": 0.1696529984474182,
|
| 1311 |
+
"learning_rate": 6.58103833448412e-05,
|
| 1312 |
+
"loss": 0.0222,
|
| 1313 |
+
"step": 1520
|
| 1314 |
+
},
|
| 1315 |
+
{
|
| 1316 |
+
"epoch": 2.3031329381879764,
|
| 1317 |
+
"grad_norm": 0.3020496666431427,
|
| 1318 |
+
"learning_rate": 6.531079083070288e-05,
|
| 1319 |
+
"loss": 0.0234,
|
| 1320 |
+
"step": 1530
|
| 1321 |
+
},
|
| 1322 |
+
{
|
| 1323 |
+
"epoch": 2.3181860946467214,
|
| 1324 |
+
"grad_norm": 0.12914994359016418,
|
| 1325 |
+
"learning_rate": 6.480950580290752e-05,
|
| 1326 |
+
"loss": 0.017,
|
| 1327 |
+
"step": 1540
|
| 1328 |
+
},
|
| 1329 |
+
{
|
| 1330 |
+
"epoch": 2.333239251105466,
|
| 1331 |
+
"grad_norm": 0.18893304467201233,
|
| 1332 |
+
"learning_rate": 6.430658367542843e-05,
|
| 1333 |
+
"loss": 0.0276,
|
| 1334 |
+
"step": 1550
|
| 1335 |
+
},
|
| 1336 |
+
{
|
| 1337 |
+
"epoch": 2.333239251105466,
|
| 1338 |
+
"eval_loss": 0.031088994815945625,
|
| 1339 |
+
"eval_runtime": 178.1568,
|
| 1340 |
+
"eval_samples_per_second": 6.629,
|
| 1341 |
+
"eval_steps_per_second": 6.629,
|
| 1342 |
+
"step": 1550
|
| 1343 |
+
},
|
| 1344 |
+
{
|
| 1345 |
+
"epoch": 2.348292407564211,
|
| 1346 |
+
"grad_norm": 0.08729825168848038,
|
| 1347 |
+
"learning_rate": 6.380208004321036e-05,
|
| 1348 |
+
"loss": 0.0247,
|
| 1349 |
+
"step": 1560
|
| 1350 |
+
},
|
| 1351 |
+
{
|
| 1352 |
+
"epoch": 2.363345564022956,
|
| 1353 |
+
"grad_norm": 0.12183941900730133,
|
| 1354 |
+
"learning_rate": 6.32960506760236e-05,
|
| 1355 |
+
"loss": 0.0181,
|
| 1356 |
+
"step": 1570
|
| 1357 |
+
},
|
| 1358 |
+
{
|
| 1359 |
+
"epoch": 2.378398720481701,
|
| 1360 |
+
"grad_norm": 0.2509489953517914,
|
| 1361 |
+
"learning_rate": 6.278855151229901e-05,
|
| 1362 |
+
"loss": 0.0158,
|
| 1363 |
+
"step": 1580
|
| 1364 |
+
},
|
| 1365 |
+
{
|
| 1366 |
+
"epoch": 2.393451876940446,
|
| 1367 |
+
"grad_norm": 0.0914265587925911,
|
| 1368 |
+
"learning_rate": 6.227963865294444e-05,
|
| 1369 |
+
"loss": 0.0168,
|
| 1370 |
+
"step": 1590
|
| 1371 |
+
},
|
| 1372 |
+
{
|
| 1373 |
+
"epoch": 2.408505033399191,
|
| 1374 |
+
"grad_norm": 0.11629947274923325,
|
| 1375 |
+
"learning_rate": 6.176936835514312e-05,
|
| 1376 |
+
"loss": 0.0146,
|
| 1377 |
+
"step": 1600
|
| 1378 |
+
},
|
| 1379 |
+
{
|
| 1380 |
+
"epoch": 2.408505033399191,
|
| 1381 |
+
"eval_loss": 0.031647298485040665,
|
| 1382 |
+
"eval_runtime": 178.063,
|
| 1383 |
+
"eval_samples_per_second": 6.632,
|
| 1384 |
+
"eval_steps_per_second": 6.632,
|
| 1385 |
+
"step": 1600
|
| 1386 |
+
},
|
| 1387 |
+
{
|
| 1388 |
+
"epoch": 2.423558189857936,
|
| 1389 |
+
"grad_norm": 0.2622014582157135,
|
| 1390 |
+
"learning_rate": 6.125779702613471e-05,
|
| 1391 |
+
"loss": 0.0176,
|
| 1392 |
+
"step": 1610
|
| 1393 |
+
},
|
| 1394 |
+
{
|
| 1395 |
+
"epoch": 2.438611346316681,
|
| 1396 |
+
"grad_norm": 0.177884042263031,
|
| 1397 |
+
"learning_rate": 6.074498121697983e-05,
|
| 1398 |
+
"loss": 0.0207,
|
| 1399 |
+
"step": 1620
|
| 1400 |
+
},
|
| 1401 |
+
{
|
| 1402 |
+
"epoch": 2.4536645027754256,
|
| 1403 |
+
"grad_norm": 0.20466935634613037,
|
| 1404 |
+
"learning_rate": 6.023097761630879e-05,
|
| 1405 |
+
"loss": 0.0254,
|
| 1406 |
+
"step": 1630
|
| 1407 |
+
},
|
| 1408 |
+
{
|
| 1409 |
+
"epoch": 2.4687176592341706,
|
| 1410 |
+
"grad_norm": 0.1373012661933899,
|
| 1411 |
+
"learning_rate": 5.971584304405489e-05,
|
| 1412 |
+
"loss": 0.0164,
|
| 1413 |
+
"step": 1640
|
| 1414 |
+
},
|
| 1415 |
+
{
|
| 1416 |
+
"epoch": 2.4837708156929157,
|
| 1417 |
+
"grad_norm": 0.23096352815628052,
|
| 1418 |
+
"learning_rate": 5.919963444517338e-05,
|
| 1419 |
+
"loss": 0.0165,
|
| 1420 |
+
"step": 1650
|
| 1421 |
+
},
|
| 1422 |
+
{
|
| 1423 |
+
"epoch": 2.4837708156929157,
|
| 1424 |
+
"eval_loss": 0.03288305178284645,
|
| 1425 |
+
"eval_runtime": 178.0988,
|
| 1426 |
+
"eval_samples_per_second": 6.631,
|
| 1427 |
+
"eval_steps_per_second": 6.631,
|
| 1428 |
+
"step": 1650
|
| 1429 |
+
},
|
| 1430 |
+
{
|
| 1431 |
+
"epoch": 2.4988239721516607,
|
| 1432 |
+
"grad_norm": 0.2017652541399002,
|
| 1433 |
+
"learning_rate": 5.868240888334653e-05,
|
| 1434 |
+
"loss": 0.0181,
|
| 1435 |
+
"step": 1660
|
| 1436 |
+
},
|
| 1437 |
+
{
|
| 1438 |
+
"epoch": 2.5138771286104054,
|
| 1439 |
+
"grad_norm": 0.07514988631010056,
|
| 1440 |
+
"learning_rate": 5.816422353467562e-05,
|
| 1441 |
+
"loss": 0.0148,
|
| 1442 |
+
"step": 1670
|
| 1443 |
+
},
|
| 1444 |
+
{
|
| 1445 |
+
"epoch": 2.5289302850691504,
|
| 1446 |
+
"grad_norm": 0.11077038943767548,
|
| 1447 |
+
"learning_rate": 5.7645135681360496e-05,
|
| 1448 |
+
"loss": 0.0183,
|
| 1449 |
+
"step": 1680
|
| 1450 |
+
},
|
| 1451 |
+
{
|
| 1452 |
+
"epoch": 2.5439834415278955,
|
| 1453 |
+
"grad_norm": 0.24594064056873322,
|
| 1454 |
+
"learning_rate": 5.7125202705367234e-05,
|
| 1455 |
+
"loss": 0.0228,
|
| 1456 |
+
"step": 1690
|
| 1457 |
+
},
|
| 1458 |
+
{
|
| 1459 |
+
"epoch": 2.5590365979866405,
|
| 1460 |
+
"grad_norm": 0.2879762649536133,
|
| 1461 |
+
"learning_rate": 5.660448208208513e-05,
|
| 1462 |
+
"loss": 0.0242,
|
| 1463 |
+
"step": 1700
|
| 1464 |
+
},
|
| 1465 |
+
{
|
| 1466 |
+
"epoch": 2.5590365979866405,
|
| 1467 |
+
"eval_loss": 0.03144305199384689,
|
| 1468 |
+
"eval_runtime": 178.0564,
|
| 1469 |
+
"eval_samples_per_second": 6.633,
|
| 1470 |
+
"eval_steps_per_second": 6.633,
|
| 1471 |
+
"step": 1700
|
| 1472 |
+
},
|
| 1473 |
+
{
|
| 1474 |
+
"epoch": 2.574089754445385,
|
| 1475 |
+
"grad_norm": 0.060580622404813766,
|
| 1476 |
+
"learning_rate": 5.608303137397294e-05,
|
| 1477 |
+
"loss": 0.0156,
|
| 1478 |
+
"step": 1710
|
| 1479 |
+
},
|
| 1480 |
+
{
|
| 1481 |
+
"epoch": 2.58914291090413,
|
| 1482 |
+
"grad_norm": 0.13122287392616272,
|
| 1483 |
+
"learning_rate": 5.5560908224195886e-05,
|
| 1484 |
+
"loss": 0.0218,
|
| 1485 |
+
"step": 1720
|
| 1486 |
+
},
|
| 1487 |
+
{
|
| 1488 |
+
"epoch": 2.6041960673628752,
|
| 1489 |
+
"grad_norm": 0.2391199916601181,
|
| 1490 |
+
"learning_rate": 5.503817035025342e-05,
|
| 1491 |
+
"loss": 0.0187,
|
| 1492 |
+
"step": 1730
|
| 1493 |
+
},
|
| 1494 |
+
{
|
| 1495 |
+
"epoch": 2.61924922382162,
|
| 1496 |
+
"grad_norm": 0.2503703236579895,
|
| 1497 |
+
"learning_rate": 5.4514875537598985e-05,
|
| 1498 |
+
"loss": 0.0143,
|
| 1499 |
+
"step": 1740
|
| 1500 |
+
},
|
| 1501 |
+
{
|
| 1502 |
+
"epoch": 2.634302380280365,
|
| 1503 |
+
"grad_norm": 0.2618468105792999,
|
| 1504 |
+
"learning_rate": 5.399108163325217e-05,
|
| 1505 |
+
"loss": 0.0273,
|
| 1506 |
+
"step": 1750
|
| 1507 |
+
},
|
| 1508 |
+
{
|
| 1509 |
+
"epoch": 2.634302380280365,
|
| 1510 |
+
"eval_loss": 0.032075975090265274,
|
| 1511 |
+
"eval_runtime": 178.055,
|
| 1512 |
+
"eval_samples_per_second": 6.633,
|
| 1513 |
+
"eval_steps_per_second": 6.633,
|
| 1514 |
+
"step": 1750
|
| 1515 |
+
},
|
| 1516 |
+
{
|
| 1517 |
+
"epoch": 2.64935553673911,
|
| 1518 |
+
"grad_norm": 0.15629079937934875,
|
| 1519 |
+
"learning_rate": 5.346684653940408e-05,
|
| 1520 |
+
"loss": 0.0217,
|
| 1521 |
+
"step": 1760
|
| 1522 |
+
},
|
| 1523 |
+
{
|
| 1524 |
+
"epoch": 2.664408693197855,
|
| 1525 |
+
"grad_norm": 0.09079688042402267,
|
| 1526 |
+
"learning_rate": 5.294222820701661e-05,
|
| 1527 |
+
"loss": 0.0204,
|
| 1528 |
+
"step": 1770
|
| 1529 |
+
},
|
| 1530 |
+
{
|
| 1531 |
+
"epoch": 2.6794618496566,
|
| 1532 |
+
"grad_norm": 0.29432734847068787,
|
| 1533 |
+
"learning_rate": 5.24172846294163e-05,
|
| 1534 |
+
"loss": 0.0148,
|
| 1535 |
+
"step": 1780
|
| 1536 |
+
},
|
| 1537 |
+
{
|
| 1538 |
+
"epoch": 2.6945150061153447,
|
| 1539 |
+
"grad_norm": 0.17906472086906433,
|
| 1540 |
+
"learning_rate": 5.1892073835883524e-05,
|
| 1541 |
+
"loss": 0.0189,
|
| 1542 |
+
"step": 1790
|
| 1543 |
+
},
|
| 1544 |
+
{
|
| 1545 |
+
"epoch": 2.7095681625740897,
|
| 1546 |
+
"grad_norm": 0.37694233655929565,
|
| 1547 |
+
"learning_rate": 5.136665388523778e-05,
|
| 1548 |
+
"loss": 0.0204,
|
| 1549 |
+
"step": 1800
|
| 1550 |
+
},
|
| 1551 |
+
{
|
| 1552 |
+
"epoch": 2.7095681625740897,
|
| 1553 |
+
"eval_loss": 0.03231995552778244,
|
| 1554 |
+
"eval_runtime": 177.9391,
|
| 1555 |
+
"eval_samples_per_second": 6.637,
|
| 1556 |
+
"eval_steps_per_second": 6.637,
|
| 1557 |
+
"step": 1800
|
| 1558 |
+
},
|
| 1559 |
+
{
|
| 1560 |
+
"epoch": 2.7246213190328348,
|
| 1561 |
+
"grad_norm": 0.15220922231674194,
|
| 1562 |
+
"learning_rate": 5.0841082859419585e-05,
|
| 1563 |
+
"loss": 0.0243,
|
| 1564 |
+
"step": 1810
|
| 1565 |
+
},
|
| 1566 |
+
{
|
| 1567 |
+
"epoch": 2.7396744754915794,
|
| 1568 |
+
"grad_norm": 0.1869482398033142,
|
| 1569 |
+
"learning_rate": 5.031541885706987e-05,
|
| 1570 |
+
"loss": 0.0215,
|
| 1571 |
+
"step": 1820
|
| 1572 |
+
},
|
| 1573 |
+
{
|
| 1574 |
+
"epoch": 2.7547276319503244,
|
| 1575 |
+
"grad_norm": 0.12930846214294434,
|
| 1576 |
+
"learning_rate": 4.9789719987107545e-05,
|
| 1577 |
+
"loss": 0.0223,
|
| 1578 |
+
"step": 1830
|
| 1579 |
+
},
|
| 1580 |
+
{
|
| 1581 |
+
"epoch": 2.7697807884090695,
|
| 1582 |
+
"grad_norm": 0.4493294060230255,
|
| 1583 |
+
"learning_rate": 4.926404436230596e-05,
|
| 1584 |
+
"loss": 0.0201,
|
| 1585 |
+
"step": 1840
|
| 1586 |
+
},
|
| 1587 |
+
{
|
| 1588 |
+
"epoch": 2.7848339448678145,
|
| 1589 |
+
"grad_norm": 0.3657067120075226,
|
| 1590 |
+
"learning_rate": 4.8738450092868785e-05,
|
| 1591 |
+
"loss": 0.021,
|
| 1592 |
+
"step": 1850
|
| 1593 |
+
},
|
| 1594 |
+
{
|
| 1595 |
+
"epoch": 2.7848339448678145,
|
| 1596 |
+
"eval_loss": 0.03066716156899929,
|
| 1597 |
+
"eval_runtime": 177.9619,
|
| 1598 |
+
"eval_samples_per_second": 6.636,
|
| 1599 |
+
"eval_steps_per_second": 6.636,
|
| 1600 |
+
"step": 1850
|
| 1601 |
+
},
|
| 1602 |
+
{
|
| 1603 |
+
"epoch": 2.7998871013265596,
|
| 1604 |
+
"grad_norm": 0.35673412680625916,
|
| 1605 |
+
"learning_rate": 4.8212995280006426e-05,
|
| 1606 |
+
"loss": 0.0186,
|
| 1607 |
+
"step": 1860
|
| 1608 |
+
},
|
| 1609 |
+
{
|
| 1610 |
+
"epoch": 2.814940257785304,
|
| 1611 |
+
"grad_norm": 0.20455271005630493,
|
| 1612 |
+
"learning_rate": 4.76877380095132e-05,
|
| 1613 |
+
"loss": 0.0183,
|
| 1614 |
+
"step": 1870
|
| 1615 |
+
},
|
| 1616 |
+
{
|
| 1617 |
+
"epoch": 2.8299934142440493,
|
| 1618 |
+
"grad_norm": 0.25493428111076355,
|
| 1619 |
+
"learning_rate": 4.7162736345346303e-05,
|
| 1620 |
+
"loss": 0.0149,
|
| 1621 |
+
"step": 1880
|
| 1622 |
+
},
|
| 1623 |
+
{
|
| 1624 |
+
"epoch": 2.8450465707027943,
|
| 1625 |
+
"grad_norm": 0.25855717062950134,
|
| 1626 |
+
"learning_rate": 4.663804832320726e-05,
|
| 1627 |
+
"loss": 0.0134,
|
| 1628 |
+
"step": 1890
|
| 1629 |
+
},
|
| 1630 |
+
{
|
| 1631 |
+
"epoch": 2.860099727161539,
|
| 1632 |
+
"grad_norm": 0.16750505566596985,
|
| 1633 |
+
"learning_rate": 4.6113731944126406e-05,
|
| 1634 |
+
"loss": 0.0195,
|
| 1635 |
+
"step": 1900
|
| 1636 |
+
},
|
| 1637 |
+
{
|
| 1638 |
+
"epoch": 2.860099727161539,
|
| 1639 |
+
"eval_loss": 0.033523257821798325,
|
| 1640 |
+
"eval_runtime": 178.3199,
|
| 1641 |
+
"eval_samples_per_second": 6.623,
|
| 1642 |
+
"eval_steps_per_second": 6.623,
|
| 1643 |
+
"step": 1900
|
| 1644 |
+
},
|
| 1645 |
+
{
|
| 1646 |
+
"epoch": 2.875152883620284,
|
| 1647 |
+
"grad_norm": 0.17557500302791595,
|
| 1648 |
+
"learning_rate": 4.558984516805118e-05,
|
| 1649 |
+
"loss": 0.0177,
|
| 1650 |
+
"step": 1910
|
| 1651 |
+
},
|
| 1652 |
+
{
|
| 1653 |
+
"epoch": 2.890206040079029,
|
| 1654 |
+
"grad_norm": 0.32916489243507385,
|
| 1655 |
+
"learning_rate": 4.5066445907439104e-05,
|
| 1656 |
+
"loss": 0.0132,
|
| 1657 |
+
"step": 1920
|
| 1658 |
+
},
|
| 1659 |
+
{
|
| 1660 |
+
"epoch": 2.905259196537774,
|
| 1661 |
+
"grad_norm": 0.3346385359764099,
|
| 1662 |
+
"learning_rate": 4.454359202085582e-05,
|
| 1663 |
+
"loss": 0.0233,
|
| 1664 |
+
"step": 1930
|
| 1665 |
+
},
|
| 1666 |
+
{
|
| 1667 |
+
"epoch": 2.920312352996519,
|
| 1668 |
+
"grad_norm": 0.5354165434837341,
|
| 1669 |
+
"learning_rate": 4.402134130657925e-05,
|
| 1670 |
+
"loss": 0.0163,
|
| 1671 |
+
"step": 1940
|
| 1672 |
+
},
|
| 1673 |
+
{
|
| 1674 |
+
"epoch": 2.9353655094552638,
|
| 1675 |
+
"grad_norm": 0.24359606206417084,
|
| 1676 |
+
"learning_rate": 4.349975149621039e-05,
|
| 1677 |
+
"loss": 0.0221,
|
| 1678 |
+
"step": 1950
|
| 1679 |
+
},
|
| 1680 |
+
{
|
| 1681 |
+
"epoch": 2.9353655094552638,
|
| 1682 |
+
"eval_loss": 0.030188467353582382,
|
| 1683 |
+
"eval_runtime": 177.9643,
|
| 1684 |
+
"eval_samples_per_second": 6.636,
|
| 1685 |
+
"eval_steps_per_second": 6.636,
|
| 1686 |
+
"step": 1950
|
| 1687 |
+
},
|
| 1688 |
+
{
|
| 1689 |
+
"epoch": 2.950418665914009,
|
| 1690 |
+
"grad_norm": 0.11338020116090775,
|
| 1691 |
+
"learning_rate": 4.297888024829126e-05,
|
| 1692 |
+
"loss": 0.0163,
|
| 1693 |
+
"step": 1960
|
| 1694 |
+
},
|
| 1695 |
+
{
|
| 1696 |
+
"epoch": 2.965471822372754,
|
| 1697 |
+
"grad_norm": 0.1252482682466507,
|
| 1698 |
+
"learning_rate": 4.2458785141931314e-05,
|
| 1699 |
+
"loss": 0.0181,
|
| 1700 |
+
"step": 1970
|
| 1701 |
+
},
|
| 1702 |
+
{
|
| 1703 |
+
"epoch": 2.9805249788314985,
|
| 1704 |
+
"grad_norm": 0.0980340763926506,
|
| 1705 |
+
"learning_rate": 4.1939523670442316e-05,
|
| 1706 |
+
"loss": 0.0171,
|
| 1707 |
+
"step": 1980
|
| 1708 |
+
},
|
| 1709 |
+
{
|
| 1710 |
+
"epoch": 2.9955781352902435,
|
| 1711 |
+
"grad_norm": 0.1608283817768097,
|
| 1712 |
+
"learning_rate": 4.14211532349828e-05,
|
| 1713 |
+
"loss": 0.0177,
|
| 1714 |
+
"step": 1990
|
| 1715 |
+
},
|
| 1716 |
+
{
|
| 1717 |
+
"epoch": 3.0106312917489886,
|
| 1718 |
+
"grad_norm": 0.07570704817771912,
|
| 1719 |
+
"learning_rate": 4.090373113821281e-05,
|
| 1720 |
+
"loss": 0.0129,
|
| 1721 |
+
"step": 2000
|
| 1722 |
+
},
|
| 1723 |
+
{
|
| 1724 |
+
"epoch": 3.0106312917489886,
|
| 1725 |
+
"eval_loss": 0.03050847351551056,
|
| 1726 |
+
"eval_runtime": 177.9373,
|
| 1727 |
+
"eval_samples_per_second": 6.637,
|
| 1728 |
+
"eval_steps_per_second": 6.637,
|
| 1729 |
+
"step": 2000
|
| 1730 |
+
},
|
| 1731 |
+
{
|
| 1732 |
+
"epoch": 3.0256844482077336,
|
| 1733 |
+
"grad_norm": 0.15597139298915863,
|
| 1734 |
+
"learning_rate": 4.0387314577959315e-05,
|
| 1735 |
+
"loss": 0.0113,
|
| 1736 |
+
"step": 2010
|
| 1737 |
+
},
|
| 1738 |
+
{
|
| 1739 |
+
"epoch": 3.0407376046664787,
|
| 1740 |
+
"grad_norm": 0.055645011365413666,
|
| 1741 |
+
"learning_rate": 3.987196064089346e-05,
|
| 1742 |
+
"loss": 0.0091,
|
| 1743 |
+
"step": 2020
|
| 1744 |
+
},
|
| 1745 |
+
{
|
| 1746 |
+
"epoch": 3.0557907611252233,
|
| 1747 |
+
"grad_norm": 0.10545477271080017,
|
| 1748 |
+
"learning_rate": 3.935772629621995e-05,
|
| 1749 |
+
"loss": 0.0079,
|
| 1750 |
+
"step": 2030
|
| 1751 |
+
},
|
| 1752 |
+
{
|
| 1753 |
+
"epoch": 3.0708439175839684,
|
| 1754 |
+
"grad_norm": 0.14946284890174866,
|
| 1755 |
+
"learning_rate": 3.8844668389379396e-05,
|
| 1756 |
+
"loss": 0.0102,
|
| 1757 |
+
"step": 2040
|
| 1758 |
+
},
|
| 1759 |
+
{
|
| 1760 |
+
"epoch": 3.0858970740427134,
|
| 1761 |
+
"grad_norm": 0.2742091119289398,
|
| 1762 |
+
"learning_rate": 3.833284363576447e-05,
|
| 1763 |
+
"loss": 0.0127,
|
| 1764 |
+
"step": 2050
|
| 1765 |
+
},
|
| 1766 |
+
{
|
| 1767 |
+
"epoch": 3.0858970740427134,
|
| 1768 |
+
"eval_loss": 0.03302207961678505,
|
| 1769 |
+
"eval_runtime": 178.0745,
|
| 1770 |
+
"eval_samples_per_second": 6.632,
|
| 1771 |
+
"eval_steps_per_second": 6.632,
|
| 1772 |
+
"step": 2050
|
| 1773 |
+
},
|
| 1774 |
+
{
|
| 1775 |
+
"epoch": 3.1009502305014585,
|
| 1776 |
+
"grad_norm": 0.19018790125846863,
|
| 1777 |
+
"learning_rate": 3.7822308614450406e-05,
|
| 1778 |
+
"loss": 0.0096,
|
| 1779 |
+
"step": 2060
|
| 1780 |
+
},
|
| 1781 |
+
{
|
| 1782 |
+
"epoch": 3.116003386960203,
|
| 1783 |
+
"grad_norm": 0.04711037501692772,
|
| 1784 |
+
"learning_rate": 3.7313119761940375e-05,
|
| 1785 |
+
"loss": 0.0142,
|
| 1786 |
+
"step": 2070
|
| 1787 |
+
},
|
| 1788 |
+
{
|
| 1789 |
+
"epoch": 3.131056543418948,
|
| 1790 |
+
"grad_norm": 0.22224950790405273,
|
| 1791 |
+
"learning_rate": 3.680533336592694e-05,
|
| 1792 |
+
"loss": 0.0091,
|
| 1793 |
+
"step": 2080
|
| 1794 |
+
},
|
| 1795 |
+
{
|
| 1796 |
+
"epoch": 3.146109699877693,
|
| 1797 |
+
"grad_norm": 0.09899937361478806,
|
| 1798 |
+
"learning_rate": 3.62990055590697e-05,
|
| 1799 |
+
"loss": 0.0101,
|
| 1800 |
+
"step": 2090
|
| 1801 |
+
},
|
| 1802 |
+
{
|
| 1803 |
+
"epoch": 3.1611628563364382,
|
| 1804 |
+
"grad_norm": 0.09017643332481384,
|
| 1805 |
+
"learning_rate": 3.579419231279023e-05,
|
| 1806 |
+
"loss": 0.0122,
|
| 1807 |
+
"step": 2100
|
| 1808 |
+
},
|
| 1809 |
+
{
|
| 1810 |
+
"epoch": 3.1611628563364382,
|
| 1811 |
+
"eval_loss": 0.03305284306406975,
|
| 1812 |
+
"eval_runtime": 178.3274,
|
| 1813 |
+
"eval_samples_per_second": 6.623,
|
| 1814 |
+
"eval_steps_per_second": 6.623,
|
| 1815 |
+
"step": 2100
|
| 1816 |
+
},
|
| 1817 |
+
{
|
| 1818 |
+
"epoch": 3.176216012795183,
|
| 1819 |
+
"grad_norm": 0.12764662504196167,
|
| 1820 |
+
"learning_rate": 3.529094943108475e-05,
|
| 1821 |
+
"loss": 0.0085,
|
| 1822 |
+
"step": 2110
|
| 1823 |
+
},
|
| 1824 |
+
{
|
| 1825 |
+
"epoch": 3.191269169253928,
|
| 1826 |
+
"grad_norm": 0.12719255685806274,
|
| 1827 |
+
"learning_rate": 3.478933254435534e-05,
|
| 1828 |
+
"loss": 0.0144,
|
| 1829 |
+
"step": 2120
|
| 1830 |
+
},
|
| 1831 |
+
{
|
| 1832 |
+
"epoch": 3.206322325712673,
|
| 1833 |
+
"grad_norm": 0.164920374751091,
|
| 1834 |
+
"learning_rate": 3.4289397103260346e-05,
|
| 1835 |
+
"loss": 0.0125,
|
| 1836 |
+
"step": 2130
|
| 1837 |
+
},
|
| 1838 |
+
{
|
| 1839 |
+
"epoch": 3.221375482171418,
|
| 1840 |
+
"grad_norm": 0.22153641283512115,
|
| 1841 |
+
"learning_rate": 3.3791198372584664e-05,
|
| 1842 |
+
"loss": 0.0129,
|
| 1843 |
+
"step": 2140
|
| 1844 |
+
},
|
| 1845 |
+
{
|
| 1846 |
+
"epoch": 3.2364286386301626,
|
| 1847 |
+
"grad_norm": 0.11729541420936584,
|
| 1848 |
+
"learning_rate": 3.329479142513051e-05,
|
| 1849 |
+
"loss": 0.0129,
|
| 1850 |
+
"step": 2150
|
| 1851 |
+
},
|
| 1852 |
+
{
|
| 1853 |
+
"epoch": 3.2364286386301626,
|
| 1854 |
+
"eval_loss": 0.03324244171380997,
|
| 1855 |
+
"eval_runtime": 178.3386,
|
| 1856 |
+
"eval_samples_per_second": 6.622,
|
| 1857 |
+
"eval_steps_per_second": 6.622,
|
| 1858 |
+
"step": 2150
|
| 1859 |
+
},
|
| 1860 |
+
{
|
| 1861 |
+
"epoch": 3.2514817950889077,
|
| 1862 |
+
"grad_norm": 0.344887912273407,
|
| 1863 |
+
"learning_rate": 3.280023113562957e-05,
|
| 1864 |
+
"loss": 0.0136,
|
| 1865 |
+
"step": 2160
|
| 1866 |
+
},
|
| 1867 |
+
{
|
| 1868 |
+
"epoch": 3.2665349515476527,
|
| 1869 |
+
"grad_norm": 0.1558815985918045,
|
| 1870 |
+
"learning_rate": 3.230757217467677e-05,
|
| 1871 |
+
"loss": 0.0171,
|
| 1872 |
+
"step": 2170
|
| 1873 |
+
},
|
| 1874 |
+
{
|
| 1875 |
+
"epoch": 3.2815881080063978,
|
| 1876 |
+
"grad_norm": 0.19974495470523834,
|
| 1877 |
+
"learning_rate": 3.1816869002686936e-05,
|
| 1878 |
+
"loss": 0.0108,
|
| 1879 |
+
"step": 2180
|
| 1880 |
+
},
|
| 1881 |
+
{
|
| 1882 |
+
"epoch": 3.2966412644651424,
|
| 1883 |
+
"grad_norm": 0.13594548404216766,
|
| 1884 |
+
"learning_rate": 3.1328175863874464e-05,
|
| 1885 |
+
"loss": 0.0086,
|
| 1886 |
+
"step": 2190
|
| 1887 |
+
},
|
| 1888 |
+
{
|
| 1889 |
+
"epoch": 3.3116944209238874,
|
| 1890 |
+
"grad_norm": 0.21001030504703522,
|
| 1891 |
+
"learning_rate": 3.084154678025692e-05,
|
| 1892 |
+
"loss": 0.0125,
|
| 1893 |
+
"step": 2200
|
| 1894 |
+
},
|
| 1895 |
+
{
|
| 1896 |
+
"epoch": 3.3116944209238874,
|
| 1897 |
+
"eval_loss": 0.03326999396085739,
|
| 1898 |
+
"eval_runtime": 178.6788,
|
| 1899 |
+
"eval_samples_per_second": 6.61,
|
| 1900 |
+
"eval_steps_per_second": 6.61,
|
| 1901 |
+
"step": 2200
|
| 1902 |
+
},
|
| 1903 |
+
{
|
| 1904 |
+
"epoch": 3.3267475773826325,
|
| 1905 |
+
"grad_norm": 0.2704913914203644,
|
| 1906 |
+
"learning_rate": 3.035703554568331e-05,
|
| 1907 |
+
"loss": 0.0126,
|
| 1908 |
+
"step": 2210
|
| 1909 |
+
},
|
| 1910 |
+
{
|
| 1911 |
+
"epoch": 3.3418007338413775,
|
| 1912 |
+
"grad_norm": 0.1079394668340683,
|
| 1913 |
+
"learning_rate": 2.9874695719887464e-05,
|
| 1914 |
+
"loss": 0.0109,
|
| 1915 |
+
"step": 2220
|
| 1916 |
+
},
|
| 1917 |
+
{
|
| 1918 |
+
"epoch": 3.356853890300122,
|
| 1919 |
+
"grad_norm": 0.3655363917350769,
|
| 1920 |
+
"learning_rate": 2.9394580622567312e-05,
|
| 1921 |
+
"loss": 0.0076,
|
| 1922 |
+
"step": 2230
|
| 1923 |
+
},
|
| 1924 |
+
{
|
| 1925 |
+
"epoch": 3.371907046758867,
|
| 1926 |
+
"grad_norm": 0.09557830542325974,
|
| 1927 |
+
"learning_rate": 2.8916743327490803e-05,
|
| 1928 |
+
"loss": 0.0116,
|
| 1929 |
+
"step": 2240
|
| 1930 |
+
},
|
| 1931 |
+
{
|
| 1932 |
+
"epoch": 3.3869602032176123,
|
| 1933 |
+
"grad_norm": 0.22106143832206726,
|
| 1934 |
+
"learning_rate": 2.8441236656628828e-05,
|
| 1935 |
+
"loss": 0.0132,
|
| 1936 |
+
"step": 2250
|
| 1937 |
+
},
|
| 1938 |
+
{
|
| 1939 |
+
"epoch": 3.3869602032176123,
|
| 1940 |
+
"eval_loss": 0.03426986560225487,
|
| 1941 |
+
"eval_runtime": 179.2113,
|
| 1942 |
+
"eval_samples_per_second": 6.59,
|
| 1943 |
+
"eval_steps_per_second": 6.59,
|
| 1944 |
+
"step": 2250
|
| 1945 |
+
},
|
| 1946 |
+
{
|
| 1947 |
+
"epoch": 3.4020133596763573,
|
| 1948 |
+
"grad_norm": 0.1936042606830597,
|
| 1949 |
+
"learning_rate": 2.79681131743161e-05,
|
| 1950 |
+
"loss": 0.0064,
|
| 1951 |
+
"step": 2260
|
| 1952 |
+
},
|
| 1953 |
+
{
|
| 1954 |
+
"epoch": 3.417066516135102,
|
| 1955 |
+
"grad_norm": 0.18822342157363892,
|
| 1956 |
+
"learning_rate": 2.7497425181440607e-05,
|
| 1957 |
+
"loss": 0.0131,
|
| 1958 |
+
"step": 2270
|
| 1959 |
+
},
|
| 1960 |
+
{
|
| 1961 |
+
"epoch": 3.432119672593847,
|
| 1962 |
+
"grad_norm": 0.13977766036987305,
|
| 1963 |
+
"learning_rate": 2.702922470966187e-05,
|
| 1964 |
+
"loss": 0.0067,
|
| 1965 |
+
"step": 2280
|
| 1966 |
+
},
|
| 1967 |
+
{
|
| 1968 |
+
"epoch": 3.447172829052592,
|
| 1969 |
+
"grad_norm": 0.1867881566286087,
|
| 1970 |
+
"learning_rate": 2.6563563515659306e-05,
|
| 1971 |
+
"loss": 0.0104,
|
| 1972 |
+
"step": 2290
|
| 1973 |
+
},
|
| 1974 |
+
{
|
| 1975 |
+
"epoch": 3.462225985511337,
|
| 1976 |
+
"grad_norm": 0.035174448043107986,
|
| 1977 |
+
"learning_rate": 2.6100493075410848e-05,
|
| 1978 |
+
"loss": 0.0084,
|
| 1979 |
+
"step": 2300
|
| 1980 |
+
},
|
| 1981 |
+
{
|
| 1982 |
+
"epoch": 3.462225985511337,
|
| 1983 |
+
"eval_loss": 0.03515166416764259,
|
| 1984 |
+
"eval_runtime": 179.2383,
|
| 1985 |
+
"eval_samples_per_second": 6.589,
|
| 1986 |
+
"eval_steps_per_second": 6.589,
|
| 1987 |
+
"step": 2300
|
| 1988 |
+
},
|
| 1989 |
+
{
|
| 1990 |
+
"epoch": 3.4772791419700817,
|
| 1991 |
+
"grad_norm": 0.3701817989349365,
|
| 1992 |
+
"learning_rate": 2.5640064578502497e-05,
|
| 1993 |
+
"loss": 0.0153,
|
| 1994 |
+
"step": 2310
|
| 1995 |
+
},
|
| 1996 |
+
{
|
| 1997 |
+
"epoch": 3.4923322984288268,
|
| 1998 |
+
"grad_norm": 0.25794854760169983,
|
| 1999 |
+
"learning_rate": 2.5182328922469723e-05,
|
| 2000 |
+
"loss": 0.0108,
|
| 2001 |
+
"step": 2320
|
| 2002 |
+
},
|
| 2003 |
+
{
|
| 2004 |
+
"epoch": 3.507385454887572,
|
| 2005 |
+
"grad_norm": 0.29802244901657104,
|
| 2006 |
+
"learning_rate": 2.4727336707170973e-05,
|
| 2007 |
+
"loss": 0.0102,
|
| 2008 |
+
"step": 2330
|
| 2009 |
+
},
|
| 2010 |
+
{
|
| 2011 |
+
"epoch": 3.522438611346317,
|
| 2012 |
+
"grad_norm": 0.10990750789642334,
|
| 2013 |
+
"learning_rate": 2.427513822919424e-05,
|
| 2014 |
+
"loss": 0.0115,
|
| 2015 |
+
"step": 2340
|
| 2016 |
+
},
|
| 2017 |
+
{
|
| 2018 |
+
"epoch": 3.5374917678050615,
|
| 2019 |
+
"grad_norm": 0.282993346452713,
|
| 2020 |
+
"learning_rate": 2.3825783476297087e-05,
|
| 2021 |
+
"loss": 0.0086,
|
| 2022 |
+
"step": 2350
|
| 2023 |
+
},
|
| 2024 |
+
{
|
| 2025 |
+
"epoch": 3.5374917678050615,
|
| 2026 |
+
"eval_loss": 0.03473927080631256,
|
| 2027 |
+
"eval_runtime": 179.3483,
|
| 2028 |
+
"eval_samples_per_second": 6.585,
|
| 2029 |
+
"eval_steps_per_second": 6.585,
|
| 2030 |
+
"step": 2350
|
| 2031 |
+
},
|
| 2032 |
+
{
|
| 2033 |
+
"epoch": 3.5525449242638065,
|
| 2034 |
+
"grad_norm": 0.27041804790496826,
|
| 2035 |
+
"learning_rate": 2.337932212188073e-05,
|
| 2036 |
+
"loss": 0.0109,
|
| 2037 |
+
"step": 2360
|
| 2038 |
+
},
|
| 2039 |
+
{
|
| 2040 |
+
"epoch": 3.5675980807225516,
|
| 2041 |
+
"grad_norm": 0.2581341862678528,
|
| 2042 |
+
"learning_rate": 2.2935803519499e-05,
|
| 2043 |
+
"loss": 0.0122,
|
| 2044 |
+
"step": 2370
|
| 2045 |
+
},
|
| 2046 |
+
{
|
| 2047 |
+
"epoch": 3.582651237181296,
|
| 2048 |
+
"grad_norm": 0.13770440220832825,
|
| 2049 |
+
"learning_rate": 2.2495276697402662e-05,
|
| 2050 |
+
"loss": 0.0083,
|
| 2051 |
+
"step": 2380
|
| 2052 |
+
},
|
| 2053 |
+
{
|
| 2054 |
+
"epoch": 3.5977043936400412,
|
| 2055 |
+
"grad_norm": 0.13981565833091736,
|
| 2056 |
+
"learning_rate": 2.2057790353119535e-05,
|
| 2057 |
+
"loss": 0.0105,
|
| 2058 |
+
"step": 2390
|
| 2059 |
+
},
|
| 2060 |
+
{
|
| 2061 |
+
"epoch": 3.6127575500987863,
|
| 2062 |
+
"grad_norm": 0.27494239807128906,
|
| 2063 |
+
"learning_rate": 2.1623392848071354e-05,
|
| 2064 |
+
"loss": 0.0112,
|
| 2065 |
+
"step": 2400
|
| 2066 |
+
},
|
| 2067 |
+
{
|
| 2068 |
+
"epoch": 3.6127575500987863,
|
| 2069 |
+
"eval_loss": 0.03513345867395401,
|
| 2070 |
+
"eval_runtime": 179.5449,
|
| 2071 |
+
"eval_samples_per_second": 6.578,
|
| 2072 |
+
"eval_steps_per_second": 6.578,
|
| 2073 |
+
"step": 2400
|
| 2074 |
+
},
|
| 2075 |
+
{
|
| 2076 |
+
"epoch": 3.6278107065575314,
|
| 2077 |
+
"grad_norm": 0.18222074210643768,
|
| 2078 |
+
"learning_rate": 2.1192132202227677e-05,
|
| 2079 |
+
"loss": 0.0103,
|
| 2080 |
+
"step": 2410
|
| 2081 |
+
},
|
| 2082 |
+
{
|
| 2083 |
+
"epoch": 3.6428638630162764,
|
| 2084 |
+
"grad_norm": 0.27348509430885315,
|
| 2085 |
+
"learning_rate": 2.0764056088797645e-05,
|
| 2086 |
+
"loss": 0.0096,
|
| 2087 |
+
"step": 2420
|
| 2088 |
+
},
|
| 2089 |
+
{
|
| 2090 |
+
"epoch": 3.657917019475021,
|
| 2091 |
+
"grad_norm": 0.39703041315078735,
|
| 2092 |
+
"learning_rate": 2.0339211828959904e-05,
|
| 2093 |
+
"loss": 0.0098,
|
| 2094 |
+
"step": 2430
|
| 2095 |
+
},
|
| 2096 |
+
{
|
| 2097 |
+
"epoch": 3.672970175933766,
|
| 2098 |
+
"grad_norm": 0.21773886680603027,
|
| 2099 |
+
"learning_rate": 1.9917646386631577e-05,
|
| 2100 |
+
"loss": 0.0094,
|
| 2101 |
+
"step": 2440
|
| 2102 |
+
},
|
| 2103 |
+
{
|
| 2104 |
+
"epoch": 3.688023332392511,
|
| 2105 |
+
"grad_norm": 0.16702532768249512,
|
| 2106 |
+
"learning_rate": 1.949940636327671e-05,
|
| 2107 |
+
"loss": 0.012,
|
| 2108 |
+
"step": 2450
|
| 2109 |
+
},
|
| 2110 |
+
{
|
| 2111 |
+
"epoch": 3.688023332392511,
|
| 2112 |
+
"eval_loss": 0.03648751974105835,
|
| 2113 |
+
"eval_runtime": 179.7341,
|
| 2114 |
+
"eval_samples_per_second": 6.571,
|
| 2115 |
+
"eval_steps_per_second": 6.571,
|
| 2116 |
+
"step": 2450
|
| 2117 |
+
},
|
| 2118 |
+
{
|
| 2119 |
+
"epoch": 3.7030764888512557,
|
| 2120 |
+
"grad_norm": 0.32012760639190674,
|
| 2121 |
+
"learning_rate": 1.9084537992754792e-05,
|
| 2122 |
+
"loss": 0.01,
|
| 2123 |
+
"step": 2460
|
| 2124 |
+
},
|
| 2125 |
+
{
|
| 2126 |
+
"epoch": 3.718129645310001,
|
| 2127 |
+
"grad_norm": 0.24596737325191498,
|
| 2128 |
+
"learning_rate": 1.8673087136209803e-05,
|
| 2129 |
+
"loss": 0.0076,
|
| 2130 |
+
"step": 2470
|
| 2131 |
+
},
|
| 2132 |
+
{
|
| 2133 |
+
"epoch": 3.733182801768746,
|
| 2134 |
+
"grad_norm": 0.07540785521268845,
|
| 2135 |
+
"learning_rate": 1.8265099277000614e-05,
|
| 2136 |
+
"loss": 0.0067,
|
| 2137 |
+
"step": 2480
|
| 2138 |
+
},
|
| 2139 |
+
{
|
| 2140 |
+
"epoch": 3.748235958227491,
|
| 2141 |
+
"grad_norm": 0.2560862600803375,
|
| 2142 |
+
"learning_rate": 1.7860619515673033e-05,
|
| 2143 |
+
"loss": 0.0107,
|
| 2144 |
+
"step": 2490
|
| 2145 |
+
},
|
| 2146 |
+
{
|
| 2147 |
+
"epoch": 3.763289114686236,
|
| 2148 |
+
"grad_norm": 0.09346114844083786,
|
| 2149 |
+
"learning_rate": 1.7459692564974316e-05,
|
| 2150 |
+
"loss": 0.011,
|
| 2151 |
+
"step": 2500
|
| 2152 |
+
},
|
| 2153 |
+
{
|
| 2154 |
+
"epoch": 3.763289114686236,
|
| 2155 |
+
"eval_loss": 0.03612372279167175,
|
| 2156 |
+
"eval_runtime": 179.868,
|
| 2157 |
+
"eval_samples_per_second": 6.566,
|
| 2158 |
+
"eval_steps_per_second": 6.566,
|
| 2159 |
+
"step": 2500
|
| 2160 |
+
},
|
| 2161 |
+
{
|
| 2162 |
+
"epoch": 3.7783422711449806,
|
| 2163 |
+
"grad_norm": 0.07731559872627258,
|
| 2164 |
+
"learning_rate": 1.7062362744910322e-05,
|
| 2165 |
+
"loss": 0.0059,
|
| 2166 |
+
"step": 2510
|
| 2167 |
+
},
|
| 2168 |
+
{
|
| 2169 |
+
"epoch": 3.7933954276037256,
|
| 2170 |
+
"grad_norm": 0.35534602403640747,
|
| 2171 |
+
"learning_rate": 1.6668673977846254e-05,
|
| 2172 |
+
"loss": 0.0122,
|
| 2173 |
+
"step": 2520
|
| 2174 |
+
},
|
| 2175 |
+
{
|
| 2176 |
+
"epoch": 3.8084485840624707,
|
| 2177 |
+
"grad_norm": 0.4245568513870239,
|
| 2178 |
+
"learning_rate": 1.6278669783651395e-05,
|
| 2179 |
+
"loss": 0.0089,
|
| 2180 |
+
"step": 2530
|
| 2181 |
+
},
|
| 2182 |
+
{
|
| 2183 |
+
"epoch": 3.8235017405212153,
|
| 2184 |
+
"grad_norm": 0.0772927924990654,
|
| 2185 |
+
"learning_rate": 1.589239327488812e-05,
|
| 2186 |
+
"loss": 0.0051,
|
| 2187 |
+
"step": 2540
|
| 2188 |
+
},
|
| 2189 |
+
{
|
| 2190 |
+
"epoch": 3.8385548969799603,
|
| 2191 |
+
"grad_norm": 0.1755327433347702,
|
| 2192 |
+
"learning_rate": 1.5509887152046137e-05,
|
| 2193 |
+
"loss": 0.0079,
|
| 2194 |
+
"step": 2550
|
| 2195 |
+
},
|
| 2196 |
+
{
|
| 2197 |
+
"epoch": 3.8385548969799603,
|
| 2198 |
+
"eval_loss": 0.03893940895795822,
|
| 2199 |
+
"eval_runtime": 180.0653,
|
| 2200 |
+
"eval_samples_per_second": 6.559,
|
| 2201 |
+
"eval_steps_per_second": 6.559,
|
| 2202 |
+
"step": 2550
|
| 2203 |
+
},
|
| 2204 |
+
{
|
| 2205 |
+
"epoch": 3.8536080534387054,
|
| 2206 |
+
"grad_norm": 0.23755531013011932,
|
| 2207 |
+
"learning_rate": 1.5131193698822232e-05,
|
| 2208 |
+
"loss": 0.0137,
|
| 2209 |
+
"step": 2560
|
| 2210 |
+
},
|
| 2211 |
+
{
|
| 2212 |
+
"epoch": 3.8686612098974504,
|
| 2213 |
+
"grad_norm": 0.21640031039714813,
|
| 2214 |
+
"learning_rate": 1.4756354777446001e-05,
|
| 2215 |
+
"loss": 0.0089,
|
| 2216 |
+
"step": 2570
|
| 2217 |
+
},
|
| 2218 |
+
{
|
| 2219 |
+
"epoch": 3.8837143663561955,
|
| 2220 |
+
"grad_norm": 0.3311944305896759,
|
| 2221 |
+
"learning_rate": 1.4385411824052342e-05,
|
| 2222 |
+
"loss": 0.0093,
|
| 2223 |
+
"step": 2580
|
| 2224 |
+
},
|
| 2225 |
+
{
|
| 2226 |
+
"epoch": 3.89876752281494,
|
| 2227 |
+
"grad_norm": 0.2809070646762848,
|
| 2228 |
+
"learning_rate": 1.4018405844100812e-05,
|
| 2229 |
+
"loss": 0.0123,
|
| 2230 |
+
"step": 2590
|
| 2231 |
+
},
|
| 2232 |
+
{
|
| 2233 |
+
"epoch": 3.913820679273685,
|
| 2234 |
+
"grad_norm": 0.19817812740802765,
|
| 2235 |
+
"learning_rate": 1.3655377407842812e-05,
|
| 2236 |
+
"loss": 0.0089,
|
| 2237 |
+
"step": 2600
|
| 2238 |
+
},
|
| 2239 |
+
{
|
| 2240 |
+
"epoch": 3.913820679273685,
|
| 2241 |
+
"eval_loss": 0.03675345331430435,
|
| 2242 |
+
"eval_runtime": 180.0963,
|
| 2243 |
+
"eval_samples_per_second": 6.558,
|
| 2244 |
+
"eval_steps_per_second": 6.558,
|
| 2245 |
+
"step": 2600
|
| 2246 |
+
},
|
| 2247 |
+
{
|
| 2248 |
+
"epoch": 3.92887383573243,
|
| 2249 |
+
"grad_norm": 0.15485577285289764,
|
| 2250 |
+
"learning_rate": 1.3296366645836822e-05,
|
| 2251 |
+
"loss": 0.0074,
|
| 2252 |
+
"step": 2610
|
| 2253 |
+
},
|
| 2254 |
+
{
|
| 2255 |
+
"epoch": 3.943926992191175,
|
| 2256 |
+
"grad_norm": 0.2657122313976288,
|
| 2257 |
+
"learning_rate": 1.2941413244512113e-05,
|
| 2258 |
+
"loss": 0.0128,
|
| 2259 |
+
"step": 2620
|
| 2260 |
+
},
|
| 2261 |
+
{
|
| 2262 |
+
"epoch": 3.95898014864992,
|
| 2263 |
+
"grad_norm": 0.3065131902694702,
|
| 2264 |
+
"learning_rate": 1.2590556441781725e-05,
|
| 2265 |
+
"loss": 0.0077,
|
| 2266 |
+
"step": 2630
|
| 2267 |
+
},
|
| 2268 |
+
{
|
| 2269 |
+
"epoch": 3.974033305108665,
|
| 2270 |
+
"grad_norm": 0.487378865480423,
|
| 2271 |
+
"learning_rate": 1.2243835022705003e-05,
|
| 2272 |
+
"loss": 0.0115,
|
| 2273 |
+
"step": 2640
|
| 2274 |
+
},
|
| 2275 |
+
{
|
| 2276 |
+
"epoch": 3.98908646156741,
|
| 2277 |
+
"grad_norm": 0.25239965319633484,
|
| 2278 |
+
"learning_rate": 1.1901287315199977e-05,
|
| 2279 |
+
"loss": 0.009,
|
| 2280 |
+
"step": 2650
|
| 2281 |
+
},
|
| 2282 |
+
{
|
| 2283 |
+
"epoch": 3.98908646156741,
|
| 2284 |
+
"eval_loss": 0.03644752502441406,
|
| 2285 |
+
"eval_runtime": 180.2417,
|
| 2286 |
+
"eval_samples_per_second": 6.552,
|
| 2287 |
+
"eval_steps_per_second": 6.552,
|
| 2288 |
+
"step": 2650
|
| 2289 |
+
},
|
| 2290 |
+
{
|
| 2291 |
+
"epoch": 4.004139618026155,
|
| 2292 |
+
"grad_norm": 0.24973592162132263,
|
| 2293 |
+
"learning_rate": 1.1562951185806676e-05,
|
| 2294 |
+
"loss": 0.0078,
|
| 2295 |
+
"step": 2660
|
| 2296 |
+
},
|
| 2297 |
+
{
|
| 2298 |
+
"epoch": 4.0191927744849,
|
| 2299 |
+
"grad_norm": 0.11293565481901169,
|
| 2300 |
+
"learning_rate": 1.1228864035501069e-05,
|
| 2301 |
+
"loss": 0.0064,
|
| 2302 |
+
"step": 2670
|
| 2303 |
+
},
|
| 2304 |
+
{
|
| 2305 |
+
"epoch": 4.034245930943645,
|
| 2306 |
+
"grad_norm": 0.24446244537830353,
|
| 2307 |
+
"learning_rate": 1.0899062795560573e-05,
|
| 2308 |
+
"loss": 0.0042,
|
| 2309 |
+
"step": 2680
|
| 2310 |
+
},
|
| 2311 |
+
{
|
| 2312 |
+
"epoch": 4.04929908740239,
|
| 2313 |
+
"grad_norm": 0.29575204849243164,
|
| 2314 |
+
"learning_rate": 1.0573583923481711e-05,
|
| 2315 |
+
"loss": 0.0087,
|
| 2316 |
+
"step": 2690
|
| 2317 |
+
},
|
| 2318 |
+
{
|
| 2319 |
+
"epoch": 4.064352243861134,
|
| 2320 |
+
"grad_norm": 0.18624891340732574,
|
| 2321 |
+
"learning_rate": 1.0252463398949792e-05,
|
| 2322 |
+
"loss": 0.0086,
|
| 2323 |
+
"step": 2700
|
| 2324 |
+
},
|
| 2325 |
+
{
|
| 2326 |
+
"epoch": 4.064352243861134,
|
| 2327 |
+
"eval_loss": 0.037868741899728775,
|
| 2328 |
+
"eval_runtime": 180.3651,
|
| 2329 |
+
"eval_samples_per_second": 6.548,
|
| 2330 |
+
"eval_steps_per_second": 6.548,
|
| 2331 |
+
"step": 2700
|
| 2332 |
+
},
|
| 2333 |
+
{
|
| 2334 |
+
"epoch": 4.07940540031988,
|
| 2335 |
+
"grad_norm": 0.10995467007160187,
|
| 2336 |
+
"learning_rate": 9.935736719861622e-06,
|
| 2337 |
+
"loss": 0.0053,
|
| 2338 |
+
"step": 2710
|
| 2339 |
+
},
|
| 2340 |
+
{
|
| 2341 |
+
"epoch": 4.0944585567786245,
|
| 2342 |
+
"grad_norm": 0.09725473076105118,
|
| 2343 |
+
"learning_rate": 9.62343889840151e-06,
|
| 2344 |
+
"loss": 0.0051,
|
| 2345 |
+
"step": 2720
|
| 2346 |
+
},
|
| 2347 |
+
{
|
| 2348 |
+
"epoch": 4.109511713237369,
|
| 2349 |
+
"grad_norm": 0.036554962396621704,
|
| 2350 |
+
"learning_rate": 9.315604457170768e-06,
|
| 2351 |
+
"loss": 0.0033,
|
| 2352 |
+
"step": 2730
|
| 2353 |
+
},
|
| 2354 |
+
{
|
| 2355 |
+
"epoch": 4.124564869696115,
|
| 2356 |
+
"grad_norm": 0.10706653445959091,
|
| 2357 |
+
"learning_rate": 9.012267425371513e-06,
|
| 2358 |
+
"loss": 0.0053,
|
| 2359 |
+
"step": 2740
|
| 2360 |
+
},
|
| 2361 |
+
{
|
| 2362 |
+
"epoch": 4.139618026154859,
|
| 2363 |
+
"grad_norm": 0.11202232539653778,
|
| 2364 |
+
"learning_rate": 8.71346133504498e-06,
|
| 2365 |
+
"loss": 0.005,
|
| 2366 |
+
"step": 2750
|
| 2367 |
+
},
|
| 2368 |
+
{
|
| 2369 |
+
"epoch": 4.139618026154859,
|
| 2370 |
+
"eval_loss": 0.03884652629494667,
|
| 2371 |
+
"eval_runtime": 180.4257,
|
| 2372 |
+
"eval_samples_per_second": 6.546,
|
| 2373 |
+
"eval_steps_per_second": 6.546,
|
| 2374 |
+
"step": 2750
|
| 2375 |
+
},
|
| 2376 |
+
{
|
| 2377 |
+
"epoch": 4.154671182613605,
|
| 2378 |
+
"grad_norm": 0.05641934648156166,
|
| 2379 |
+
"learning_rate": 8.419219217364654e-06,
|
| 2380 |
+
"loss": 0.0037,
|
| 2381 |
+
"step": 2760
|
| 2382 |
+
},
|
| 2383 |
+
{
|
| 2384 |
+
"epoch": 4.169724339072349,
|
| 2385 |
+
"grad_norm": 0.17391756176948547,
|
| 2386 |
+
"learning_rate": 8.129573598984997e-06,
|
| 2387 |
+
"loss": 0.0054,
|
| 2388 |
+
"step": 2770
|
| 2389 |
+
},
|
| 2390 |
+
{
|
| 2391 |
+
"epoch": 4.184777495531094,
|
| 2392 |
+
"grad_norm": 0.08039785176515579,
|
| 2393 |
+
"learning_rate": 7.844556498445788e-06,
|
| 2394 |
+
"loss": 0.0046,
|
| 2395 |
+
"step": 2780
|
| 2396 |
+
},
|
| 2397 |
+
{
|
| 2398 |
+
"epoch": 4.199830651989839,
|
| 2399 |
+
"grad_norm": 0.24674569070339203,
|
| 2400 |
+
"learning_rate": 7.564199422632579e-06,
|
| 2401 |
+
"loss": 0.0067,
|
| 2402 |
+
"step": 2790
|
| 2403 |
+
},
|
| 2404 |
+
{
|
| 2405 |
+
"epoch": 4.214883808448584,
|
| 2406 |
+
"grad_norm": 0.05541248247027397,
|
| 2407 |
+
"learning_rate": 7.288533363293959e-06,
|
| 2408 |
+
"loss": 0.0051,
|
| 2409 |
+
"step": 2800
|
| 2410 |
+
},
|
| 2411 |
+
{
|
| 2412 |
+
"epoch": 4.214883808448584,
|
| 2413 |
+
"eval_loss": 0.0397312231361866,
|
| 2414 |
+
"eval_runtime": 180.5412,
|
| 2415 |
+
"eval_samples_per_second": 6.541,
|
| 2416 |
+
"eval_steps_per_second": 6.541,
|
| 2417 |
+
"step": 2800
|
| 2418 |
+
},
|
| 2419 |
+
{
|
| 2420 |
+
"epoch": 4.229936964907329,
|
| 2421 |
+
"grad_norm": 0.30977103114128113,
|
| 2422 |
+
"learning_rate": 7.017588793615498e-06,
|
| 2423 |
+
"loss": 0.004,
|
| 2424 |
+
"step": 2810
|
| 2425 |
+
},
|
| 2426 |
+
{
|
| 2427 |
+
"epoch": 4.244990121366074,
|
| 2428 |
+
"grad_norm": 0.145084947347641,
|
| 2429 |
+
"learning_rate": 6.751395664851135e-06,
|
| 2430 |
+
"loss": 0.0038,
|
| 2431 |
+
"step": 2820
|
| 2432 |
+
},
|
| 2433 |
+
{
|
| 2434 |
+
"epoch": 4.260043277824819,
|
| 2435 |
+
"grad_norm": 0.1771937906742096,
|
| 2436 |
+
"learning_rate": 6.489983403012312e-06,
|
| 2437 |
+
"loss": 0.0047,
|
| 2438 |
+
"step": 2830
|
| 2439 |
+
},
|
| 2440 |
+
{
|
| 2441 |
+
"epoch": 4.275096434283564,
|
| 2442 |
+
"grad_norm": 0.30625805258750916,
|
| 2443 |
+
"learning_rate": 6.233380905615049e-06,
|
| 2444 |
+
"loss": 0.0042,
|
| 2445 |
+
"step": 2840
|
| 2446 |
+
},
|
| 2447 |
+
{
|
| 2448 |
+
"epoch": 4.290149590742309,
|
| 2449 |
+
"grad_norm": 0.15609732270240784,
|
| 2450 |
+
"learning_rate": 5.981616538485496e-06,
|
| 2451 |
+
"loss": 0.0055,
|
| 2452 |
+
"step": 2850
|
| 2453 |
+
},
|
| 2454 |
+
{
|
| 2455 |
+
"epoch": 4.290149590742309,
|
| 2456 |
+
"eval_loss": 0.04038526862859726,
|
| 2457 |
+
"eval_runtime": 180.0932,
|
| 2458 |
+
"eval_samples_per_second": 6.558,
|
| 2459 |
+
"eval_steps_per_second": 6.558,
|
| 2460 |
+
"step": 2850
|
| 2461 |
+
},
|
| 2462 |
+
{
|
| 2463 |
+
"epoch": 4.3052027472010534,
|
| 2464 |
+
"grad_norm": 0.150795578956604,
|
| 2465 |
+
"learning_rate": 5.73471813262435e-06,
|
| 2466 |
+
"loss": 0.0035,
|
| 2467 |
+
"step": 2860
|
| 2468 |
+
},
|
| 2469 |
+
{
|
| 2470 |
+
"epoch": 4.320255903659799,
|
| 2471 |
+
"grad_norm": 0.26626893877983093,
|
| 2472 |
+
"learning_rate": 5.4927129811301715e-06,
|
| 2473 |
+
"loss": 0.0035,
|
| 2474 |
+
"step": 2870
|
| 2475 |
+
},
|
| 2476 |
+
{
|
| 2477 |
+
"epoch": 4.335309060118544,
|
| 2478 |
+
"grad_norm": 0.042985837906599045,
|
| 2479 |
+
"learning_rate": 5.255627836182453e-06,
|
| 2480 |
+
"loss": 0.0024,
|
| 2481 |
+
"step": 2880
|
| 2482 |
+
},
|
| 2483 |
+
{
|
| 2484 |
+
"epoch": 4.350362216577288,
|
| 2485 |
+
"grad_norm": 0.2361251413822174,
|
| 2486 |
+
"learning_rate": 5.0234889060842176e-06,
|
| 2487 |
+
"loss": 0.0056,
|
| 2488 |
+
"step": 2890
|
| 2489 |
+
},
|
| 2490 |
+
{
|
| 2491 |
+
"epoch": 4.365415373036034,
|
| 2492 |
+
"grad_norm": 0.12534025311470032,
|
| 2493 |
+
"learning_rate": 4.796321852364877e-06,
|
| 2494 |
+
"loss": 0.0083,
|
| 2495 |
+
"step": 2900
|
| 2496 |
+
},
|
| 2497 |
+
{
|
| 2498 |
+
"epoch": 4.365415373036034,
|
| 2499 |
+
"eval_loss": 0.04092538729310036,
|
| 2500 |
+
"eval_runtime": 180.6762,
|
| 2501 |
+
"eval_samples_per_second": 6.537,
|
| 2502 |
+
"eval_steps_per_second": 6.537,
|
| 2503 |
+
"step": 2900
|
| 2504 |
+
},
|
| 2505 |
+
{
|
| 2506 |
+
"epoch": 4.380468529494778,
|
| 2507 |
+
"grad_norm": 0.10099391639232635,
|
| 2508 |
+
"learning_rate": 4.5741517869435706e-06,
|
| 2509 |
+
"loss": 0.0033,
|
| 2510 |
+
"step": 2910
|
| 2511 |
+
},
|
| 2512 |
+
{
|
| 2513 |
+
"epoch": 4.395521685953524,
|
| 2514 |
+
"grad_norm": 0.037613652646541595,
|
| 2515 |
+
"learning_rate": 4.357003269353105e-06,
|
| 2516 |
+
"loss": 0.0035,
|
| 2517 |
+
"step": 2920
|
| 2518 |
+
},
|
| 2519 |
+
{
|
| 2520 |
+
"epoch": 4.410574842412268,
|
| 2521 |
+
"grad_norm": 0.10834306478500366,
|
| 2522 |
+
"learning_rate": 4.144900304025101e-06,
|
| 2523 |
+
"loss": 0.0057,
|
| 2524 |
+
"step": 2930
|
| 2525 |
+
},
|
| 2526 |
+
{
|
| 2527 |
+
"epoch": 4.425627998871013,
|
| 2528 |
+
"grad_norm": 0.4370887577533722,
|
| 2529 |
+
"learning_rate": 3.937866337636459e-06,
|
| 2530 |
+
"loss": 0.0077,
|
| 2531 |
+
"step": 2940
|
| 2532 |
+
},
|
| 2533 |
+
{
|
| 2534 |
+
"epoch": 4.4406811553297585,
|
| 2535 |
+
"grad_norm": 0.2118665874004364,
|
| 2536 |
+
"learning_rate": 3.7359242565174423e-06,
|
| 2537 |
+
"loss": 0.0055,
|
| 2538 |
+
"step": 2950
|
| 2539 |
+
},
|
| 2540 |
+
{
|
| 2541 |
+
"epoch": 4.4406811553297585,
|
| 2542 |
+
"eval_loss": 0.04117078334093094,
|
| 2543 |
+
"eval_runtime": 180.7007,
|
| 2544 |
+
"eval_samples_per_second": 6.536,
|
| 2545 |
+
"eval_steps_per_second": 6.536,
|
| 2546 |
+
"step": 2950
|
| 2547 |
+
},
|
| 2548 |
+
{
|
| 2549 |
+
"epoch": 4.455734311788503,
|
| 2550 |
+
"grad_norm": 0.11663982272148132,
|
| 2551 |
+
"learning_rate": 3.539096384121743e-06,
|
| 2552 |
+
"loss": 0.005,
|
| 2553 |
+
"step": 2960
|
| 2554 |
+
},
|
| 2555 |
+
{
|
| 2556 |
+
"epoch": 4.470787468247248,
|
| 2557 |
+
"grad_norm": 0.1609174907207489,
|
| 2558 |
+
"learning_rate": 3.34740447855878e-06,
|
| 2559 |
+
"loss": 0.0025,
|
| 2560 |
+
"step": 2970
|
| 2561 |
+
},
|
| 2562 |
+
{
|
| 2563 |
+
"epoch": 4.485840624705993,
|
| 2564 |
+
"grad_norm": 0.2631778419017792,
|
| 2565 |
+
"learning_rate": 3.160869730188465e-06,
|
| 2566 |
+
"loss": 0.0046,
|
| 2567 |
+
"step": 2980
|
| 2568 |
+
},
|
| 2569 |
+
{
|
| 2570 |
+
"epoch": 4.500893781164738,
|
| 2571 |
+
"grad_norm": 0.05932645872235298,
|
| 2572 |
+
"learning_rate": 2.9795127592787186e-06,
|
| 2573 |
+
"loss": 0.0054,
|
| 2574 |
+
"step": 2990
|
| 2575 |
+
},
|
| 2576 |
+
{
|
| 2577 |
+
"epoch": 4.515946937623482,
|
| 2578 |
+
"grad_norm": 0.5128821134567261,
|
| 2579 |
+
"learning_rate": 2.803353613726056e-06,
|
| 2580 |
+
"loss": 0.0056,
|
| 2581 |
+
"step": 3000
|
| 2582 |
+
},
|
| 2583 |
+
{
|
| 2584 |
+
"epoch": 4.515946937623482,
|
| 2585 |
+
"eval_loss": 0.041392985731363297,
|
| 2586 |
+
"eval_runtime": 180.7978,
|
| 2587 |
+
"eval_samples_per_second": 6.532,
|
| 2588 |
+
"eval_steps_per_second": 6.532,
|
| 2589 |
+
"step": 3000
|
| 2590 |
+
},
|
| 2591 |
+
{
|
| 2592 |
+
"epoch": 4.531000094082228,
|
| 2593 |
+
"grad_norm": 0.12355174124240875,
|
| 2594 |
+
"learning_rate": 2.6324117668393877e-06,
|
| 2595 |
+
"loss": 0.0044,
|
| 2596 |
+
"step": 3010
|
| 2597 |
+
},
|
| 2598 |
+
{
|
| 2599 |
+
"epoch": 4.5460532505409725,
|
| 2600 |
+
"grad_norm": 0.32771605253219604,
|
| 2601 |
+
"learning_rate": 2.466706115187406e-06,
|
| 2602 |
+
"loss": 0.0029,
|
| 2603 |
+
"step": 3020
|
| 2604 |
+
},
|
| 2605 |
+
{
|
| 2606 |
+
"epoch": 4.561106406999718,
|
| 2607 |
+
"grad_norm": 0.15814469754695892,
|
| 2608 |
+
"learning_rate": 2.3062549765096364e-06,
|
| 2609 |
+
"loss": 0.0028,
|
| 2610 |
+
"step": 3030
|
| 2611 |
+
},
|
| 2612 |
+
{
|
| 2613 |
+
"epoch": 4.576159563458463,
|
| 2614 |
+
"grad_norm": 0.20907792448997498,
|
| 2615 |
+
"learning_rate": 2.1510760876915505e-06,
|
| 2616 |
+
"loss": 0.0071,
|
| 2617 |
+
"step": 3040
|
| 2618 |
+
},
|
| 2619 |
+
{
|
| 2620 |
+
"epoch": 4.591212719917207,
|
| 2621 |
+
"grad_norm": 0.3278658986091614,
|
| 2622 |
+
"learning_rate": 2.0011866028038617e-06,
|
| 2623 |
+
"loss": 0.01,
|
| 2624 |
+
"step": 3050
|
| 2625 |
+
},
|
| 2626 |
+
{
|
| 2627 |
+
"epoch": 4.591212719917207,
|
| 2628 |
+
"eval_loss": 0.04168102145195007,
|
| 2629 |
+
"eval_runtime": 180.8128,
|
| 2630 |
+
"eval_samples_per_second": 6.532,
|
| 2631 |
+
"eval_steps_per_second": 6.532,
|
| 2632 |
+
"step": 3050
|
| 2633 |
+
},
|
| 2634 |
+
{
|
| 2635 |
+
"epoch": 4.606265876375953,
|
| 2636 |
+
"grad_norm": 0.22570809721946716,
|
| 2637 |
+
"learning_rate": 1.8566030912062549e-06,
|
| 2638 |
+
"loss": 0.0058,
|
| 2639 |
+
"step": 3060
|
| 2640 |
+
},
|
| 2641 |
+
{
|
| 2642 |
+
"epoch": 4.621319032834697,
|
| 2643 |
+
"grad_norm": 0.134377583861351,
|
| 2644 |
+
"learning_rate": 1.717341535715733e-06,
|
| 2645 |
+
"loss": 0.0034,
|
| 2646 |
+
"step": 3070
|
| 2647 |
+
},
|
| 2648 |
+
{
|
| 2649 |
+
"epoch": 4.636372189293443,
|
| 2650 |
+
"grad_norm": 0.11749345809221268,
|
| 2651 |
+
"learning_rate": 1.5834173308397982e-06,
|
| 2652 |
+
"loss": 0.0045,
|
| 2653 |
+
"step": 3080
|
| 2654 |
+
},
|
| 2655 |
+
{
|
| 2656 |
+
"epoch": 4.6514253457521875,
|
| 2657 |
+
"grad_norm": 0.0782150849699974,
|
| 2658 |
+
"learning_rate": 1.4548452810747403e-06,
|
| 2659 |
+
"loss": 0.0064,
|
| 2660 |
+
"step": 3090
|
| 2661 |
+
},
|
| 2662 |
+
{
|
| 2663 |
+
"epoch": 4.666478502210932,
|
| 2664 |
+
"grad_norm": 0.13278253376483917,
|
| 2665 |
+
"learning_rate": 1.33163959926903e-06,
|
| 2666 |
+
"loss": 0.0041,
|
| 2667 |
+
"step": 3100
|
| 2668 |
+
},
|
| 2669 |
+
{
|
| 2670 |
+
"epoch": 4.666478502210932,
|
| 2671 |
+
"eval_loss": 0.041555680334568024,
|
| 2672 |
+
"eval_runtime": 180.8052,
|
| 2673 |
+
"eval_samples_per_second": 6.532,
|
| 2674 |
+
"eval_steps_per_second": 6.532,
|
| 2675 |
+
"step": 3100
|
| 2676 |
+
},
|
| 2677 |
+
{
|
| 2678 |
+
"epoch": 4.681531658669678,
|
| 2679 |
+
"grad_norm": 0.06097458675503731,
|
| 2680 |
+
"learning_rate": 1.2138139050522023e-06,
|
| 2681 |
+
"loss": 0.0046,
|
| 2682 |
+
"step": 3110
|
| 2683 |
+
},
|
| 2684 |
+
{
|
| 2685 |
+
"epoch": 4.696584815128422,
|
| 2686 |
+
"grad_norm": 0.0928138941526413,
|
| 2687 |
+
"learning_rate": 1.101381223329301e-06,
|
| 2688 |
+
"loss": 0.0042,
|
| 2689 |
+
"step": 3120
|
| 2690 |
+
},
|
| 2691 |
+
{
|
| 2692 |
+
"epoch": 4.711637971587168,
|
| 2693 |
+
"grad_norm": 0.11098574846982956,
|
| 2694 |
+
"learning_rate": 9.943539828410342e-07,
|
| 2695 |
+
"loss": 0.0041,
|
| 2696 |
+
"step": 3130
|
| 2697 |
+
},
|
| 2698 |
+
{
|
| 2699 |
+
"epoch": 4.726691128045912,
|
| 2700 |
+
"grad_norm": 0.02670971117913723,
|
| 2701 |
+
"learning_rate": 8.927440147898702e-07,
|
| 2702 |
+
"loss": 0.0053,
|
| 2703 |
+
"step": 3140
|
| 2704 |
+
},
|
| 2705 |
+
{
|
| 2706 |
+
"epoch": 4.741744284504657,
|
| 2707 |
+
"grad_norm": 0.16498462855815887,
|
| 2708 |
+
"learning_rate": 7.96562551532154e-07,
|
| 2709 |
+
"loss": 0.0046,
|
| 2710 |
+
"step": 3150
|
| 2711 |
+
},
|
| 2712 |
+
{
|
| 2713 |
+
"epoch": 4.741744284504657,
|
| 2714 |
+
"eval_loss": 0.041612498462200165,
|
| 2715 |
+
"eval_runtime": 180.8943,
|
| 2716 |
+
"eval_samples_per_second": 6.529,
|
| 2717 |
+
"eval_steps_per_second": 6.529,
|
| 2718 |
+
"step": 3150
|
| 2719 |
+
},
|
| 2720 |
+
{
|
| 2721 |
+
"epoch": 4.756797440963402,
|
| 2722 |
+
"grad_norm": 0.02326858416199684,
|
| 2723 |
+
"learning_rate": 7.05820225336451e-07,
|
| 2724 |
+
"loss": 0.0034,
|
| 2725 |
+
"step": 3160
|
| 2726 |
+
},
|
| 2727 |
+
{
|
| 2728 |
+
"epoch": 4.771850597422147,
|
| 2729 |
+
"grad_norm": 0.03126712515950203,
|
| 2730 |
+
"learning_rate": 6.20527067208232e-07,
|
| 2731 |
+
"loss": 0.0028,
|
| 2732 |
+
"step": 3170
|
| 2733 |
+
},
|
| 2734 |
+
{
|
| 2735 |
+
"epoch": 4.786903753880892,
|
| 2736 |
+
"grad_norm": 0.09830054640769958,
|
| 2737 |
+
"learning_rate": 5.406925057809653e-07,
|
| 2738 |
+
"loss": 0.0042,
|
| 2739 |
+
"step": 3180
|
| 2740 |
+
},
|
| 2741 |
+
{
|
| 2742 |
+
"epoch": 4.801956910339637,
|
| 2743 |
+
"grad_norm": 0.10579635947942734,
|
| 2744 |
+
"learning_rate": 4.6632536627386756e-07,
|
| 2745 |
+
"loss": 0.0054,
|
| 2746 |
+
"step": 3190
|
| 2747 |
+
},
|
| 2748 |
+
{
|
| 2749 |
+
"epoch": 4.817010066798382,
|
| 2750 |
+
"grad_norm": 0.15730389952659607,
|
| 2751 |
+
"learning_rate": 3.974338695163393e-07,
|
| 2752 |
+
"loss": 0.0029,
|
| 2753 |
+
"step": 3200
|
| 2754 |
+
},
|
| 2755 |
+
{
|
| 2756 |
+
"epoch": 4.817010066798382,
|
| 2757 |
+
"eval_loss": 0.04158816114068031,
|
| 2758 |
+
"eval_runtime": 180.8757,
|
| 2759 |
+
"eval_samples_per_second": 6.529,
|
| 2760 |
+
"eval_steps_per_second": 6.529,
|
| 2761 |
+
"step": 3200
|
| 2762 |
+
},
|
| 2763 |
+
{
|
| 2764 |
+
"epoch": 4.832063223257126,
|
| 2765 |
+
"grad_norm": 0.15249422192573547,
|
| 2766 |
+
"learning_rate": 3.3402563103916984e-07,
|
| 2767 |
+
"loss": 0.003,
|
| 2768 |
+
"step": 3210
|
| 2769 |
+
},
|
| 2770 |
+
{
|
| 2771 |
+
"epoch": 4.847116379715872,
|
| 2772 |
+
"grad_norm": 0.06642192602157593,
|
| 2773 |
+
"learning_rate": 2.7610766023271615e-07,
|
| 2774 |
+
"loss": 0.0032,
|
| 2775 |
+
"step": 3220
|
| 2776 |
+
},
|
| 2777 |
+
{
|
| 2778 |
+
"epoch": 4.8621695361746164,
|
| 2779 |
+
"grad_norm": 0.17876122891902924,
|
| 2780 |
+
"learning_rate": 2.2368635957205618e-07,
|
| 2781 |
+
"loss": 0.0049,
|
| 2782 |
+
"step": 3230
|
| 2783 |
+
},
|
| 2784 |
+
{
|
| 2785 |
+
"epoch": 4.877222692633362,
|
| 2786 |
+
"grad_norm": 0.2662655711174011,
|
| 2787 |
+
"learning_rate": 1.7676752390920482e-07,
|
| 2788 |
+
"loss": 0.0067,
|
| 2789 |
+
"step": 3240
|
| 2790 |
+
},
|
| 2791 |
+
{
|
| 2792 |
+
"epoch": 4.8922758490921066,
|
| 2793 |
+
"grad_norm": 0.3073679208755493,
|
| 2794 |
+
"learning_rate": 1.3535633983257078e-07,
|
| 2795 |
+
"loss": 0.0071,
|
| 2796 |
+
"step": 3250
|
| 2797 |
+
},
|
| 2798 |
+
{
|
| 2799 |
+
"epoch": 4.8922758490921066,
|
| 2800 |
+
"eval_loss": 0.04167798534035683,
|
| 2801 |
+
"eval_runtime": 180.9807,
|
| 2802 |
+
"eval_samples_per_second": 6.526,
|
| 2803 |
+
"eval_steps_per_second": 6.526,
|
| 2804 |
+
"step": 3250
|
| 2805 |
+
},
|
| 2806 |
+
{
|
| 2807 |
+
"epoch": 4.907329005550851,
|
| 2808 |
+
"grad_norm": 0.07359470427036285,
|
| 2809 |
+
"learning_rate": 9.945738509358205e-08,
|
| 2810 |
+
"loss": 0.0026,
|
| 2811 |
+
"step": 3260
|
| 2812 |
+
},
|
| 2813 |
+
{
|
| 2814 |
+
"epoch": 4.922382162009597,
|
| 2815 |
+
"grad_norm": 0.0783749371767044,
|
| 2816 |
+
"learning_rate": 6.907462810065158e-08,
|
| 2817 |
+
"loss": 0.0037,
|
| 2818 |
+
"step": 3270
|
| 2819 |
+
},
|
| 2820 |
+
{
|
| 2821 |
+
"epoch": 4.937435318468341,
|
| 2822 |
+
"grad_norm": 0.2075500339269638,
|
| 2823 |
+
"learning_rate": 4.4211427480500554e-08,
|
| 2824 |
+
"loss": 0.0079,
|
| 2825 |
+
"step": 3280
|
| 2826 |
+
},
|
| 2827 |
+
{
|
| 2828 |
+
"epoch": 4.952488474927087,
|
| 2829 |
+
"grad_norm": 0.035787180066108704,
|
| 2830 |
+
"learning_rate": 2.4870531706872034e-08,
|
| 2831 |
+
"loss": 0.0042,
|
| 2832 |
+
"step": 3290
|
| 2833 |
+
},
|
| 2834 |
+
{
|
| 2835 |
+
"epoch": 4.967541631385831,
|
| 2836 |
+
"grad_norm": 0.28970831632614136,
|
| 2837 |
+
"learning_rate": 1.105407879670728e-08,
|
| 2838 |
+
"loss": 0.0039,
|
| 2839 |
+
"step": 3300
|
| 2840 |
+
},
|
| 2841 |
+
{
|
| 2842 |
+
"epoch": 4.967541631385831,
|
| 2843 |
+
"eval_loss": 0.041562460362911224,
|
| 2844 |
+
"eval_runtime": 180.9472,
|
| 2845 |
+
"eval_samples_per_second": 6.527,
|
| 2846 |
+
"eval_steps_per_second": 6.527,
|
| 2847 |
+
"step": 3300
|
| 2848 |
+
},
|
| 2849 |
+
{
|
| 2850 |
+
"epoch": 4.982594787844576,
|
| 2851 |
+
"grad_norm": 0.1511860340833664,
|
| 2852 |
+
"learning_rate": 2.763596073807051e-09,
|
| 2853 |
+
"loss": 0.0041,
|
| 2854 |
+
"step": 3310
|
| 2855 |
+
},
|
| 2856 |
+
{
|
| 2857 |
+
"epoch": 4.9976479443033215,
|
| 2858 |
+
"grad_norm": 0.369547039270401,
|
| 2859 |
+
"learning_rate": 0.0,
|
| 2860 |
+
"loss": 0.0045,
|
| 2861 |
+
"step": 3320
|
| 2862 |
+
},
|
| 2863 |
+
{
|
| 2864 |
+
"epoch": 4.9976479443033215,
|
| 2865 |
+
"step": 3320,
|
| 2866 |
+
"total_flos": 8.981559806648648e+17,
|
| 2867 |
+
"train_loss": 0.048681382634620886,
|
| 2868 |
+
"train_runtime": 37834.6736,
|
| 2869 |
+
"train_samples_per_second": 1.405,
|
| 2870 |
+
"train_steps_per_second": 0.088
|
| 2871 |
+
}
|
| 2872 |
+
],
|
| 2873 |
+
"logging_steps": 10,
|
| 2874 |
+
"max_steps": 3320,
|
| 2875 |
+
"num_input_tokens_seen": 0,
|
| 2876 |
+
"num_train_epochs": 5,
|
| 2877 |
+
"save_steps": 50,
|
| 2878 |
+
"stateful_callbacks": {
|
| 2879 |
+
"TrainerControl": {
|
| 2880 |
+
"args": {
|
| 2881 |
+
"should_epoch_stop": false,
|
| 2882 |
+
"should_evaluate": false,
|
| 2883 |
+
"should_log": false,
|
| 2884 |
+
"should_save": true,
|
| 2885 |
+
"should_training_stop": true
|
| 2886 |
+
},
|
| 2887 |
+
"attributes": {}
|
| 2888 |
+
}
|
| 2889 |
+
},
|
| 2890 |
+
"total_flos": 8.981559806648648e+17,
|
| 2891 |
+
"train_batch_size": 1,
|
| 2892 |
+
"trial_name": null,
|
| 2893 |
+
"trial_params": null
|
| 2894 |
+
}
|
training_eval_loss.png
ADDED
|
training_loss.png
ADDED
|