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 +2837 -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-fold9
|
|
@@ -15,9 +16,9 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 15 |
|
| 16 |
# MentaLLaMA-chat-7B-PsyCourse-fold9
|
| 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-fold9
|
|
|
|
| 16 |
|
| 17 |
# MentaLLaMA-chat-7B-PsyCourse-fold9
|
| 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-fold9 dataset.
|
| 20 |
It achieves the following results on the evaluation set:
|
| 21 |
+
- Loss: 0.0305
|
| 22 |
|
| 23 |
## Model description
|
| 24 |
|
all_results.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 4.997121473805412,
|
| 3 |
+
"eval_loss": 0.030531086027622223,
|
| 4 |
+
"eval_runtime": 195.6611,
|
| 5 |
+
"eval_samples_per_second": 5.924,
|
| 6 |
+
"eval_steps_per_second": 5.924,
|
| 7 |
+
"total_flos": 8.80470547326763e+17,
|
| 8 |
+
"train_loss": 0.04897872828713943,
|
| 9 |
+
"train_runtime": 41700.3864,
|
| 10 |
+
"train_samples_per_second": 1.25,
|
| 11 |
+
"train_steps_per_second": 0.078
|
| 12 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 4.997121473805412,
|
| 3 |
+
"eval_loss": 0.030531086027622223,
|
| 4 |
+
"eval_runtime": 195.6611,
|
| 5 |
+
"eval_samples_per_second": 5.924,
|
| 6 |
+
"eval_steps_per_second": 5.924
|
| 7 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 4.997121473805412,
|
| 3 |
+
"total_flos": 8.80470547326763e+17,
|
| 4 |
+
"train_loss": 0.04897872828713943,
|
| 5 |
+
"train_runtime": 41700.3864,
|
| 6 |
+
"train_samples_per_second": 1.25,
|
| 7 |
+
"train_steps_per_second": 0.078
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,2837 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.030531086027622223,
|
| 3 |
+
"best_model_checkpoint": "saves/psy-course/MentaLLaMA-chat-7B/train/fold9/checkpoint-1450",
|
| 4 |
+
"epoch": 4.997121473805412,
|
| 5 |
+
"eval_steps": 50,
|
| 6 |
+
"global_step": 3255,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.01535213970447131,
|
| 13 |
+
"grad_norm": 1.986027717590332,
|
| 14 |
+
"learning_rate": 3.067484662576687e-06,
|
| 15 |
+
"loss": 1.7207,
|
| 16 |
+
"step": 10
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.03070427940894262,
|
| 20 |
+
"grad_norm": 1.8678655624389648,
|
| 21 |
+
"learning_rate": 6.134969325153374e-06,
|
| 22 |
+
"loss": 1.5868,
|
| 23 |
+
"step": 20
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.04605641911341393,
|
| 27 |
+
"grad_norm": 2.281944751739502,
|
| 28 |
+
"learning_rate": 9.202453987730062e-06,
|
| 29 |
+
"loss": 1.6864,
|
| 30 |
+
"step": 30
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.06140855881788524,
|
| 34 |
+
"grad_norm": 2.9772989749908447,
|
| 35 |
+
"learning_rate": 1.2269938650306748e-05,
|
| 36 |
+
"loss": 1.3561,
|
| 37 |
+
"step": 40
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.07676069852235655,
|
| 41 |
+
"grad_norm": 1.116693377494812,
|
| 42 |
+
"learning_rate": 1.5337423312883436e-05,
|
| 43 |
+
"loss": 0.8389,
|
| 44 |
+
"step": 50
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.07676069852235655,
|
| 48 |
+
"eval_loss": 0.6106540560722351,
|
| 49 |
+
"eval_runtime": 210.0434,
|
| 50 |
+
"eval_samples_per_second": 5.518,
|
| 51 |
+
"eval_steps_per_second": 5.518,
|
| 52 |
+
"step": 50
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.09211283822682786,
|
| 56 |
+
"grad_norm": 0.8406881093978882,
|
| 57 |
+
"learning_rate": 1.8404907975460123e-05,
|
| 58 |
+
"loss": 0.668,
|
| 59 |
+
"step": 60
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.10746497793129918,
|
| 63 |
+
"grad_norm": 0.87685626745224,
|
| 64 |
+
"learning_rate": 2.1472392638036813e-05,
|
| 65 |
+
"loss": 0.5046,
|
| 66 |
+
"step": 70
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.12281711763577048,
|
| 70 |
+
"grad_norm": 0.7217749953269958,
|
| 71 |
+
"learning_rate": 2.4539877300613496e-05,
|
| 72 |
+
"loss": 0.2741,
|
| 73 |
+
"step": 80
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.1381692573402418,
|
| 77 |
+
"grad_norm": 0.6710460782051086,
|
| 78 |
+
"learning_rate": 2.7607361963190186e-05,
|
| 79 |
+
"loss": 0.1939,
|
| 80 |
+
"step": 90
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.1535213970447131,
|
| 84 |
+
"grad_norm": 0.46170759201049805,
|
| 85 |
+
"learning_rate": 3.067484662576687e-05,
|
| 86 |
+
"loss": 0.1391,
|
| 87 |
+
"step": 100
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.1535213970447131,
|
| 91 |
+
"eval_loss": 0.10637813061475754,
|
| 92 |
+
"eval_runtime": 209.8019,
|
| 93 |
+
"eval_samples_per_second": 5.524,
|
| 94 |
+
"eval_steps_per_second": 5.524,
|
| 95 |
+
"step": 100
|
| 96 |
+
},
|
| 97 |
+
{
|
| 98 |
+
"epoch": 0.16887353674918443,
|
| 99 |
+
"grad_norm": 0.4916459023952484,
|
| 100 |
+
"learning_rate": 3.3742331288343556e-05,
|
| 101 |
+
"loss": 0.098,
|
| 102 |
+
"step": 110
|
| 103 |
+
},
|
| 104 |
+
{
|
| 105 |
+
"epoch": 0.18422567645365573,
|
| 106 |
+
"grad_norm": 0.3704639673233032,
|
| 107 |
+
"learning_rate": 3.6809815950920246e-05,
|
| 108 |
+
"loss": 0.089,
|
| 109 |
+
"step": 120
|
| 110 |
+
},
|
| 111 |
+
{
|
| 112 |
+
"epoch": 0.19957781615812703,
|
| 113 |
+
"grad_norm": 0.8016656041145325,
|
| 114 |
+
"learning_rate": 3.987730061349693e-05,
|
| 115 |
+
"loss": 0.0972,
|
| 116 |
+
"step": 130
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 0.21492995586259836,
|
| 120 |
+
"grad_norm": 0.8712407350540161,
|
| 121 |
+
"learning_rate": 4.2944785276073626e-05,
|
| 122 |
+
"loss": 0.0683,
|
| 123 |
+
"step": 140
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"epoch": 0.23028209556706966,
|
| 127 |
+
"grad_norm": 0.7326760292053223,
|
| 128 |
+
"learning_rate": 4.601226993865031e-05,
|
| 129 |
+
"loss": 0.0757,
|
| 130 |
+
"step": 150
|
| 131 |
+
},
|
| 132 |
+
{
|
| 133 |
+
"epoch": 0.23028209556706966,
|
| 134 |
+
"eval_loss": 0.06941775232553482,
|
| 135 |
+
"eval_runtime": 209.7144,
|
| 136 |
+
"eval_samples_per_second": 5.527,
|
| 137 |
+
"eval_steps_per_second": 5.527,
|
| 138 |
+
"step": 150
|
| 139 |
+
},
|
| 140 |
+
{
|
| 141 |
+
"epoch": 0.24563423527154096,
|
| 142 |
+
"grad_norm": 0.8024667501449585,
|
| 143 |
+
"learning_rate": 4.907975460122699e-05,
|
| 144 |
+
"loss": 0.0869,
|
| 145 |
+
"step": 160
|
| 146 |
+
},
|
| 147 |
+
{
|
| 148 |
+
"epoch": 0.2609863749760123,
|
| 149 |
+
"grad_norm": 0.7557644844055176,
|
| 150 |
+
"learning_rate": 5.214723926380368e-05,
|
| 151 |
+
"loss": 0.0752,
|
| 152 |
+
"step": 170
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 0.2763385146804836,
|
| 156 |
+
"grad_norm": 0.8181012868881226,
|
| 157 |
+
"learning_rate": 5.521472392638037e-05,
|
| 158 |
+
"loss": 0.0704,
|
| 159 |
+
"step": 180
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 0.2916906543849549,
|
| 163 |
+
"grad_norm": 0.8826432228088379,
|
| 164 |
+
"learning_rate": 5.8282208588957056e-05,
|
| 165 |
+
"loss": 0.0583,
|
| 166 |
+
"step": 190
|
| 167 |
+
},
|
| 168 |
+
{
|
| 169 |
+
"epoch": 0.3070427940894262,
|
| 170 |
+
"grad_norm": 0.5068766474723816,
|
| 171 |
+
"learning_rate": 6.134969325153375e-05,
|
| 172 |
+
"loss": 0.0639,
|
| 173 |
+
"step": 200
|
| 174 |
+
},
|
| 175 |
+
{
|
| 176 |
+
"epoch": 0.3070427940894262,
|
| 177 |
+
"eval_loss": 0.058139290660619736,
|
| 178 |
+
"eval_runtime": 209.743,
|
| 179 |
+
"eval_samples_per_second": 5.526,
|
| 180 |
+
"eval_steps_per_second": 5.526,
|
| 181 |
+
"step": 200
|
| 182 |
+
},
|
| 183 |
+
{
|
| 184 |
+
"epoch": 0.3223949337938975,
|
| 185 |
+
"grad_norm": 0.660163402557373,
|
| 186 |
+
"learning_rate": 6.441717791411042e-05,
|
| 187 |
+
"loss": 0.0634,
|
| 188 |
+
"step": 210
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 0.33774707349836885,
|
| 192 |
+
"grad_norm": 0.4490329325199127,
|
| 193 |
+
"learning_rate": 6.748466257668711e-05,
|
| 194 |
+
"loss": 0.0456,
|
| 195 |
+
"step": 220
|
| 196 |
+
},
|
| 197 |
+
{
|
| 198 |
+
"epoch": 0.35309921320284016,
|
| 199 |
+
"grad_norm": 0.5326518416404724,
|
| 200 |
+
"learning_rate": 7.055214723926382e-05,
|
| 201 |
+
"loss": 0.0758,
|
| 202 |
+
"step": 230
|
| 203 |
+
},
|
| 204 |
+
{
|
| 205 |
+
"epoch": 0.36845135290731146,
|
| 206 |
+
"grad_norm": 0.527441143989563,
|
| 207 |
+
"learning_rate": 7.361963190184049e-05,
|
| 208 |
+
"loss": 0.0547,
|
| 209 |
+
"step": 240
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 0.38380349261178276,
|
| 213 |
+
"grad_norm": 0.6042910814285278,
|
| 214 |
+
"learning_rate": 7.668711656441718e-05,
|
| 215 |
+
"loss": 0.0739,
|
| 216 |
+
"step": 250
|
| 217 |
+
},
|
| 218 |
+
{
|
| 219 |
+
"epoch": 0.38380349261178276,
|
| 220 |
+
"eval_loss": 0.0467386394739151,
|
| 221 |
+
"eval_runtime": 209.5833,
|
| 222 |
+
"eval_samples_per_second": 5.53,
|
| 223 |
+
"eval_steps_per_second": 5.53,
|
| 224 |
+
"step": 250
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 0.39915563231625406,
|
| 228 |
+
"grad_norm": 0.6599677205085754,
|
| 229 |
+
"learning_rate": 7.975460122699386e-05,
|
| 230 |
+
"loss": 0.0542,
|
| 231 |
+
"step": 260
|
| 232 |
+
},
|
| 233 |
+
{
|
| 234 |
+
"epoch": 0.41450777202072536,
|
| 235 |
+
"grad_norm": 0.4332006275653839,
|
| 236 |
+
"learning_rate": 8.282208588957055e-05,
|
| 237 |
+
"loss": 0.0519,
|
| 238 |
+
"step": 270
|
| 239 |
+
},
|
| 240 |
+
{
|
| 241 |
+
"epoch": 0.4298599117251967,
|
| 242 |
+
"grad_norm": 0.7430977821350098,
|
| 243 |
+
"learning_rate": 8.588957055214725e-05,
|
| 244 |
+
"loss": 0.0575,
|
| 245 |
+
"step": 280
|
| 246 |
+
},
|
| 247 |
+
{
|
| 248 |
+
"epoch": 0.445212051429668,
|
| 249 |
+
"grad_norm": 0.4669485092163086,
|
| 250 |
+
"learning_rate": 8.895705521472393e-05,
|
| 251 |
+
"loss": 0.0456,
|
| 252 |
+
"step": 290
|
| 253 |
+
},
|
| 254 |
+
{
|
| 255 |
+
"epoch": 0.4605641911341393,
|
| 256 |
+
"grad_norm": 0.5974306464195251,
|
| 257 |
+
"learning_rate": 9.202453987730062e-05,
|
| 258 |
+
"loss": 0.0558,
|
| 259 |
+
"step": 300
|
| 260 |
+
},
|
| 261 |
+
{
|
| 262 |
+
"epoch": 0.4605641911341393,
|
| 263 |
+
"eval_loss": 0.04392043128609657,
|
| 264 |
+
"eval_runtime": 209.0171,
|
| 265 |
+
"eval_samples_per_second": 5.545,
|
| 266 |
+
"eval_steps_per_second": 5.545,
|
| 267 |
+
"step": 300
|
| 268 |
+
},
|
| 269 |
+
{
|
| 270 |
+
"epoch": 0.4759163308386106,
|
| 271 |
+
"grad_norm": 0.5303856134414673,
|
| 272 |
+
"learning_rate": 9.50920245398773e-05,
|
| 273 |
+
"loss": 0.0436,
|
| 274 |
+
"step": 310
|
| 275 |
+
},
|
| 276 |
+
{
|
| 277 |
+
"epoch": 0.4912684705430819,
|
| 278 |
+
"grad_norm": 0.30366232991218567,
|
| 279 |
+
"learning_rate": 9.815950920245399e-05,
|
| 280 |
+
"loss": 0.0447,
|
| 281 |
+
"step": 320
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 0.5066206102475532,
|
| 285 |
+
"grad_norm": 0.31800657510757446,
|
| 286 |
+
"learning_rate": 9.999953982785432e-05,
|
| 287 |
+
"loss": 0.0523,
|
| 288 |
+
"step": 330
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"epoch": 0.5219727499520246,
|
| 292 |
+
"grad_norm": 0.38302093744277954,
|
| 293 |
+
"learning_rate": 9.999436298849151e-05,
|
| 294 |
+
"loss": 0.0413,
|
| 295 |
+
"step": 340
|
| 296 |
+
},
|
| 297 |
+
{
|
| 298 |
+
"epoch": 0.5373248896564958,
|
| 299 |
+
"grad_norm": 0.25141751766204834,
|
| 300 |
+
"learning_rate": 9.998343469212352e-05,
|
| 301 |
+
"loss": 0.0394,
|
| 302 |
+
"step": 350
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 0.5373248896564958,
|
| 306 |
+
"eval_loss": 0.04250750318169594,
|
| 307 |
+
"eval_runtime": 209.0742,
|
| 308 |
+
"eval_samples_per_second": 5.543,
|
| 309 |
+
"eval_steps_per_second": 5.543,
|
| 310 |
+
"step": 350
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.5526770293609672,
|
| 314 |
+
"grad_norm": 0.5167840719223022,
|
| 315 |
+
"learning_rate": 9.996675619596465e-05,
|
| 316 |
+
"loss": 0.0382,
|
| 317 |
+
"step": 360
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.5680291690654385,
|
| 321 |
+
"grad_norm": 0.4687771201133728,
|
| 322 |
+
"learning_rate": 9.99443294187443e-05,
|
| 323 |
+
"loss": 0.053,
|
| 324 |
+
"step": 370
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.5833813087699098,
|
| 328 |
+
"grad_norm": 0.16902729868888855,
|
| 329 |
+
"learning_rate": 9.991615694048621e-05,
|
| 330 |
+
"loss": 0.039,
|
| 331 |
+
"step": 380
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.5987334484743811,
|
| 335 |
+
"grad_norm": 0.48972830176353455,
|
| 336 |
+
"learning_rate": 9.988224200221172e-05,
|
| 337 |
+
"loss": 0.0537,
|
| 338 |
+
"step": 390
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.6140855881788524,
|
| 342 |
+
"grad_norm": 0.2853284478187561,
|
| 343 |
+
"learning_rate": 9.984258850556693e-05,
|
| 344 |
+
"loss": 0.0477,
|
| 345 |
+
"step": 400
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.6140855881788524,
|
| 349 |
+
"eval_loss": 0.04190515726804733,
|
| 350 |
+
"eval_runtime": 209.6671,
|
| 351 |
+
"eval_samples_per_second": 5.528,
|
| 352 |
+
"eval_steps_per_second": 5.528,
|
| 353 |
+
"step": 400
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.6294377278833237,
|
| 357 |
+
"grad_norm": 0.560493528842926,
|
| 358 |
+
"learning_rate": 9.979720101237375e-05,
|
| 359 |
+
"loss": 0.0418,
|
| 360 |
+
"step": 410
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 0.644789867587795,
|
| 364 |
+
"grad_norm": 0.36185890436172485,
|
| 365 |
+
"learning_rate": 9.974608474410512e-05,
|
| 366 |
+
"loss": 0.04,
|
| 367 |
+
"step": 420
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 0.6601420072922664,
|
| 371 |
+
"grad_norm": 0.6417878270149231,
|
| 372 |
+
"learning_rate": 9.968924558128445e-05,
|
| 373 |
+
"loss": 0.0423,
|
| 374 |
+
"step": 430
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.6754941469967377,
|
| 378 |
+
"grad_norm": 0.4333862066268921,
|
| 379 |
+
"learning_rate": 9.962669006280894e-05,
|
| 380 |
+
"loss": 0.0384,
|
| 381 |
+
"step": 440
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 0.690846286701209,
|
| 385 |
+
"grad_norm": 0.3614458739757538,
|
| 386 |
+
"learning_rate": 9.95584253851974e-05,
|
| 387 |
+
"loss": 0.047,
|
| 388 |
+
"step": 450
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 0.690846286701209,
|
| 392 |
+
"eval_loss": 0.043166350573301315,
|
| 393 |
+
"eval_runtime": 209.6238,
|
| 394 |
+
"eval_samples_per_second": 5.529,
|
| 395 |
+
"eval_steps_per_second": 5.529,
|
| 396 |
+
"step": 450
|
| 397 |
+
},
|
| 398 |
+
{
|
| 399 |
+
"epoch": 0.7061984264056803,
|
| 400 |
+
"grad_norm": 0.16041065752506256,
|
| 401 |
+
"learning_rate": 9.948445940176243e-05,
|
| 402 |
+
"loss": 0.0444,
|
| 403 |
+
"step": 460
|
| 404 |
+
},
|
| 405 |
+
{
|
| 406 |
+
"epoch": 0.7215505661101516,
|
| 407 |
+
"grad_norm": 0.2849539518356323,
|
| 408 |
+
"learning_rate": 9.940480062170679e-05,
|
| 409 |
+
"loss": 0.0407,
|
| 410 |
+
"step": 470
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 0.7369027058146229,
|
| 414 |
+
"grad_norm": 0.2027292102575302,
|
| 415 |
+
"learning_rate": 9.931945820914462e-05,
|
| 416 |
+
"loss": 0.0537,
|
| 417 |
+
"step": 480
|
| 418 |
+
},
|
| 419 |
+
{
|
| 420 |
+
"epoch": 0.7522548455190943,
|
| 421 |
+
"grad_norm": 0.3879440426826477,
|
| 422 |
+
"learning_rate": 9.922844198204715e-05,
|
| 423 |
+
"loss": 0.037,
|
| 424 |
+
"step": 490
|
| 425 |
+
},
|
| 426 |
+
{
|
| 427 |
+
"epoch": 0.7676069852235655,
|
| 428 |
+
"grad_norm": 0.3453935980796814,
|
| 429 |
+
"learning_rate": 9.913176241111319e-05,
|
| 430 |
+
"loss": 0.0416,
|
| 431 |
+
"step": 500
|
| 432 |
+
},
|
| 433 |
+
{
|
| 434 |
+
"epoch": 0.7676069852235655,
|
| 435 |
+
"eval_loss": 0.03759530559182167,
|
| 436 |
+
"eval_runtime": 206.4796,
|
| 437 |
+
"eval_samples_per_second": 5.613,
|
| 438 |
+
"eval_steps_per_second": 5.613,
|
| 439 |
+
"step": 500
|
| 440 |
+
},
|
| 441 |
+
{
|
| 442 |
+
"epoch": 0.7829591249280369,
|
| 443 |
+
"grad_norm": 0.3907797932624817,
|
| 444 |
+
"learning_rate": 9.902943061856456e-05,
|
| 445 |
+
"loss": 0.0356,
|
| 446 |
+
"step": 510
|
| 447 |
+
},
|
| 448 |
+
{
|
| 449 |
+
"epoch": 0.7983112646325081,
|
| 450 |
+
"grad_norm": 0.27914416790008545,
|
| 451 |
+
"learning_rate": 9.892145837686657e-05,
|
| 452 |
+
"loss": 0.052,
|
| 453 |
+
"step": 520
|
| 454 |
+
},
|
| 455 |
+
{
|
| 456 |
+
"epoch": 0.8136634043369795,
|
| 457 |
+
"grad_norm": 0.2834685742855072,
|
| 458 |
+
"learning_rate": 9.880785810737378e-05,
|
| 459 |
+
"loss": 0.048,
|
| 460 |
+
"step": 530
|
| 461 |
+
},
|
| 462 |
+
{
|
| 463 |
+
"epoch": 0.8290155440414507,
|
| 464 |
+
"grad_norm": 0.11335510015487671,
|
| 465 |
+
"learning_rate": 9.868864287890083e-05,
|
| 466 |
+
"loss": 0.037,
|
| 467 |
+
"step": 540
|
| 468 |
+
},
|
| 469 |
+
{
|
| 470 |
+
"epoch": 0.8443676837459221,
|
| 471 |
+
"grad_norm": 0.26555532217025757,
|
| 472 |
+
"learning_rate": 9.856382640621917e-05,
|
| 473 |
+
"loss": 0.0637,
|
| 474 |
+
"step": 550
|
| 475 |
+
},
|
| 476 |
+
{
|
| 477 |
+
"epoch": 0.8443676837459221,
|
| 478 |
+
"eval_loss": 0.03945336863398552,
|
| 479 |
+
"eval_runtime": 202.7088,
|
| 480 |
+
"eval_samples_per_second": 5.718,
|
| 481 |
+
"eval_steps_per_second": 5.718,
|
| 482 |
+
"step": 550
|
| 483 |
+
},
|
| 484 |
+
{
|
| 485 |
+
"epoch": 0.8597198234503934,
|
| 486 |
+
"grad_norm": 0.3383285701274872,
|
| 487 |
+
"learning_rate": 9.84334230484792e-05,
|
| 488 |
+
"loss": 0.0344,
|
| 489 |
+
"step": 560
|
| 490 |
+
},
|
| 491 |
+
{
|
| 492 |
+
"epoch": 0.8750719631548647,
|
| 493 |
+
"grad_norm": 0.11692766845226288,
|
| 494 |
+
"learning_rate": 9.82974478075583e-05,
|
| 495 |
+
"loss": 0.0392,
|
| 496 |
+
"step": 570
|
| 497 |
+
},
|
| 498 |
+
{
|
| 499 |
+
"epoch": 0.890424102859336,
|
| 500 |
+
"grad_norm": 0.3442629873752594,
|
| 501 |
+
"learning_rate": 9.815591632633509e-05,
|
| 502 |
+
"loss": 0.0343,
|
| 503 |
+
"step": 580
|
| 504 |
+
},
|
| 505 |
+
{
|
| 506 |
+
"epoch": 0.9057762425638073,
|
| 507 |
+
"grad_norm": 0.25957638025283813,
|
| 508 |
+
"learning_rate": 9.800884488688985e-05,
|
| 509 |
+
"loss": 0.0465,
|
| 510 |
+
"step": 590
|
| 511 |
+
},
|
| 512 |
+
{
|
| 513 |
+
"epoch": 0.9211283822682786,
|
| 514 |
+
"grad_norm": 0.20577019453048706,
|
| 515 |
+
"learning_rate": 9.785625040863124e-05,
|
| 516 |
+
"loss": 0.0377,
|
| 517 |
+
"step": 600
|
| 518 |
+
},
|
| 519 |
+
{
|
| 520 |
+
"epoch": 0.9211283822682786,
|
| 521 |
+
"eval_loss": 0.035671208053827286,
|
| 522 |
+
"eval_runtime": 200.7483,
|
| 523 |
+
"eval_samples_per_second": 5.773,
|
| 524 |
+
"eval_steps_per_second": 5.773,
|
| 525 |
+
"step": 600
|
| 526 |
+
},
|
| 527 |
+
{
|
| 528 |
+
"epoch": 0.93648052197275,
|
| 529 |
+
"grad_norm": 0.24516840279102325,
|
| 530 |
+
"learning_rate": 9.769815044635005e-05,
|
| 531 |
+
"loss": 0.0289,
|
| 532 |
+
"step": 610
|
| 533 |
+
},
|
| 534 |
+
{
|
| 535 |
+
"epoch": 0.9518326616772212,
|
| 536 |
+
"grad_norm": 0.3754214942455292,
|
| 537 |
+
"learning_rate": 9.753456318819946e-05,
|
| 538 |
+
"loss": 0.042,
|
| 539 |
+
"step": 620
|
| 540 |
+
},
|
| 541 |
+
{
|
| 542 |
+
"epoch": 0.9671848013816926,
|
| 543 |
+
"grad_norm": 0.35276058316230774,
|
| 544 |
+
"learning_rate": 9.736550745360292e-05,
|
| 545 |
+
"loss": 0.037,
|
| 546 |
+
"step": 630
|
| 547 |
+
},
|
| 548 |
+
{
|
| 549 |
+
"epoch": 0.9825369410861639,
|
| 550 |
+
"grad_norm": 0.14712929725646973,
|
| 551 |
+
"learning_rate": 9.719100269108872e-05,
|
| 552 |
+
"loss": 0.0341,
|
| 553 |
+
"step": 640
|
| 554 |
+
},
|
| 555 |
+
{
|
| 556 |
+
"epoch": 0.9978890807906352,
|
| 557 |
+
"grad_norm": 0.16145889461040497,
|
| 558 |
+
"learning_rate": 9.701106897605304e-05,
|
| 559 |
+
"loss": 0.0315,
|
| 560 |
+
"step": 650
|
| 561 |
+
},
|
| 562 |
+
{
|
| 563 |
+
"epoch": 0.9978890807906352,
|
| 564 |
+
"eval_loss": 0.036111604422330856,
|
| 565 |
+
"eval_runtime": 199.7828,
|
| 566 |
+
"eval_samples_per_second": 5.801,
|
| 567 |
+
"eval_steps_per_second": 5.801,
|
| 568 |
+
"step": 650
|
| 569 |
+
},
|
| 570 |
+
{
|
| 571 |
+
"epoch": 1.0132412204951065,
|
| 572 |
+
"grad_norm": 0.21693025529384613,
|
| 573 |
+
"learning_rate": 9.682572700845006e-05,
|
| 574 |
+
"loss": 0.0254,
|
| 575 |
+
"step": 660
|
| 576 |
+
},
|
| 577 |
+
{
|
| 578 |
+
"epoch": 1.0285933601995778,
|
| 579 |
+
"grad_norm": 0.2023308128118515,
|
| 580 |
+
"learning_rate": 9.663499811041082e-05,
|
| 581 |
+
"loss": 0.0339,
|
| 582 |
+
"step": 670
|
| 583 |
+
},
|
| 584 |
+
{
|
| 585 |
+
"epoch": 1.0439454999040492,
|
| 586 |
+
"grad_norm": 0.3511405885219574,
|
| 587 |
+
"learning_rate": 9.643890422379018e-05,
|
| 588 |
+
"loss": 0.0334,
|
| 589 |
+
"step": 680
|
| 590 |
+
},
|
| 591 |
+
{
|
| 592 |
+
"epoch": 1.0592976396085205,
|
| 593 |
+
"grad_norm": 0.23378199338912964,
|
| 594 |
+
"learning_rate": 9.623746790764261e-05,
|
| 595 |
+
"loss": 0.0312,
|
| 596 |
+
"step": 690
|
| 597 |
+
},
|
| 598 |
+
{
|
| 599 |
+
"epoch": 1.0746497793129917,
|
| 600 |
+
"grad_norm": 0.3404940664768219,
|
| 601 |
+
"learning_rate": 9.603071233562695e-05,
|
| 602 |
+
"loss": 0.0332,
|
| 603 |
+
"step": 700
|
| 604 |
+
},
|
| 605 |
+
{
|
| 606 |
+
"epoch": 1.0746497793129917,
|
| 607 |
+
"eval_loss": 0.0359707772731781,
|
| 608 |
+
"eval_runtime": 199.3622,
|
| 609 |
+
"eval_samples_per_second": 5.814,
|
| 610 |
+
"eval_steps_per_second": 5.814,
|
| 611 |
+
"step": 700
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 1.090001919017463,
|
| 615 |
+
"grad_norm": 0.10679006576538086,
|
| 616 |
+
"learning_rate": 9.581866129334044e-05,
|
| 617 |
+
"loss": 0.0407,
|
| 618 |
+
"step": 710
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 1.1053540587219344,
|
| 622 |
+
"grad_norm": 0.1712823063135147,
|
| 623 |
+
"learning_rate": 9.560133917558242e-05,
|
| 624 |
+
"loss": 0.0431,
|
| 625 |
+
"step": 720
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 1.1207061984264057,
|
| 629 |
+
"grad_norm": 0.3909640610218048,
|
| 630 |
+
"learning_rate": 9.537877098354786e-05,
|
| 631 |
+
"loss": 0.033,
|
| 632 |
+
"step": 730
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 1.136058338130877,
|
| 636 |
+
"grad_norm": 0.18858745694160461,
|
| 637 |
+
"learning_rate": 9.51509823219512e-05,
|
| 638 |
+
"loss": 0.0278,
|
| 639 |
+
"step": 740
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 1.1514104778353482,
|
| 643 |
+
"grad_norm": 0.3083574175834656,
|
| 644 |
+
"learning_rate": 9.491799939608065e-05,
|
| 645 |
+
"loss": 0.0322,
|
| 646 |
+
"step": 750
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 1.1514104778353482,
|
| 650 |
+
"eval_loss": 0.035209909081459045,
|
| 651 |
+
"eval_runtime": 199.0974,
|
| 652 |
+
"eval_samples_per_second": 5.821,
|
| 653 |
+
"eval_steps_per_second": 5.821,
|
| 654 |
+
"step": 750
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 1.1667626175398196,
|
| 658 |
+
"grad_norm": 0.30565544962882996,
|
| 659 |
+
"learning_rate": 9.467984900878364e-05,
|
| 660 |
+
"loss": 0.0321,
|
| 661 |
+
"step": 760
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 1.182114757244291,
|
| 665 |
+
"grad_norm": 0.18242968618869781,
|
| 666 |
+
"learning_rate": 9.443655855738321e-05,
|
| 667 |
+
"loss": 0.0295,
|
| 668 |
+
"step": 770
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 1.1974668969487623,
|
| 672 |
+
"grad_norm": 0.269875168800354,
|
| 673 |
+
"learning_rate": 9.41881560305262e-05,
|
| 674 |
+
"loss": 0.0279,
|
| 675 |
+
"step": 780
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 1.2128190366532334,
|
| 679 |
+
"grad_norm": 0.30843907594680786,
|
| 680 |
+
"learning_rate": 9.393467000496344e-05,
|
| 681 |
+
"loss": 0.0288,
|
| 682 |
+
"step": 790
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 1.2281711763577048,
|
| 686 |
+
"grad_norm": 0.21833810210227966,
|
| 687 |
+
"learning_rate": 9.367612964226218e-05,
|
| 688 |
+
"loss": 0.0321,
|
| 689 |
+
"step": 800
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 1.2281711763577048,
|
| 693 |
+
"eval_loss": 0.03326805680990219,
|
| 694 |
+
"eval_runtime": 198.7649,
|
| 695 |
+
"eval_samples_per_second": 5.831,
|
| 696 |
+
"eval_steps_per_second": 5.831,
|
| 697 |
+
"step": 800
|
| 698 |
+
},
|
| 699 |
+
{
|
| 700 |
+
"epoch": 1.2435233160621761,
|
| 701 |
+
"grad_norm": 0.08195030689239502,
|
| 702 |
+
"learning_rate": 9.341256468545122e-05,
|
| 703 |
+
"loss": 0.0211,
|
| 704 |
+
"step": 810
|
| 705 |
+
},
|
| 706 |
+
{
|
| 707 |
+
"epoch": 1.2588754557666475,
|
| 708 |
+
"grad_norm": 0.22833506762981415,
|
| 709 |
+
"learning_rate": 9.314400545559934e-05,
|
| 710 |
+
"loss": 0.029,
|
| 711 |
+
"step": 820
|
| 712 |
+
},
|
| 713 |
+
{
|
| 714 |
+
"epoch": 1.2742275954711189,
|
| 715 |
+
"grad_norm": 0.3788512945175171,
|
| 716 |
+
"learning_rate": 9.287048284832698e-05,
|
| 717 |
+
"loss": 0.0291,
|
| 718 |
+
"step": 830
|
| 719 |
+
},
|
| 720 |
+
{
|
| 721 |
+
"epoch": 1.28957973517559,
|
| 722 |
+
"grad_norm": 0.31537115573883057,
|
| 723 |
+
"learning_rate": 9.2592028330252e-05,
|
| 724 |
+
"loss": 0.033,
|
| 725 |
+
"step": 840
|
| 726 |
+
},
|
| 727 |
+
{
|
| 728 |
+
"epoch": 1.3049318748800613,
|
| 729 |
+
"grad_norm": 0.21554504334926605,
|
| 730 |
+
"learning_rate": 9.230867393536972e-05,
|
| 731 |
+
"loss": 0.0299,
|
| 732 |
+
"step": 850
|
| 733 |
+
},
|
| 734 |
+
{
|
| 735 |
+
"epoch": 1.3049318748800613,
|
| 736 |
+
"eval_loss": 0.03233664855360985,
|
| 737 |
+
"eval_runtime": 198.455,
|
| 738 |
+
"eval_samples_per_second": 5.84,
|
| 739 |
+
"eval_steps_per_second": 5.84,
|
| 740 |
+
"step": 850
|
| 741 |
+
},
|
| 742 |
+
{
|
| 743 |
+
"epoch": 1.3202840145845327,
|
| 744 |
+
"grad_norm": 0.2527925670146942,
|
| 745 |
+
"learning_rate": 9.202045226136757e-05,
|
| 746 |
+
"loss": 0.0253,
|
| 747 |
+
"step": 860
|
| 748 |
+
},
|
| 749 |
+
{
|
| 750 |
+
"epoch": 1.335636154289004,
|
| 751 |
+
"grad_norm": 0.16620400547981262,
|
| 752 |
+
"learning_rate": 9.172739646587509e-05,
|
| 753 |
+
"loss": 0.0222,
|
| 754 |
+
"step": 870
|
| 755 |
+
},
|
| 756 |
+
{
|
| 757 |
+
"epoch": 1.3509882939934754,
|
| 758 |
+
"grad_norm": 0.12310031056404114,
|
| 759 |
+
"learning_rate": 9.142954026264931e-05,
|
| 760 |
+
"loss": 0.0262,
|
| 761 |
+
"step": 880
|
| 762 |
+
},
|
| 763 |
+
{
|
| 764 |
+
"epoch": 1.3663404336979466,
|
| 765 |
+
"grad_norm": 0.16346514225006104,
|
| 766 |
+
"learning_rate": 9.112691791769634e-05,
|
| 767 |
+
"loss": 0.028,
|
| 768 |
+
"step": 890
|
| 769 |
+
},
|
| 770 |
+
{
|
| 771 |
+
"epoch": 1.381692573402418,
|
| 772 |
+
"grad_norm": 0.31864503026008606,
|
| 773 |
+
"learning_rate": 9.081956424532926e-05,
|
| 774 |
+
"loss": 0.0286,
|
| 775 |
+
"step": 900
|
| 776 |
+
},
|
| 777 |
+
{
|
| 778 |
+
"epoch": 1.381692573402418,
|
| 779 |
+
"eval_loss": 0.034016843885183334,
|
| 780 |
+
"eval_runtime": 198.4293,
|
| 781 |
+
"eval_samples_per_second": 5.841,
|
| 782 |
+
"eval_steps_per_second": 5.841,
|
| 783 |
+
"step": 900
|
| 784 |
+
},
|
| 785 |
+
{
|
| 786 |
+
"epoch": 1.3970447131068893,
|
| 787 |
+
"grad_norm": 0.36539778113365173,
|
| 788 |
+
"learning_rate": 9.050751460416305e-05,
|
| 789 |
+
"loss": 0.032,
|
| 790 |
+
"step": 910
|
| 791 |
+
},
|
| 792 |
+
{
|
| 793 |
+
"epoch": 1.4123968528113606,
|
| 794 |
+
"grad_norm": 0.13684117794036865,
|
| 795 |
+
"learning_rate": 9.019080489304685e-05,
|
| 796 |
+
"loss": 0.0305,
|
| 797 |
+
"step": 920
|
| 798 |
+
},
|
| 799 |
+
{
|
| 800 |
+
"epoch": 1.427748992515832,
|
| 801 |
+
"grad_norm": 0.32496410608291626,
|
| 802 |
+
"learning_rate": 8.986947154693408e-05,
|
| 803 |
+
"loss": 0.0395,
|
| 804 |
+
"step": 930
|
| 805 |
+
},
|
| 806 |
+
{
|
| 807 |
+
"epoch": 1.4431011322203031,
|
| 808 |
+
"grad_norm": 0.29375186562538147,
|
| 809 |
+
"learning_rate": 8.954355153269088e-05,
|
| 810 |
+
"loss": 0.0381,
|
| 811 |
+
"step": 940
|
| 812 |
+
},
|
| 813 |
+
{
|
| 814 |
+
"epoch": 1.4584532719247745,
|
| 815 |
+
"grad_norm": 0.20962753891944885,
|
| 816 |
+
"learning_rate": 8.921308234484336e-05,
|
| 817 |
+
"loss": 0.0266,
|
| 818 |
+
"step": 950
|
| 819 |
+
},
|
| 820 |
+
{
|
| 821 |
+
"epoch": 1.4584532719247745,
|
| 822 |
+
"eval_loss": 0.03320714086294174,
|
| 823 |
+
"eval_runtime": 198.354,
|
| 824 |
+
"eval_samples_per_second": 5.843,
|
| 825 |
+
"eval_steps_per_second": 5.843,
|
| 826 |
+
"step": 950
|
| 827 |
+
},
|
| 828 |
+
{
|
| 829 |
+
"epoch": 1.4738054116292458,
|
| 830 |
+
"grad_norm": 0.13867303729057312,
|
| 831 |
+
"learning_rate": 8.887810200126419e-05,
|
| 832 |
+
"loss": 0.0266,
|
| 833 |
+
"step": 960
|
| 834 |
+
},
|
| 835 |
+
{
|
| 836 |
+
"epoch": 1.4891575513337172,
|
| 837 |
+
"grad_norm": 0.26519712805747986,
|
| 838 |
+
"learning_rate": 8.853864903879889e-05,
|
| 839 |
+
"loss": 0.0323,
|
| 840 |
+
"step": 970
|
| 841 |
+
},
|
| 842 |
+
{
|
| 843 |
+
"epoch": 1.5045096910381885,
|
| 844 |
+
"grad_norm": 0.11606111377477646,
|
| 845 |
+
"learning_rate": 8.81947625088325e-05,
|
| 846 |
+
"loss": 0.0304,
|
| 847 |
+
"step": 980
|
| 848 |
+
},
|
| 849 |
+
{
|
| 850 |
+
"epoch": 1.5198618307426597,
|
| 851 |
+
"grad_norm": 0.1499972641468048,
|
| 852 |
+
"learning_rate": 8.784648197279701e-05,
|
| 853 |
+
"loss": 0.0297,
|
| 854 |
+
"step": 990
|
| 855 |
+
},
|
| 856 |
+
{
|
| 857 |
+
"epoch": 1.535213970447131,
|
| 858 |
+
"grad_norm": 0.2889319062232971,
|
| 859 |
+
"learning_rate": 8.749384749762015e-05,
|
| 860 |
+
"loss": 0.0296,
|
| 861 |
+
"step": 1000
|
| 862 |
+
},
|
| 863 |
+
{
|
| 864 |
+
"epoch": 1.535213970447131,
|
| 865 |
+
"eval_loss": 0.032004013657569885,
|
| 866 |
+
"eval_runtime": 197.6915,
|
| 867 |
+
"eval_samples_per_second": 5.863,
|
| 868 |
+
"eval_steps_per_second": 5.863,
|
| 869 |
+
"step": 1000
|
| 870 |
+
},
|
| 871 |
+
{
|
| 872 |
+
"epoch": 1.5505661101516024,
|
| 873 |
+
"grad_norm": 0.18784977495670319,
|
| 874 |
+
"learning_rate": 8.713689965111602e-05,
|
| 875 |
+
"loss": 0.022,
|
| 876 |
+
"step": 1010
|
| 877 |
+
},
|
| 878 |
+
{
|
| 879 |
+
"epoch": 1.5659182498560735,
|
| 880 |
+
"grad_norm": 0.21896380186080933,
|
| 881 |
+
"learning_rate": 8.677567949731801e-05,
|
| 882 |
+
"loss": 0.0315,
|
| 883 |
+
"step": 1020
|
| 884 |
+
},
|
| 885 |
+
{
|
| 886 |
+
"epoch": 1.581270389560545,
|
| 887 |
+
"grad_norm": 0.3686976730823517,
|
| 888 |
+
"learning_rate": 8.64102285917548e-05,
|
| 889 |
+
"loss": 0.0445,
|
| 890 |
+
"step": 1030
|
| 891 |
+
},
|
| 892 |
+
{
|
| 893 |
+
"epoch": 1.5966225292650162,
|
| 894 |
+
"grad_norm": 0.16907860338687897,
|
| 895 |
+
"learning_rate": 8.604058897666962e-05,
|
| 896 |
+
"loss": 0.0327,
|
| 897 |
+
"step": 1040
|
| 898 |
+
},
|
| 899 |
+
{
|
| 900 |
+
"epoch": 1.6119746689694876,
|
| 901 |
+
"grad_norm": 0.20729367434978485,
|
| 902 |
+
"learning_rate": 8.566680317618377e-05,
|
| 903 |
+
"loss": 0.022,
|
| 904 |
+
"step": 1050
|
| 905 |
+
},
|
| 906 |
+
{
|
| 907 |
+
"epoch": 1.6119746689694876,
|
| 908 |
+
"eval_loss": 0.03071741759777069,
|
| 909 |
+
"eval_runtime": 197.565,
|
| 910 |
+
"eval_samples_per_second": 5.866,
|
| 911 |
+
"eval_steps_per_second": 5.866,
|
| 912 |
+
"step": 1050
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 1.627326808673959,
|
| 916 |
+
"grad_norm": 0.13343840837478638,
|
| 917 |
+
"learning_rate": 8.528891419140438e-05,
|
| 918 |
+
"loss": 0.0327,
|
| 919 |
+
"step": 1060
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 1.64267894837843,
|
| 923 |
+
"grad_norm": 0.12018793076276779,
|
| 924 |
+
"learning_rate": 8.490696549547761e-05,
|
| 925 |
+
"loss": 0.0212,
|
| 926 |
+
"step": 1070
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 1.6580310880829017,
|
| 930 |
+
"grad_norm": 0.36419177055358887,
|
| 931 |
+
"learning_rate": 8.452100102858734e-05,
|
| 932 |
+
"loss": 0.0389,
|
| 933 |
+
"step": 1080
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 1.6733832277873728,
|
| 937 |
+
"grad_norm": 0.18621623516082764,
|
| 938 |
+
"learning_rate": 8.413106519290023e-05,
|
| 939 |
+
"loss": 0.028,
|
| 940 |
+
"step": 1090
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 1.6887353674918442,
|
| 944 |
+
"grad_norm": 0.13192850351333618,
|
| 945 |
+
"learning_rate": 8.373720284745757e-05,
|
| 946 |
+
"loss": 0.0292,
|
| 947 |
+
"step": 1100
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 1.6887353674918442,
|
| 951 |
+
"eval_loss": 0.031201008707284927,
|
| 952 |
+
"eval_runtime": 197.3374,
|
| 953 |
+
"eval_samples_per_second": 5.873,
|
| 954 |
+
"eval_steps_per_second": 5.873,
|
| 955 |
+
"step": 1100
|
| 956 |
+
},
|
| 957 |
+
{
|
| 958 |
+
"epoch": 1.7040875071963155,
|
| 959 |
+
"grad_norm": 0.15397463738918304,
|
| 960 |
+
"learning_rate": 8.333945930301459e-05,
|
| 961 |
+
"loss": 0.0275,
|
| 962 |
+
"step": 1110
|
| 963 |
+
},
|
| 964 |
+
{
|
| 965 |
+
"epoch": 1.7194396469007867,
|
| 966 |
+
"grad_norm": 0.1929481476545334,
|
| 967 |
+
"learning_rate": 8.293788031682789e-05,
|
| 968 |
+
"loss": 0.0301,
|
| 969 |
+
"step": 1120
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 1.7347917866052582,
|
| 973 |
+
"grad_norm": 0.1465729922056198,
|
| 974 |
+
"learning_rate": 8.253251208739137e-05,
|
| 975 |
+
"loss": 0.0236,
|
| 976 |
+
"step": 1130
|
| 977 |
+
},
|
| 978 |
+
{
|
| 979 |
+
"epoch": 1.7501439263097294,
|
| 980 |
+
"grad_norm": 0.27143338322639465,
|
| 981 |
+
"learning_rate": 8.21234012491215e-05,
|
| 982 |
+
"loss": 0.0238,
|
| 983 |
+
"step": 1140
|
| 984 |
+
},
|
| 985 |
+
{
|
| 986 |
+
"epoch": 1.7654960660142007,
|
| 987 |
+
"grad_norm": 0.16392782330513,
|
| 988 |
+
"learning_rate": 8.171059486699224e-05,
|
| 989 |
+
"loss": 0.0269,
|
| 990 |
+
"step": 1150
|
| 991 |
+
},
|
| 992 |
+
{
|
| 993 |
+
"epoch": 1.7654960660142007,
|
| 994 |
+
"eval_loss": 0.03296918794512749,
|
| 995 |
+
"eval_runtime": 197.4211,
|
| 996 |
+
"eval_samples_per_second": 5.871,
|
| 997 |
+
"eval_steps_per_second": 5.871,
|
| 998 |
+
"step": 1150
|
| 999 |
+
},
|
| 1000 |
+
{
|
| 1001 |
+
"epoch": 1.780848205718672,
|
| 1002 |
+
"grad_norm": 0.12399645149707794,
|
| 1003 |
+
"learning_rate": 8.129414043112087e-05,
|
| 1004 |
+
"loss": 0.029,
|
| 1005 |
+
"step": 1160
|
| 1006 |
+
},
|
| 1007 |
+
{
|
| 1008 |
+
"epoch": 1.7962003454231432,
|
| 1009 |
+
"grad_norm": 0.15697214007377625,
|
| 1010 |
+
"learning_rate": 8.087408585130438e-05,
|
| 1011 |
+
"loss": 0.0271,
|
| 1012 |
+
"step": 1170
|
| 1013 |
+
},
|
| 1014 |
+
{
|
| 1015 |
+
"epoch": 1.8115524851276148,
|
| 1016 |
+
"grad_norm": 0.11034557968378067,
|
| 1017 |
+
"learning_rate": 8.045047945150796e-05,
|
| 1018 |
+
"loss": 0.0312,
|
| 1019 |
+
"step": 1180
|
| 1020 |
+
},
|
| 1021 |
+
{
|
| 1022 |
+
"epoch": 1.826904624832086,
|
| 1023 |
+
"grad_norm": 0.12432292848825455,
|
| 1024 |
+
"learning_rate": 8.002336996430561e-05,
|
| 1025 |
+
"loss": 0.0265,
|
| 1026 |
+
"step": 1190
|
| 1027 |
+
},
|
| 1028 |
+
{
|
| 1029 |
+
"epoch": 1.8422567645365573,
|
| 1030 |
+
"grad_norm": 0.14914625883102417,
|
| 1031 |
+
"learning_rate": 7.959280652527394e-05,
|
| 1032 |
+
"loss": 0.0204,
|
| 1033 |
+
"step": 1200
|
| 1034 |
+
},
|
| 1035 |
+
{
|
| 1036 |
+
"epoch": 1.8422567645365573,
|
| 1037 |
+
"eval_loss": 0.030618194490671158,
|
| 1038 |
+
"eval_runtime": 197.353,
|
| 1039 |
+
"eval_samples_per_second": 5.873,
|
| 1040 |
+
"eval_steps_per_second": 5.873,
|
| 1041 |
+
"step": 1200
|
| 1042 |
+
},
|
| 1043 |
+
{
|
| 1044 |
+
"epoch": 1.8576089042410286,
|
| 1045 |
+
"grad_norm": 0.14602138102054596,
|
| 1046 |
+
"learning_rate": 7.915883866733946e-05,
|
| 1047 |
+
"loss": 0.0295,
|
| 1048 |
+
"step": 1210
|
| 1049 |
+
},
|
| 1050 |
+
{
|
| 1051 |
+
"epoch": 1.8729610439454998,
|
| 1052 |
+
"grad_norm": 0.30296677350997925,
|
| 1053 |
+
"learning_rate": 7.872151631508022e-05,
|
| 1054 |
+
"loss": 0.0301,
|
| 1055 |
+
"step": 1220
|
| 1056 |
+
},
|
| 1057 |
+
{
|
| 1058 |
+
"epoch": 1.8883131836499714,
|
| 1059 |
+
"grad_norm": 0.1722898930311203,
|
| 1060 |
+
"learning_rate": 7.828088977898234e-05,
|
| 1061 |
+
"loss": 0.0292,
|
| 1062 |
+
"step": 1230
|
| 1063 |
+
},
|
| 1064 |
+
{
|
| 1065 |
+
"epoch": 1.9036653233544425,
|
| 1066 |
+
"grad_norm": 0.21365255117416382,
|
| 1067 |
+
"learning_rate": 7.783700974965225e-05,
|
| 1068 |
+
"loss": 0.036,
|
| 1069 |
+
"step": 1240
|
| 1070 |
+
},
|
| 1071 |
+
{
|
| 1072 |
+
"epoch": 1.9190174630589139,
|
| 1073 |
+
"grad_norm": 0.15060608088970184,
|
| 1074 |
+
"learning_rate": 7.738992729198511e-05,
|
| 1075 |
+
"loss": 0.0306,
|
| 1076 |
+
"step": 1250
|
| 1077 |
+
},
|
| 1078 |
+
{
|
| 1079 |
+
"epoch": 1.9190174630589139,
|
| 1080 |
+
"eval_loss": 0.030947549268603325,
|
| 1081 |
+
"eval_runtime": 197.5552,
|
| 1082 |
+
"eval_samples_per_second": 5.867,
|
| 1083 |
+
"eval_steps_per_second": 5.867,
|
| 1084 |
+
"step": 1250
|
| 1085 |
+
},
|
| 1086 |
+
{
|
| 1087 |
+
"epoch": 1.9343696027633852,
|
| 1088 |
+
"grad_norm": 0.1173550933599472,
|
| 1089 |
+
"learning_rate": 7.693969383929017e-05,
|
| 1090 |
+
"loss": 0.02,
|
| 1091 |
+
"step": 1260
|
| 1092 |
+
},
|
| 1093 |
+
{
|
| 1094 |
+
"epoch": 1.9497217424678563,
|
| 1095 |
+
"grad_norm": 0.2480284571647644,
|
| 1096 |
+
"learning_rate": 7.648636118737385e-05,
|
| 1097 |
+
"loss": 0.0353,
|
| 1098 |
+
"step": 1270
|
| 1099 |
+
},
|
| 1100 |
+
{
|
| 1101 |
+
"epoch": 1.965073882172328,
|
| 1102 |
+
"grad_norm": 0.2729029655456543,
|
| 1103 |
+
"learning_rate": 7.602998148858089e-05,
|
| 1104 |
+
"loss": 0.024,
|
| 1105 |
+
"step": 1280
|
| 1106 |
+
},
|
| 1107 |
+
{
|
| 1108 |
+
"epoch": 1.980426021876799,
|
| 1109 |
+
"grad_norm": 0.29488909244537354,
|
| 1110 |
+
"learning_rate": 7.557060724579484e-05,
|
| 1111 |
+
"loss": 0.0276,
|
| 1112 |
+
"step": 1290
|
| 1113 |
+
},
|
| 1114 |
+
{
|
| 1115 |
+
"epoch": 1.9957781615812704,
|
| 1116 |
+
"grad_norm": 0.451770544052124,
|
| 1117 |
+
"learning_rate": 7.51082913063978e-05,
|
| 1118 |
+
"loss": 0.0364,
|
| 1119 |
+
"step": 1300
|
| 1120 |
+
},
|
| 1121 |
+
{
|
| 1122 |
+
"epoch": 1.9957781615812704,
|
| 1123 |
+
"eval_loss": 0.03137281537055969,
|
| 1124 |
+
"eval_runtime": 197.4511,
|
| 1125 |
+
"eval_samples_per_second": 5.87,
|
| 1126 |
+
"eval_steps_per_second": 5.87,
|
| 1127 |
+
"step": 1300
|
| 1128 |
+
},
|
| 1129 |
+
{
|
| 1130 |
+
"epoch": 2.0111303012857418,
|
| 1131 |
+
"grad_norm": 0.20203380286693573,
|
| 1132 |
+
"learning_rate": 7.464308685619099e-05,
|
| 1133 |
+
"loss": 0.0225,
|
| 1134 |
+
"step": 1310
|
| 1135 |
+
},
|
| 1136 |
+
{
|
| 1137 |
+
"epoch": 2.026482440990213,
|
| 1138 |
+
"grad_norm": 0.2726834714412689,
|
| 1139 |
+
"learning_rate": 7.417504741327587e-05,
|
| 1140 |
+
"loss": 0.017,
|
| 1141 |
+
"step": 1320
|
| 1142 |
+
},
|
| 1143 |
+
{
|
| 1144 |
+
"epoch": 2.0418345806946845,
|
| 1145 |
+
"grad_norm": 0.30587098002433777,
|
| 1146 |
+
"learning_rate": 7.370422682189755e-05,
|
| 1147 |
+
"loss": 0.0183,
|
| 1148 |
+
"step": 1330
|
| 1149 |
+
},
|
| 1150 |
+
{
|
| 1151 |
+
"epoch": 2.0571867203991556,
|
| 1152 |
+
"grad_norm": 0.11235737800598145,
|
| 1153 |
+
"learning_rate": 7.323067924625024e-05,
|
| 1154 |
+
"loss": 0.0171,
|
| 1155 |
+
"step": 1340
|
| 1156 |
+
},
|
| 1157 |
+
{
|
| 1158 |
+
"epoch": 2.0725388601036268,
|
| 1159 |
+
"grad_norm": 0.19976893067359924,
|
| 1160 |
+
"learning_rate": 7.275445916424627e-05,
|
| 1161 |
+
"loss": 0.0194,
|
| 1162 |
+
"step": 1350
|
| 1163 |
+
},
|
| 1164 |
+
{
|
| 1165 |
+
"epoch": 2.0725388601036268,
|
| 1166 |
+
"eval_loss": 0.0319148451089859,
|
| 1167 |
+
"eval_runtime": 193.9364,
|
| 1168 |
+
"eval_samples_per_second": 5.976,
|
| 1169 |
+
"eval_steps_per_second": 5.976,
|
| 1170 |
+
"step": 1350
|
| 1171 |
+
},
|
| 1172 |
+
{
|
| 1173 |
+
"epoch": 2.0878909998080983,
|
| 1174 |
+
"grad_norm": 0.26515206694602966,
|
| 1175 |
+
"learning_rate": 7.227562136124864e-05,
|
| 1176 |
+
"loss": 0.0141,
|
| 1177 |
+
"step": 1360
|
| 1178 |
+
},
|
| 1179 |
+
{
|
| 1180 |
+
"epoch": 2.1032431395125695,
|
| 1181 |
+
"grad_norm": 0.15498653054237366,
|
| 1182 |
+
"learning_rate": 7.179422092376856e-05,
|
| 1183 |
+
"loss": 0.0159,
|
| 1184 |
+
"step": 1370
|
| 1185 |
+
},
|
| 1186 |
+
{
|
| 1187 |
+
"epoch": 2.118595279217041,
|
| 1188 |
+
"grad_norm": 0.32679131627082825,
|
| 1189 |
+
"learning_rate": 7.13103132331281e-05,
|
| 1190 |
+
"loss": 0.0198,
|
| 1191 |
+
"step": 1380
|
| 1192 |
+
},
|
| 1193 |
+
{
|
| 1194 |
+
"epoch": 2.133947418921512,
|
| 1195 |
+
"grad_norm": 0.28593504428863525,
|
| 1196 |
+
"learning_rate": 7.082395395908903e-05,
|
| 1197 |
+
"loss": 0.0221,
|
| 1198 |
+
"step": 1390
|
| 1199 |
+
},
|
| 1200 |
+
{
|
| 1201 |
+
"epoch": 2.1492995586259833,
|
| 1202 |
+
"grad_norm": 0.30774274468421936,
|
| 1203 |
+
"learning_rate": 7.033519905344846e-05,
|
| 1204 |
+
"loss": 0.0148,
|
| 1205 |
+
"step": 1400
|
| 1206 |
+
},
|
| 1207 |
+
{
|
| 1208 |
+
"epoch": 2.1492995586259833,
|
| 1209 |
+
"eval_loss": 0.03176383674144745,
|
| 1210 |
+
"eval_runtime": 196.6326,
|
| 1211 |
+
"eval_samples_per_second": 5.894,
|
| 1212 |
+
"eval_steps_per_second": 5.894,
|
| 1213 |
+
"step": 1400
|
| 1214 |
+
},
|
| 1215 |
+
{
|
| 1216 |
+
"epoch": 2.164651698330455,
|
| 1217 |
+
"grad_norm": 0.15440446138381958,
|
| 1218 |
+
"learning_rate": 6.984410474360195e-05,
|
| 1219 |
+
"loss": 0.0181,
|
| 1220 |
+
"step": 1410
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 2.180003838034926,
|
| 1224 |
+
"grad_norm": 0.17429502308368683,
|
| 1225 |
+
"learning_rate": 6.935072752607511e-05,
|
| 1226 |
+
"loss": 0.0228,
|
| 1227 |
+
"step": 1420
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"epoch": 2.1953559777393976,
|
| 1231 |
+
"grad_norm": 0.1579236090183258,
|
| 1232 |
+
"learning_rate": 6.885512416002412e-05,
|
| 1233 |
+
"loss": 0.0247,
|
| 1234 |
+
"step": 1430
|
| 1235 |
+
},
|
| 1236 |
+
{
|
| 1237 |
+
"epoch": 2.2107081174438687,
|
| 1238 |
+
"grad_norm": 0.18410032987594604,
|
| 1239 |
+
"learning_rate": 6.835735166070587e-05,
|
| 1240 |
+
"loss": 0.0179,
|
| 1241 |
+
"step": 1440
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 2.22606025714834,
|
| 1245 |
+
"grad_norm": 0.21408940851688385,
|
| 1246 |
+
"learning_rate": 6.785746729291897e-05,
|
| 1247 |
+
"loss": 0.0161,
|
| 1248 |
+
"step": 1450
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 2.22606025714834,
|
| 1252 |
+
"eval_loss": 0.030531086027622223,
|
| 1253 |
+
"eval_runtime": 196.8522,
|
| 1254 |
+
"eval_samples_per_second": 5.888,
|
| 1255 |
+
"eval_steps_per_second": 5.888,
|
| 1256 |
+
"step": 1450
|
| 1257 |
+
},
|
| 1258 |
+
{
|
| 1259 |
+
"epoch": 2.2414123968528115,
|
| 1260 |
+
"grad_norm": 0.1800948679447174,
|
| 1261 |
+
"learning_rate": 6.735552856441585e-05,
|
| 1262 |
+
"loss": 0.0216,
|
| 1263 |
+
"step": 1460
|
| 1264 |
+
},
|
| 1265 |
+
{
|
| 1266 |
+
"epoch": 2.2567645365572826,
|
| 1267 |
+
"grad_norm": 0.19375424087047577,
|
| 1268 |
+
"learning_rate": 6.685159321928691e-05,
|
| 1269 |
+
"loss": 0.0189,
|
| 1270 |
+
"step": 1470
|
| 1271 |
+
},
|
| 1272 |
+
{
|
| 1273 |
+
"epoch": 2.272116676261754,
|
| 1274 |
+
"grad_norm": 0.2888384163379669,
|
| 1275 |
+
"learning_rate": 6.634571923131756e-05,
|
| 1276 |
+
"loss": 0.0177,
|
| 1277 |
+
"step": 1480
|
| 1278 |
+
},
|
| 1279 |
+
{
|
| 1280 |
+
"epoch": 2.2874688159662253,
|
| 1281 |
+
"grad_norm": 0.1332111358642578,
|
| 1282 |
+
"learning_rate": 6.583796479731872e-05,
|
| 1283 |
+
"loss": 0.0172,
|
| 1284 |
+
"step": 1490
|
| 1285 |
+
},
|
| 1286 |
+
{
|
| 1287 |
+
"epoch": 2.3028209556706964,
|
| 1288 |
+
"grad_norm": 0.09404625743627548,
|
| 1289 |
+
"learning_rate": 6.532838833043189e-05,
|
| 1290 |
+
"loss": 0.0293,
|
| 1291 |
+
"step": 1500
|
| 1292 |
+
},
|
| 1293 |
+
{
|
| 1294 |
+
"epoch": 2.3028209556706964,
|
| 1295 |
+
"eval_loss": 0.03225649148225784,
|
| 1296 |
+
"eval_runtime": 196.8453,
|
| 1297 |
+
"eval_samples_per_second": 5.888,
|
| 1298 |
+
"eval_steps_per_second": 5.888,
|
| 1299 |
+
"step": 1500
|
| 1300 |
+
},
|
| 1301 |
+
{
|
| 1302 |
+
"epoch": 2.318173095375168,
|
| 1303 |
+
"grad_norm": 0.1909513920545578,
|
| 1304 |
+
"learning_rate": 6.481704845340894e-05,
|
| 1305 |
+
"loss": 0.0207,
|
| 1306 |
+
"step": 1510
|
| 1307 |
+
},
|
| 1308 |
+
{
|
| 1309 |
+
"epoch": 2.333525235079639,
|
| 1310 |
+
"grad_norm": 0.34329941868782043,
|
| 1311 |
+
"learning_rate": 6.43040039918683e-05,
|
| 1312 |
+
"loss": 0.0178,
|
| 1313 |
+
"step": 1520
|
| 1314 |
+
},
|
| 1315 |
+
{
|
| 1316 |
+
"epoch": 2.3488773747841103,
|
| 1317 |
+
"grad_norm": 0.28039267659187317,
|
| 1318 |
+
"learning_rate": 6.37893139675273e-05,
|
| 1319 |
+
"loss": 0.0178,
|
| 1320 |
+
"step": 1530
|
| 1321 |
+
},
|
| 1322 |
+
{
|
| 1323 |
+
"epoch": 2.364229514488582,
|
| 1324 |
+
"grad_norm": 0.2201843112707138,
|
| 1325 |
+
"learning_rate": 6.327303759141235e-05,
|
| 1326 |
+
"loss": 0.0146,
|
| 1327 |
+
"step": 1540
|
| 1328 |
+
},
|
| 1329 |
+
{
|
| 1330 |
+
"epoch": 2.379581654193053,
|
| 1331 |
+
"grad_norm": 0.37142327427864075,
|
| 1332 |
+
"learning_rate": 6.275523425704707e-05,
|
| 1333 |
+
"loss": 0.0203,
|
| 1334 |
+
"step": 1550
|
| 1335 |
+
},
|
| 1336 |
+
{
|
| 1337 |
+
"epoch": 2.379581654193053,
|
| 1338 |
+
"eval_loss": 0.03294830769300461,
|
| 1339 |
+
"eval_runtime": 197.0925,
|
| 1340 |
+
"eval_samples_per_second": 5.88,
|
| 1341 |
+
"eval_steps_per_second": 5.88,
|
| 1342 |
+
"step": 1550
|
| 1343 |
+
},
|
| 1344 |
+
{
|
| 1345 |
+
"epoch": 2.3949337938975246,
|
| 1346 |
+
"grad_norm": 0.135298952460289,
|
| 1347 |
+
"learning_rate": 6.223596353361961e-05,
|
| 1348 |
+
"loss": 0.0174,
|
| 1349 |
+
"step": 1560
|
| 1350 |
+
},
|
| 1351 |
+
{
|
| 1352 |
+
"epoch": 2.4102859336019957,
|
| 1353 |
+
"grad_norm": 0.07320668548345566,
|
| 1354 |
+
"learning_rate": 6.171528515912965e-05,
|
| 1355 |
+
"loss": 0.0144,
|
| 1356 |
+
"step": 1570
|
| 1357 |
+
},
|
| 1358 |
+
{
|
| 1359 |
+
"epoch": 2.425638073306467,
|
| 1360 |
+
"grad_norm": 0.2828701436519623,
|
| 1361 |
+
"learning_rate": 6.119325903351599e-05,
|
| 1362 |
+
"loss": 0.019,
|
| 1363 |
+
"step": 1580
|
| 1364 |
+
},
|
| 1365 |
+
{
|
| 1366 |
+
"epoch": 2.4409902130109384,
|
| 1367 |
+
"grad_norm": 0.2670722007751465,
|
| 1368 |
+
"learning_rate": 6.0669945211765585e-05,
|
| 1369 |
+
"loss": 0.027,
|
| 1370 |
+
"step": 1590
|
| 1371 |
+
},
|
| 1372 |
+
{
|
| 1373 |
+
"epoch": 2.4563423527154096,
|
| 1374 |
+
"grad_norm": 0.2617368698120117,
|
| 1375 |
+
"learning_rate": 6.0145403897004696e-05,
|
| 1376 |
+
"loss": 0.0235,
|
| 1377 |
+
"step": 1600
|
| 1378 |
+
},
|
| 1379 |
+
{
|
| 1380 |
+
"epoch": 2.4563423527154096,
|
| 1381 |
+
"eval_loss": 0.032713185995817184,
|
| 1382 |
+
"eval_runtime": 196.9905,
|
| 1383 |
+
"eval_samples_per_second": 5.884,
|
| 1384 |
+
"eval_steps_per_second": 5.884,
|
| 1385 |
+
"step": 1600
|
| 1386 |
+
},
|
| 1387 |
+
{
|
| 1388 |
+
"epoch": 2.471694492419881,
|
| 1389 |
+
"grad_norm": 0.29310551285743713,
|
| 1390 |
+
"learning_rate": 5.961969543357292e-05,
|
| 1391 |
+
"loss": 0.0233,
|
| 1392 |
+
"step": 1610
|
| 1393 |
+
},
|
| 1394 |
+
{
|
| 1395 |
+
"epoch": 2.4870466321243523,
|
| 1396 |
+
"grad_norm": 0.16717247664928436,
|
| 1397 |
+
"learning_rate": 5.9092880300081123e-05,
|
| 1398 |
+
"loss": 0.0243,
|
| 1399 |
+
"step": 1620
|
| 1400 |
+
},
|
| 1401 |
+
{
|
| 1402 |
+
"epoch": 2.5023987718288234,
|
| 1403 |
+
"grad_norm": 0.36758190393447876,
|
| 1404 |
+
"learning_rate": 5.8565019102453844e-05,
|
| 1405 |
+
"loss": 0.0221,
|
| 1406 |
+
"step": 1630
|
| 1407 |
+
},
|
| 1408 |
+
{
|
| 1409 |
+
"epoch": 2.517750911533295,
|
| 1410 |
+
"grad_norm": 0.21562321484088898,
|
| 1411 |
+
"learning_rate": 5.8036172566957006e-05,
|
| 1412 |
+
"loss": 0.0193,
|
| 1413 |
+
"step": 1640
|
| 1414 |
+
},
|
| 1415 |
+
{
|
| 1416 |
+
"epoch": 2.533103051237766,
|
| 1417 |
+
"grad_norm": 0.20650765299797058,
|
| 1418 |
+
"learning_rate": 5.750640153321194e-05,
|
| 1419 |
+
"loss": 0.0234,
|
| 1420 |
+
"step": 1650
|
| 1421 |
+
},
|
| 1422 |
+
{
|
| 1423 |
+
"epoch": 2.533103051237766,
|
| 1424 |
+
"eval_loss": 0.031079526990652084,
|
| 1425 |
+
"eval_runtime": 197.1562,
|
| 1426 |
+
"eval_samples_per_second": 5.879,
|
| 1427 |
+
"eval_steps_per_second": 5.879,
|
| 1428 |
+
"step": 1650
|
| 1429 |
+
},
|
| 1430 |
+
{
|
| 1431 |
+
"epoch": 2.5484551909422377,
|
| 1432 |
+
"grad_norm": 0.43781033158302307,
|
| 1433 |
+
"learning_rate": 5.697576694719616e-05,
|
| 1434 |
+
"loss": 0.0183,
|
| 1435 |
+
"step": 1660
|
| 1436 |
+
},
|
| 1437 |
+
{
|
| 1438 |
+
"epoch": 2.563807330646709,
|
| 1439 |
+
"grad_norm": 0.14761963486671448,
|
| 1440 |
+
"learning_rate": 5.644432985423206e-05,
|
| 1441 |
+
"loss": 0.0185,
|
| 1442 |
+
"step": 1670
|
| 1443 |
+
},
|
| 1444 |
+
{
|
| 1445 |
+
"epoch": 2.57915947035118,
|
| 1446 |
+
"grad_norm": 0.32953789830207825,
|
| 1447 |
+
"learning_rate": 5.591215139196414e-05,
|
| 1448 |
+
"loss": 0.0246,
|
| 1449 |
+
"step": 1680
|
| 1450 |
+
},
|
| 1451 |
+
{
|
| 1452 |
+
"epoch": 2.5945116100556516,
|
| 1453 |
+
"grad_norm": 0.2492523342370987,
|
| 1454 |
+
"learning_rate": 5.5379292783325585e-05,
|
| 1455 |
+
"loss": 0.0239,
|
| 1456 |
+
"step": 1690
|
| 1457 |
+
},
|
| 1458 |
+
{
|
| 1459 |
+
"epoch": 2.6098637497601227,
|
| 1460 |
+
"grad_norm": 0.38178038597106934,
|
| 1461 |
+
"learning_rate": 5.4845815329495054e-05,
|
| 1462 |
+
"loss": 0.0227,
|
| 1463 |
+
"step": 1700
|
| 1464 |
+
},
|
| 1465 |
+
{
|
| 1466 |
+
"epoch": 2.6098637497601227,
|
| 1467 |
+
"eval_loss": 0.030748112127184868,
|
| 1468 |
+
"eval_runtime": 197.0897,
|
| 1469 |
+
"eval_samples_per_second": 5.881,
|
| 1470 |
+
"eval_steps_per_second": 5.881,
|
| 1471 |
+
"step": 1700
|
| 1472 |
+
},
|
| 1473 |
+
{
|
| 1474 |
+
"epoch": 2.6252158894645943,
|
| 1475 |
+
"grad_norm": 0.11092711985111237,
|
| 1476 |
+
"learning_rate": 5.431178040284446e-05,
|
| 1477 |
+
"loss": 0.0184,
|
| 1478 |
+
"step": 1710
|
| 1479 |
+
},
|
| 1480 |
+
{
|
| 1481 |
+
"epoch": 2.6405680291690654,
|
| 1482 |
+
"grad_norm": 0.25804007053375244,
|
| 1483 |
+
"learning_rate": 5.377724943987855e-05,
|
| 1484 |
+
"loss": 0.0162,
|
| 1485 |
+
"step": 1720
|
| 1486 |
+
},
|
| 1487 |
+
{
|
| 1488 |
+
"epoch": 2.6559201688735365,
|
| 1489 |
+
"grad_norm": 0.22017376124858856,
|
| 1490 |
+
"learning_rate": 5.324228393416718e-05,
|
| 1491 |
+
"loss": 0.0163,
|
| 1492 |
+
"step": 1730
|
| 1493 |
+
},
|
| 1494 |
+
{
|
| 1495 |
+
"epoch": 2.671272308578008,
|
| 1496 |
+
"grad_norm": 0.515049934387207,
|
| 1497 |
+
"learning_rate": 5.270694542927088e-05,
|
| 1498 |
+
"loss": 0.0191,
|
| 1499 |
+
"step": 1740
|
| 1500 |
+
},
|
| 1501 |
+
{
|
| 1502 |
+
"epoch": 2.6866244482824793,
|
| 1503 |
+
"grad_norm": 0.16297389566898346,
|
| 1504 |
+
"learning_rate": 5.21712955116608e-05,
|
| 1505 |
+
"loss": 0.0147,
|
| 1506 |
+
"step": 1750
|
| 1507 |
+
},
|
| 1508 |
+
{
|
| 1509 |
+
"epoch": 2.6866244482824793,
|
| 1510 |
+
"eval_loss": 0.031325813382864,
|
| 1511 |
+
"eval_runtime": 196.8694,
|
| 1512 |
+
"eval_samples_per_second": 5.887,
|
| 1513 |
+
"eval_steps_per_second": 5.887,
|
| 1514 |
+
"step": 1750
|
| 1515 |
+
},
|
| 1516 |
+
{
|
| 1517 |
+
"epoch": 2.701976587986951,
|
| 1518 |
+
"grad_norm": 0.1251741498708725,
|
| 1519 |
+
"learning_rate": 5.1635395803633666e-05,
|
| 1520 |
+
"loss": 0.0183,
|
| 1521 |
+
"step": 1760
|
| 1522 |
+
},
|
| 1523 |
+
{
|
| 1524 |
+
"epoch": 2.717328727691422,
|
| 1525 |
+
"grad_norm": 0.0920635238289833,
|
| 1526 |
+
"learning_rate": 5.109930795622265e-05,
|
| 1527 |
+
"loss": 0.0139,
|
| 1528 |
+
"step": 1770
|
| 1529 |
+
},
|
| 1530 |
+
{
|
| 1531 |
+
"epoch": 2.732680867395893,
|
| 1532 |
+
"grad_norm": 0.08902430534362793,
|
| 1533 |
+
"learning_rate": 5.056309364210483e-05,
|
| 1534 |
+
"loss": 0.0155,
|
| 1535 |
+
"step": 1780
|
| 1536 |
+
},
|
| 1537 |
+
{
|
| 1538 |
+
"epoch": 2.7480330071003647,
|
| 1539 |
+
"grad_norm": 0.2637597322463989,
|
| 1540 |
+
"learning_rate": 5.002681454850632e-05,
|
| 1541 |
+
"loss": 0.0259,
|
| 1542 |
+
"step": 1790
|
| 1543 |
+
},
|
| 1544 |
+
{
|
| 1545 |
+
"epoch": 2.763385146804836,
|
| 1546 |
+
"grad_norm": 0.17477376759052277,
|
| 1547 |
+
"learning_rate": 4.949053237010554e-05,
|
| 1548 |
+
"loss": 0.0202,
|
| 1549 |
+
"step": 1800
|
| 1550 |
+
},
|
| 1551 |
+
{
|
| 1552 |
+
"epoch": 2.763385146804836,
|
| 1553 |
+
"eval_loss": 0.032162439078092575,
|
| 1554 |
+
"eval_runtime": 196.5492,
|
| 1555 |
+
"eval_samples_per_second": 5.897,
|
| 1556 |
+
"eval_steps_per_second": 5.897,
|
| 1557 |
+
"step": 1800
|
| 1558 |
+
},
|
| 1559 |
+
{
|
| 1560 |
+
"epoch": 2.7787372865093074,
|
| 1561 |
+
"grad_norm": 0.17457661032676697,
|
| 1562 |
+
"learning_rate": 4.89543088019359e-05,
|
| 1563 |
+
"loss": 0.0141,
|
| 1564 |
+
"step": 1810
|
| 1565 |
+
},
|
| 1566 |
+
{
|
| 1567 |
+
"epoch": 2.7940894262137785,
|
| 1568 |
+
"grad_norm": 0.22364993393421173,
|
| 1569 |
+
"learning_rate": 4.841820553228805e-05,
|
| 1570 |
+
"loss": 0.0236,
|
| 1571 |
+
"step": 1820
|
| 1572 |
+
},
|
| 1573 |
+
{
|
| 1574 |
+
"epoch": 2.8094415659182497,
|
| 1575 |
+
"grad_norm": 0.25010716915130615,
|
| 1576 |
+
"learning_rate": 4.7882284235613324e-05,
|
| 1577 |
+
"loss": 0.0269,
|
| 1578 |
+
"step": 1830
|
| 1579 |
+
},
|
| 1580 |
+
{
|
| 1581 |
+
"epoch": 2.8247937056227213,
|
| 1582 |
+
"grad_norm": 0.08725937455892563,
|
| 1583 |
+
"learning_rate": 4.734660656542846e-05,
|
| 1584 |
+
"loss": 0.0172,
|
| 1585 |
+
"step": 1840
|
| 1586 |
+
},
|
| 1587 |
+
{
|
| 1588 |
+
"epoch": 2.8401458453271924,
|
| 1589 |
+
"grad_norm": 0.17154793441295624,
|
| 1590 |
+
"learning_rate": 4.681123414722291e-05,
|
| 1591 |
+
"loss": 0.0203,
|
| 1592 |
+
"step": 1850
|
| 1593 |
+
},
|
| 1594 |
+
{
|
| 1595 |
+
"epoch": 2.8401458453271924,
|
| 1596 |
+
"eval_loss": 0.031308364123106,
|
| 1597 |
+
"eval_runtime": 196.5695,
|
| 1598 |
+
"eval_samples_per_second": 5.896,
|
| 1599 |
+
"eval_steps_per_second": 5.896,
|
| 1600 |
+
"step": 1850
|
| 1601 |
+
},
|
| 1602 |
+
{
|
| 1603 |
+
"epoch": 2.855497985031664,
|
| 1604 |
+
"grad_norm": 0.14235727488994598,
|
| 1605 |
+
"learning_rate": 4.627622857136929e-05,
|
| 1606 |
+
"loss": 0.0157,
|
| 1607 |
+
"step": 1860
|
| 1608 |
+
},
|
| 1609 |
+
{
|
| 1610 |
+
"epoch": 2.870850124736135,
|
| 1611 |
+
"grad_norm": 0.3842136561870575,
|
| 1612 |
+
"learning_rate": 4.5741651386037883e-05,
|
| 1613 |
+
"loss": 0.0216,
|
| 1614 |
+
"step": 1870
|
| 1615 |
+
},
|
| 1616 |
+
{
|
| 1617 |
+
"epoch": 2.8862022644406062,
|
| 1618 |
+
"grad_norm": 0.25185850262641907,
|
| 1619 |
+
"learning_rate": 4.5207564090116064e-05,
|
| 1620 |
+
"loss": 0.0151,
|
| 1621 |
+
"step": 1880
|
| 1622 |
+
},
|
| 1623 |
+
{
|
| 1624 |
+
"epoch": 2.901554404145078,
|
| 1625 |
+
"grad_norm": 0.23601651191711426,
|
| 1626 |
+
"learning_rate": 4.467402812613323e-05,
|
| 1627 |
+
"loss": 0.0148,
|
| 1628 |
+
"step": 1890
|
| 1629 |
+
},
|
| 1630 |
+
{
|
| 1631 |
+
"epoch": 2.916906543849549,
|
| 1632 |
+
"grad_norm": 0.17588479816913605,
|
| 1633 |
+
"learning_rate": 4.414110487319244e-05,
|
| 1634 |
+
"loss": 0.0199,
|
| 1635 |
+
"step": 1900
|
| 1636 |
+
},
|
| 1637 |
+
{
|
| 1638 |
+
"epoch": 2.916906543849549,
|
| 1639 |
+
"eval_loss": 0.031001152470707893,
|
| 1640 |
+
"eval_runtime": 196.7891,
|
| 1641 |
+
"eval_samples_per_second": 5.89,
|
| 1642 |
+
"eval_steps_per_second": 5.89,
|
| 1643 |
+
"step": 1900
|
| 1644 |
+
},
|
| 1645 |
+
{
|
| 1646 |
+
"epoch": 2.9322586835540205,
|
| 1647 |
+
"grad_norm": 0.21399495005607605,
|
| 1648 |
+
"learning_rate": 4.360885563990919e-05,
|
| 1649 |
+
"loss": 0.0174,
|
| 1650 |
+
"step": 1910
|
| 1651 |
+
},
|
| 1652 |
+
{
|
| 1653 |
+
"epoch": 2.9476108232584917,
|
| 1654 |
+
"grad_norm": 0.18000049889087677,
|
| 1655 |
+
"learning_rate": 4.307734165735829e-05,
|
| 1656 |
+
"loss": 0.0123,
|
| 1657 |
+
"step": 1920
|
| 1658 |
+
},
|
| 1659 |
+
{
|
| 1660 |
+
"epoch": 2.962962962962963,
|
| 1661 |
+
"grad_norm": 0.19970905780792236,
|
| 1662 |
+
"learning_rate": 4.254662407202976e-05,
|
| 1663 |
+
"loss": 0.0175,
|
| 1664 |
+
"step": 1930
|
| 1665 |
+
},
|
| 1666 |
+
{
|
| 1667 |
+
"epoch": 2.9783151026674344,
|
| 1668 |
+
"grad_norm": 0.21200795471668243,
|
| 1669 |
+
"learning_rate": 4.201676393879446e-05,
|
| 1670 |
+
"loss": 0.0197,
|
| 1671 |
+
"step": 1940
|
| 1672 |
+
},
|
| 1673 |
+
{
|
| 1674 |
+
"epoch": 2.9936672423719055,
|
| 1675 |
+
"grad_norm": 0.15390953421592712,
|
| 1676 |
+
"learning_rate": 4.148782221388007e-05,
|
| 1677 |
+
"loss": 0.0152,
|
| 1678 |
+
"step": 1950
|
| 1679 |
+
},
|
| 1680 |
+
{
|
| 1681 |
+
"epoch": 2.9936672423719055,
|
| 1682 |
+
"eval_loss": 0.0314854197204113,
|
| 1683 |
+
"eval_runtime": 197.0087,
|
| 1684 |
+
"eval_samples_per_second": 5.883,
|
| 1685 |
+
"eval_steps_per_second": 5.883,
|
| 1686 |
+
"step": 1950
|
| 1687 |
+
},
|
| 1688 |
+
{
|
| 1689 |
+
"epoch": 3.009019382076377,
|
| 1690 |
+
"grad_norm": 0.1752149611711502,
|
| 1691 |
+
"learning_rate": 4.0959859747858706e-05,
|
| 1692 |
+
"loss": 0.0116,
|
| 1693 |
+
"step": 1960
|
| 1694 |
+
},
|
| 1695 |
+
{
|
| 1696 |
+
"epoch": 3.0243715217808482,
|
| 1697 |
+
"grad_norm": 0.22070646286010742,
|
| 1698 |
+
"learning_rate": 4.043293727864644e-05,
|
| 1699 |
+
"loss": 0.0081,
|
| 1700 |
+
"step": 1970
|
| 1701 |
+
},
|
| 1702 |
+
{
|
| 1703 |
+
"epoch": 3.0397236614853194,
|
| 1704 |
+
"grad_norm": 0.39557531476020813,
|
| 1705 |
+
"learning_rate": 3.990711542451591e-05,
|
| 1706 |
+
"loss": 0.0103,
|
| 1707 |
+
"step": 1980
|
| 1708 |
+
},
|
| 1709 |
+
{
|
| 1710 |
+
"epoch": 3.055075801189791,
|
| 1711 |
+
"grad_norm": 0.1403462290763855,
|
| 1712 |
+
"learning_rate": 3.9382454677122704e-05,
|
| 1713 |
+
"loss": 0.0105,
|
| 1714 |
+
"step": 1990
|
| 1715 |
+
},
|
| 1716 |
+
{
|
| 1717 |
+
"epoch": 3.070427940894262,
|
| 1718 |
+
"grad_norm": 0.15649639070034027,
|
| 1719 |
+
"learning_rate": 3.885901539454623e-05,
|
| 1720 |
+
"loss": 0.0065,
|
| 1721 |
+
"step": 2000
|
| 1722 |
+
},
|
| 1723 |
+
{
|
| 1724 |
+
"epoch": 3.070427940894262,
|
| 1725 |
+
"eval_loss": 0.034673649817705154,
|
| 1726 |
+
"eval_runtime": 196.6895,
|
| 1727 |
+
"eval_samples_per_second": 5.893,
|
| 1728 |
+
"eval_steps_per_second": 5.893,
|
| 1729 |
+
"step": 2000
|
| 1730 |
+
},
|
| 1731 |
+
{
|
| 1732 |
+
"epoch": 3.0857800805987337,
|
| 1733 |
+
"grad_norm": 0.26310181617736816,
|
| 1734 |
+
"learning_rate": 3.833685779434597e-05,
|
| 1735 |
+
"loss": 0.0124,
|
| 1736 |
+
"step": 2010
|
| 1737 |
+
},
|
| 1738 |
+
{
|
| 1739 |
+
"epoch": 3.101132220303205,
|
| 1740 |
+
"grad_norm": 0.12814761698246002,
|
| 1741 |
+
"learning_rate": 3.7816041946634024e-05,
|
| 1742 |
+
"loss": 0.0092,
|
| 1743 |
+
"step": 2020
|
| 1744 |
+
},
|
| 1745 |
+
{
|
| 1746 |
+
"epoch": 3.116484360007676,
|
| 1747 |
+
"grad_norm": 0.2951210141181946,
|
| 1748 |
+
"learning_rate": 3.729662776716439e-05,
|
| 1749 |
+
"loss": 0.0085,
|
| 1750 |
+
"step": 2030
|
| 1751 |
+
},
|
| 1752 |
+
{
|
| 1753 |
+
"epoch": 3.1318364997121475,
|
| 1754 |
+
"grad_norm": 0.13861867785453796,
|
| 1755 |
+
"learning_rate": 3.677867501044029e-05,
|
| 1756 |
+
"loss": 0.0067,
|
| 1757 |
+
"step": 2040
|
| 1758 |
+
},
|
| 1759 |
+
{
|
| 1760 |
+
"epoch": 3.1471886394166186,
|
| 1761 |
+
"grad_norm": 0.17344604432582855,
|
| 1762 |
+
"learning_rate": 3.6262243262839654e-05,
|
| 1763 |
+
"loss": 0.0155,
|
| 1764 |
+
"step": 2050
|
| 1765 |
+
},
|
| 1766 |
+
{
|
| 1767 |
+
"epoch": 3.1471886394166186,
|
| 1768 |
+
"eval_loss": 0.034481730312108994,
|
| 1769 |
+
"eval_runtime": 197.6943,
|
| 1770 |
+
"eval_samples_per_second": 5.863,
|
| 1771 |
+
"eval_steps_per_second": 5.863,
|
| 1772 |
+
"step": 2050
|
| 1773 |
+
},
|
| 1774 |
+
{
|
| 1775 |
+
"epoch": 3.16254077912109,
|
| 1776 |
+
"grad_norm": 0.09056749194860458,
|
| 1777 |
+
"learning_rate": 3.574739193576042e-05,
|
| 1778 |
+
"loss": 0.0077,
|
| 1779 |
+
"step": 2060
|
| 1780 |
+
},
|
| 1781 |
+
{
|
| 1782 |
+
"epoch": 3.1778929188255614,
|
| 1783 |
+
"grad_norm": 0.10517852753400803,
|
| 1784 |
+
"learning_rate": 3.5234180258785554e-05,
|
| 1785 |
+
"loss": 0.0092,
|
| 1786 |
+
"step": 2070
|
| 1787 |
+
},
|
| 1788 |
+
{
|
| 1789 |
+
"epoch": 3.1932450585300325,
|
| 1790 |
+
"grad_norm": 0.12550629675388336,
|
| 1791 |
+
"learning_rate": 3.472266727286928e-05,
|
| 1792 |
+
"loss": 0.0077,
|
| 1793 |
+
"step": 2080
|
| 1794 |
+
},
|
| 1795 |
+
{
|
| 1796 |
+
"epoch": 3.208597198234504,
|
| 1797 |
+
"grad_norm": 0.15013106167316437,
|
| 1798 |
+
"learning_rate": 3.4212911823544746e-05,
|
| 1799 |
+
"loss": 0.011,
|
| 1800 |
+
"step": 2090
|
| 1801 |
+
},
|
| 1802 |
+
{
|
| 1803 |
+
"epoch": 3.223949337938975,
|
| 1804 |
+
"grad_norm": 0.10141454637050629,
|
| 1805 |
+
"learning_rate": 3.370497255415443e-05,
|
| 1806 |
+
"loss": 0.0087,
|
| 1807 |
+
"step": 2100
|
| 1808 |
+
},
|
| 1809 |
+
{
|
| 1810 |
+
"epoch": 3.223949337938975,
|
| 1811 |
+
"eval_loss": 0.03667592629790306,
|
| 1812 |
+
"eval_runtime": 198.0424,
|
| 1813 |
+
"eval_samples_per_second": 5.852,
|
| 1814 |
+
"eval_steps_per_second": 5.852,
|
| 1815 |
+
"step": 2100
|
| 1816 |
+
},
|
| 1817 |
+
{
|
| 1818 |
+
"epoch": 3.239301477643447,
|
| 1819 |
+
"grad_norm": 0.34847912192344666,
|
| 1820 |
+
"learning_rate": 3.319890789910364e-05,
|
| 1821 |
+
"loss": 0.009,
|
| 1822 |
+
"step": 2110
|
| 1823 |
+
},
|
| 1824 |
+
{
|
| 1825 |
+
"epoch": 3.254653617347918,
|
| 1826 |
+
"grad_norm": 0.09903811663389206,
|
| 1827 |
+
"learning_rate": 3.269477607713802e-05,
|
| 1828 |
+
"loss": 0.0129,
|
| 1829 |
+
"step": 2120
|
| 1830 |
+
},
|
| 1831 |
+
{
|
| 1832 |
+
"epoch": 3.270005757052389,
|
| 1833 |
+
"grad_norm": 0.35761070251464844,
|
| 1834 |
+
"learning_rate": 3.219263508464614e-05,
|
| 1835 |
+
"loss": 0.0109,
|
| 1836 |
+
"step": 2130
|
| 1837 |
+
},
|
| 1838 |
+
{
|
| 1839 |
+
"epoch": 3.2853578967568606,
|
| 1840 |
+
"grad_norm": 0.8299241662025452,
|
| 1841 |
+
"learning_rate": 3.169254268898725e-05,
|
| 1842 |
+
"loss": 0.0097,
|
| 1843 |
+
"step": 2140
|
| 1844 |
+
},
|
| 1845 |
+
{
|
| 1846 |
+
"epoch": 3.3007100364613318,
|
| 1847 |
+
"grad_norm": 0.034059587866067886,
|
| 1848 |
+
"learning_rate": 3.119455642184572e-05,
|
| 1849 |
+
"loss": 0.0107,
|
| 1850 |
+
"step": 2150
|
| 1851 |
+
},
|
| 1852 |
+
{
|
| 1853 |
+
"epoch": 3.3007100364613318,
|
| 1854 |
+
"eval_loss": 0.03527729585766792,
|
| 1855 |
+
"eval_runtime": 197.9624,
|
| 1856 |
+
"eval_samples_per_second": 5.855,
|
| 1857 |
+
"eval_steps_per_second": 5.855,
|
| 1858 |
+
"step": 2150
|
| 1859 |
+
},
|
| 1860 |
+
{
|
| 1861 |
+
"epoch": 3.3160621761658033,
|
| 1862 |
+
"grad_norm": 0.17827175557613373,
|
| 1863 |
+
"learning_rate": 3.069873357261249e-05,
|
| 1864 |
+
"loss": 0.0101,
|
| 1865 |
+
"step": 2160
|
| 1866 |
+
},
|
| 1867 |
+
{
|
| 1868 |
+
"epoch": 3.3314143158702745,
|
| 1869 |
+
"grad_norm": 0.14705407619476318,
|
| 1870 |
+
"learning_rate": 3.020513118179428e-05,
|
| 1871 |
+
"loss": 0.0128,
|
| 1872 |
+
"step": 2170
|
| 1873 |
+
},
|
| 1874 |
+
{
|
| 1875 |
+
"epoch": 3.3467664555747456,
|
| 1876 |
+
"grad_norm": 0.3249284327030182,
|
| 1877 |
+
"learning_rate": 2.9713806034451652e-05,
|
| 1878 |
+
"loss": 0.0085,
|
| 1879 |
+
"step": 2180
|
| 1880 |
+
},
|
| 1881 |
+
{
|
| 1882 |
+
"epoch": 3.362118595279217,
|
| 1883 |
+
"grad_norm": 0.035756666213274,
|
| 1884 |
+
"learning_rate": 2.9224814653666242e-05,
|
| 1885 |
+
"loss": 0.0075,
|
| 1886 |
+
"step": 2190
|
| 1887 |
+
},
|
| 1888 |
+
{
|
| 1889 |
+
"epoch": 3.3774707349836883,
|
| 1890 |
+
"grad_norm": 0.1439007669687271,
|
| 1891 |
+
"learning_rate": 2.8738213294038212e-05,
|
| 1892 |
+
"loss": 0.0113,
|
| 1893 |
+
"step": 2200
|
| 1894 |
+
},
|
| 1895 |
+
{
|
| 1896 |
+
"epoch": 3.3774707349836883,
|
| 1897 |
+
"eval_loss": 0.03770239278674126,
|
| 1898 |
+
"eval_runtime": 197.5106,
|
| 1899 |
+
"eval_samples_per_second": 5.868,
|
| 1900 |
+
"eval_steps_per_second": 5.868,
|
| 1901 |
+
"step": 2200
|
| 1902 |
+
},
|
| 1903 |
+
{
|
| 1904 |
+
"epoch": 3.39282287468816,
|
| 1905 |
+
"grad_norm": 0.29063209891319275,
|
| 1906 |
+
"learning_rate": 2.8254057935214735e-05,
|
| 1907 |
+
"loss": 0.0108,
|
| 1908 |
+
"step": 2210
|
| 1909 |
+
},
|
| 1910 |
+
{
|
| 1911 |
+
"epoch": 3.408175014392631,
|
| 1912 |
+
"grad_norm": 0.23443534970283508,
|
| 1913 |
+
"learning_rate": 2.7772404275449825e-05,
|
| 1914 |
+
"loss": 0.0109,
|
| 1915 |
+
"step": 2220
|
| 1916 |
+
},
|
| 1917 |
+
{
|
| 1918 |
+
"epoch": 3.423527154097102,
|
| 1919 |
+
"grad_norm": 0.2173459231853485,
|
| 1920 |
+
"learning_rate": 2.7293307725196793e-05,
|
| 1921 |
+
"loss": 0.0095,
|
| 1922 |
+
"step": 2230
|
| 1923 |
+
},
|
| 1924 |
+
{
|
| 1925 |
+
"epoch": 3.4388792938015738,
|
| 1926 |
+
"grad_norm": 0.19652967154979706,
|
| 1927 |
+
"learning_rate": 2.6816823400733625e-05,
|
| 1928 |
+
"loss": 0.0119,
|
| 1929 |
+
"step": 2240
|
| 1930 |
+
},
|
| 1931 |
+
{
|
| 1932 |
+
"epoch": 3.454231433506045,
|
| 1933 |
+
"grad_norm": 0.49155566096305847,
|
| 1934 |
+
"learning_rate": 2.6343006117822437e-05,
|
| 1935 |
+
"loss": 0.0115,
|
| 1936 |
+
"step": 2250
|
| 1937 |
+
},
|
| 1938 |
+
{
|
| 1939 |
+
"epoch": 3.454231433506045,
|
| 1940 |
+
"eval_loss": 0.03576047718524933,
|
| 1941 |
+
"eval_runtime": 198.0048,
|
| 1942 |
+
"eval_samples_per_second": 5.853,
|
| 1943 |
+
"eval_steps_per_second": 5.853,
|
| 1944 |
+
"step": 2250
|
| 1945 |
+
},
|
| 1946 |
+
{
|
| 1947 |
+
"epoch": 3.469583573210516,
|
| 1948 |
+
"grad_norm": 0.04775138571858406,
|
| 1949 |
+
"learning_rate": 2.587191038540317e-05,
|
| 1950 |
+
"loss": 0.0079,
|
| 1951 |
+
"step": 2260
|
| 1952 |
+
},
|
| 1953 |
+
{
|
| 1954 |
+
"epoch": 3.4849357129149876,
|
| 1955 |
+
"grad_norm": 0.3148171603679657,
|
| 1956 |
+
"learning_rate": 2.5403590399322886e-05,
|
| 1957 |
+
"loss": 0.0057,
|
| 1958 |
+
"step": 2270
|
| 1959 |
+
},
|
| 1960 |
+
{
|
| 1961 |
+
"epoch": 3.5002878526194587,
|
| 1962 |
+
"grad_norm": 0.296872615814209,
|
| 1963 |
+
"learning_rate": 2.493810003610092e-05,
|
| 1964 |
+
"loss": 0.0143,
|
| 1965 |
+
"step": 2280
|
| 1966 |
+
},
|
| 1967 |
+
{
|
| 1968 |
+
"epoch": 3.51563999232393,
|
| 1969 |
+
"grad_norm": 0.27500900626182556,
|
| 1970 |
+
"learning_rate": 2.4475492846730835e-05,
|
| 1971 |
+
"loss": 0.0096,
|
| 1972 |
+
"step": 2290
|
| 1973 |
+
},
|
| 1974 |
+
{
|
| 1975 |
+
"epoch": 3.5309921320284015,
|
| 1976 |
+
"grad_norm": 0.2594282031059265,
|
| 1977 |
+
"learning_rate": 2.4015822050519794e-05,
|
| 1978 |
+
"loss": 0.0087,
|
| 1979 |
+
"step": 2300
|
| 1980 |
+
},
|
| 1981 |
+
{
|
| 1982 |
+
"epoch": 3.5309921320284015,
|
| 1983 |
+
"eval_loss": 0.03767295554280281,
|
| 1984 |
+
"eval_runtime": 198.052,
|
| 1985 |
+
"eval_samples_per_second": 5.852,
|
| 1986 |
+
"eval_steps_per_second": 5.852,
|
| 1987 |
+
"step": 2300
|
| 1988 |
+
},
|
| 1989 |
+
{
|
| 1990 |
+
"epoch": 3.546344271732873,
|
| 1991 |
+
"grad_norm": 0.24976056814193726,
|
| 1992 |
+
"learning_rate": 2.3559140528966074e-05,
|
| 1993 |
+
"loss": 0.0072,
|
| 1994 |
+
"step": 2310
|
| 1995 |
+
},
|
| 1996 |
+
{
|
| 1997 |
+
"epoch": 3.561696411437344,
|
| 1998 |
+
"grad_norm": 0.1786753237247467,
|
| 1999 |
+
"learning_rate": 2.3105500819675495e-05,
|
| 2000 |
+
"loss": 0.0096,
|
| 2001 |
+
"step": 2320
|
| 2002 |
+
},
|
| 2003 |
+
{
|
| 2004 |
+
"epoch": 3.5770485511418153,
|
| 2005 |
+
"grad_norm": 0.23791159689426422,
|
| 2006 |
+
"learning_rate": 2.265495511031742e-05,
|
| 2007 |
+
"loss": 0.0129,
|
| 2008 |
+
"step": 2330
|
| 2009 |
+
},
|
| 2010 |
+
{
|
| 2011 |
+
"epoch": 3.5924006908462864,
|
| 2012 |
+
"grad_norm": 0.21382766962051392,
|
| 2013 |
+
"learning_rate": 2.2207555232620893e-05,
|
| 2014 |
+
"loss": 0.0125,
|
| 2015 |
+
"step": 2340
|
| 2016 |
+
},
|
| 2017 |
+
{
|
| 2018 |
+
"epoch": 3.607752830550758,
|
| 2019 |
+
"grad_norm": 0.06649874895811081,
|
| 2020 |
+
"learning_rate": 2.1763352656411785e-05,
|
| 2021 |
+
"loss": 0.0099,
|
| 2022 |
+
"step": 2350
|
| 2023 |
+
},
|
| 2024 |
+
{
|
| 2025 |
+
"epoch": 3.607752830550758,
|
| 2026 |
+
"eval_loss": 0.03737160563468933,
|
| 2027 |
+
"eval_runtime": 198.0602,
|
| 2028 |
+
"eval_samples_per_second": 5.852,
|
| 2029 |
+
"eval_steps_per_second": 5.852,
|
| 2030 |
+
"step": 2350
|
| 2031 |
+
},
|
| 2032 |
+
{
|
| 2033 |
+
"epoch": 3.6231049702552296,
|
| 2034 |
+
"grad_norm": 0.295950710773468,
|
| 2035 |
+
"learning_rate": 2.1322398483691787e-05,
|
| 2036 |
+
"loss": 0.0063,
|
| 2037 |
+
"step": 2360
|
| 2038 |
+
},
|
| 2039 |
+
{
|
| 2040 |
+
"epoch": 3.6384571099597007,
|
| 2041 |
+
"grad_norm": 0.25773197412490845,
|
| 2042 |
+
"learning_rate": 2.08847434427593e-05,
|
| 2043 |
+
"loss": 0.0101,
|
| 2044 |
+
"step": 2370
|
| 2045 |
+
},
|
| 2046 |
+
{
|
| 2047 |
+
"epoch": 3.653809249664172,
|
| 2048 |
+
"grad_norm": 0.4787651002407074,
|
| 2049 |
+
"learning_rate": 2.0450437882373697e-05,
|
| 2050 |
+
"loss": 0.0099,
|
| 2051 |
+
"step": 2380
|
| 2052 |
+
},
|
| 2053 |
+
{
|
| 2054 |
+
"epoch": 3.669161389368643,
|
| 2055 |
+
"grad_norm": 0.1840725690126419,
|
| 2056 |
+
"learning_rate": 2.0019531765962995e-05,
|
| 2057 |
+
"loss": 0.0078,
|
| 2058 |
+
"step": 2390
|
| 2059 |
+
},
|
| 2060 |
+
{
|
| 2061 |
+
"epoch": 3.6845135290731146,
|
| 2062 |
+
"grad_norm": 0.30393171310424805,
|
| 2063 |
+
"learning_rate": 1.9592074665876026e-05,
|
| 2064 |
+
"loss": 0.0075,
|
| 2065 |
+
"step": 2400
|
| 2066 |
+
},
|
| 2067 |
+
{
|
| 2068 |
+
"epoch": 3.6845135290731146,
|
| 2069 |
+
"eval_loss": 0.038148432970047,
|
| 2070 |
+
"eval_runtime": 198.4246,
|
| 2071 |
+
"eval_samples_per_second": 5.841,
|
| 2072 |
+
"eval_steps_per_second": 5.841,
|
| 2073 |
+
"step": 2400
|
| 2074 |
+
},
|
| 2075 |
+
{
|
| 2076 |
+
"epoch": 3.6998656687775857,
|
| 2077 |
+
"grad_norm": 0.07026088237762451,
|
| 2078 |
+
"learning_rate": 1.9168115757679535e-05,
|
| 2079 |
+
"loss": 0.0127,
|
| 2080 |
+
"step": 2410
|
| 2081 |
+
},
|
| 2082 |
+
{
|
| 2083 |
+
"epoch": 3.7152178084820573,
|
| 2084 |
+
"grad_norm": 0.27099308371543884,
|
| 2085 |
+
"learning_rate": 1.8747703814500866e-05,
|
| 2086 |
+
"loss": 0.0086,
|
| 2087 |
+
"step": 2420
|
| 2088 |
+
},
|
| 2089 |
+
{
|
| 2090 |
+
"epoch": 3.7305699481865284,
|
| 2091 |
+
"grad_norm": 0.2625258266925812,
|
| 2092 |
+
"learning_rate": 1.833088720141698e-05,
|
| 2093 |
+
"loss": 0.0085,
|
| 2094 |
+
"step": 2430
|
| 2095 |
+
},
|
| 2096 |
+
{
|
| 2097 |
+
"epoch": 3.7459220878909996,
|
| 2098 |
+
"grad_norm": 0.38898608088493347,
|
| 2099 |
+
"learning_rate": 1.7917713869890557e-05,
|
| 2100 |
+
"loss": 0.0088,
|
| 2101 |
+
"step": 2440
|
| 2102 |
+
},
|
| 2103 |
+
{
|
| 2104 |
+
"epoch": 3.761274227595471,
|
| 2105 |
+
"grad_norm": 0.18794888257980347,
|
| 2106 |
+
"learning_rate": 1.7508231352253435e-05,
|
| 2107 |
+
"loss": 0.0064,
|
| 2108 |
+
"step": 2450
|
| 2109 |
+
},
|
| 2110 |
+
{
|
| 2111 |
+
"epoch": 3.761274227595471,
|
| 2112 |
+
"eval_loss": 0.038376811891794205,
|
| 2113 |
+
"eval_runtime": 198.1296,
|
| 2114 |
+
"eval_samples_per_second": 5.85,
|
| 2115 |
+
"eval_steps_per_second": 5.85,
|
| 2116 |
+
"step": 2450
|
| 2117 |
+
},
|
| 2118 |
+
{
|
| 2119 |
+
"epoch": 3.7766263672999423,
|
| 2120 |
+
"grad_norm": 0.07355306297540665,
|
| 2121 |
+
"learning_rate": 1.7102486756238435e-05,
|
| 2122 |
+
"loss": 0.0086,
|
| 2123 |
+
"step": 2460
|
| 2124 |
+
},
|
| 2125 |
+
{
|
| 2126 |
+
"epoch": 3.791978507004414,
|
| 2127 |
+
"grad_norm": 0.2125598043203354,
|
| 2128 |
+
"learning_rate": 1.6700526759560002e-05,
|
| 2129 |
+
"loss": 0.0109,
|
| 2130 |
+
"step": 2470
|
| 2131 |
+
},
|
| 2132 |
+
{
|
| 2133 |
+
"epoch": 3.807330646708885,
|
| 2134 |
+
"grad_norm": 0.12951001524925232,
|
| 2135 |
+
"learning_rate": 1.6302397604544257e-05,
|
| 2136 |
+
"loss": 0.0119,
|
| 2137 |
+
"step": 2480
|
| 2138 |
+
},
|
| 2139 |
+
{
|
| 2140 |
+
"epoch": 3.822682786413356,
|
| 2141 |
+
"grad_norm": 0.15507496893405914,
|
| 2142 |
+
"learning_rate": 1.5908145092809272e-05,
|
| 2143 |
+
"loss": 0.0066,
|
| 2144 |
+
"step": 2490
|
| 2145 |
+
},
|
| 2146 |
+
{
|
| 2147 |
+
"epoch": 3.8380349261178277,
|
| 2148 |
+
"grad_norm": 0.08303289115428925,
|
| 2149 |
+
"learning_rate": 1.551781457999586e-05,
|
| 2150 |
+
"loss": 0.0111,
|
| 2151 |
+
"step": 2500
|
| 2152 |
+
},
|
| 2153 |
+
{
|
| 2154 |
+
"epoch": 3.8380349261178277,
|
| 2155 |
+
"eval_loss": 0.0382462814450264,
|
| 2156 |
+
"eval_runtime": 197.7703,
|
| 2157 |
+
"eval_samples_per_second": 5.86,
|
| 2158 |
+
"eval_steps_per_second": 5.86,
|
| 2159 |
+
"step": 2500
|
| 2160 |
+
},
|
| 2161 |
+
{
|
| 2162 |
+
"epoch": 3.853387065822299,
|
| 2163 |
+
"grad_norm": 0.11872826516628265,
|
| 2164 |
+
"learning_rate": 1.513145097054977e-05,
|
| 2165 |
+
"loss": 0.0098,
|
| 2166 |
+
"step": 2510
|
| 2167 |
+
},
|
| 2168 |
+
{
|
| 2169 |
+
"epoch": 3.8687392055267704,
|
| 2170 |
+
"grad_norm": 0.14837545156478882,
|
| 2171 |
+
"learning_rate": 1.4749098712555854e-05,
|
| 2172 |
+
"loss": 0.0094,
|
| 2173 |
+
"step": 2520
|
| 2174 |
+
},
|
| 2175 |
+
{
|
| 2176 |
+
"epoch": 3.8840913452312416,
|
| 2177 |
+
"grad_norm": 0.1325119286775589,
|
| 2178 |
+
"learning_rate": 1.4370801792624656e-05,
|
| 2179 |
+
"loss": 0.0079,
|
| 2180 |
+
"step": 2530
|
| 2181 |
+
},
|
| 2182 |
+
{
|
| 2183 |
+
"epoch": 3.8994434849357127,
|
| 2184 |
+
"grad_norm": 0.35325750708580017,
|
| 2185 |
+
"learning_rate": 1.399660373083203e-05,
|
| 2186 |
+
"loss": 0.0096,
|
| 2187 |
+
"step": 2540
|
| 2188 |
+
},
|
| 2189 |
+
{
|
| 2190 |
+
"epoch": 3.9147956246401843,
|
| 2191 |
+
"grad_norm": 0.5195595622062683,
|
| 2192 |
+
"learning_rate": 1.3626547575712545e-05,
|
| 2193 |
+
"loss": 0.0154,
|
| 2194 |
+
"step": 2550
|
| 2195 |
+
},
|
| 2196 |
+
{
|
| 2197 |
+
"epoch": 3.9147956246401843,
|
| 2198 |
+
"eval_loss": 0.03803320229053497,
|
| 2199 |
+
"eval_runtime": 197.0492,
|
| 2200 |
+
"eval_samples_per_second": 5.882,
|
| 2201 |
+
"eval_steps_per_second": 5.882,
|
| 2202 |
+
"step": 2550
|
| 2203 |
+
},
|
| 2204 |
+
{
|
| 2205 |
+
"epoch": 3.9301477643446554,
|
| 2206 |
+
"grad_norm": 0.5137061476707458,
|
| 2207 |
+
"learning_rate": 1.3260675899307096e-05,
|
| 2208 |
+
"loss": 0.0167,
|
| 2209 |
+
"step": 2560
|
| 2210 |
+
},
|
| 2211 |
+
{
|
| 2212 |
+
"epoch": 3.945499904049127,
|
| 2213 |
+
"grad_norm": 0.12673397362232208,
|
| 2214 |
+
"learning_rate": 1.2899030792265349e-05,
|
| 2215 |
+
"loss": 0.0079,
|
| 2216 |
+
"step": 2570
|
| 2217 |
+
},
|
| 2218 |
+
{
|
| 2219 |
+
"epoch": 3.960852043753598,
|
| 2220 |
+
"grad_norm": 0.15755097568035126,
|
| 2221 |
+
"learning_rate": 1.2541653859003437e-05,
|
| 2222 |
+
"loss": 0.0108,
|
| 2223 |
+
"step": 2580
|
| 2224 |
+
},
|
| 2225 |
+
{
|
| 2226 |
+
"epoch": 3.9762041834580693,
|
| 2227 |
+
"grad_norm": 0.23179736733436584,
|
| 2228 |
+
"learning_rate": 1.2188586212917846e-05,
|
| 2229 |
+
"loss": 0.0108,
|
| 2230 |
+
"step": 2590
|
| 2231 |
+
},
|
| 2232 |
+
{
|
| 2233 |
+
"epoch": 3.991556323162541,
|
| 2234 |
+
"grad_norm": 0.23065230250358582,
|
| 2235 |
+
"learning_rate": 1.1839868471655523e-05,
|
| 2236 |
+
"loss": 0.0087,
|
| 2237 |
+
"step": 2600
|
| 2238 |
+
},
|
| 2239 |
+
{
|
| 2240 |
+
"epoch": 3.991556323162541,
|
| 2241 |
+
"eval_loss": 0.03791119158267975,
|
| 2242 |
+
"eval_runtime": 196.3594,
|
| 2243 |
+
"eval_samples_per_second": 5.902,
|
| 2244 |
+
"eval_steps_per_second": 5.902,
|
| 2245 |
+
"step": 2600
|
| 2246 |
+
},
|
| 2247 |
+
{
|
| 2248 |
+
"epoch": 4.006908462867012,
|
| 2249 |
+
"grad_norm": 0.07067421078681946,
|
| 2250 |
+
"learning_rate": 1.1495540752441235e-05,
|
| 2251 |
+
"loss": 0.0085,
|
| 2252 |
+
"step": 2610
|
| 2253 |
+
},
|
| 2254 |
+
{
|
| 2255 |
+
"epoch": 4.0222606025714835,
|
| 2256 |
+
"grad_norm": 0.06453356891870499,
|
| 2257 |
+
"learning_rate": 1.1155642667462318e-05,
|
| 2258 |
+
"loss": 0.0062,
|
| 2259 |
+
"step": 2620
|
| 2260 |
+
},
|
| 2261 |
+
{
|
| 2262 |
+
"epoch": 4.037612742275955,
|
| 2263 |
+
"grad_norm": 0.10147488862276077,
|
| 2264 |
+
"learning_rate": 1.082021331931164e-05,
|
| 2265 |
+
"loss": 0.005,
|
| 2266 |
+
"step": 2630
|
| 2267 |
+
},
|
| 2268 |
+
{
|
| 2269 |
+
"epoch": 4.052964881980426,
|
| 2270 |
+
"grad_norm": 0.18546722829341888,
|
| 2271 |
+
"learning_rate": 1.0489291296489152e-05,
|
| 2272 |
+
"loss": 0.0056,
|
| 2273 |
+
"step": 2640
|
| 2274 |
+
},
|
| 2275 |
+
{
|
| 2276 |
+
"epoch": 4.068317021684897,
|
| 2277 |
+
"grad_norm": 0.09176095575094223,
|
| 2278 |
+
"learning_rate": 1.0162914668962631e-05,
|
| 2279 |
+
"loss": 0.0042,
|
| 2280 |
+
"step": 2650
|
| 2281 |
+
},
|
| 2282 |
+
{
|
| 2283 |
+
"epoch": 4.068317021684897,
|
| 2284 |
+
"eval_loss": 0.039211783558130264,
|
| 2285 |
+
"eval_runtime": 196.0944,
|
| 2286 |
+
"eval_samples_per_second": 5.91,
|
| 2287 |
+
"eval_steps_per_second": 5.91,
|
| 2288 |
+
"step": 2650
|
| 2289 |
+
},
|
| 2290 |
+
{
|
| 2291 |
+
"epoch": 4.083669161389369,
|
| 2292 |
+
"grad_norm": 0.22541296482086182,
|
| 2293 |
+
"learning_rate": 9.841120983787915e-06,
|
| 2294 |
+
"loss": 0.0039,
|
| 2295 |
+
"step": 2660
|
| 2296 |
+
},
|
| 2297 |
+
{
|
| 2298 |
+
"epoch": 4.09902130109384,
|
| 2299 |
+
"grad_norm": 0.12679553031921387,
|
| 2300 |
+
"learning_rate": 9.523947260789546e-06,
|
| 2301 |
+
"loss": 0.0052,
|
| 2302 |
+
"step": 2670
|
| 2303 |
+
},
|
| 2304 |
+
{
|
| 2305 |
+
"epoch": 4.114373440798311,
|
| 2306 |
+
"grad_norm": 0.30610185861587524,
|
| 2307 |
+
"learning_rate": 9.211429988301823e-06,
|
| 2308 |
+
"loss": 0.0046,
|
| 2309 |
+
"step": 2680
|
| 2310 |
+
},
|
| 2311 |
+
{
|
| 2312 |
+
"epoch": 4.129725580502782,
|
| 2313 |
+
"grad_norm": 0.2688978910446167,
|
| 2314 |
+
"learning_rate": 8.90360511897121e-06,
|
| 2315 |
+
"loss": 0.0049,
|
| 2316 |
+
"step": 2690
|
| 2317 |
+
},
|
| 2318 |
+
{
|
| 2319 |
+
"epoch": 4.1450777202072535,
|
| 2320 |
+
"grad_norm": 0.11524350196123123,
|
| 2321 |
+
"learning_rate": 8.600508065620161e-06,
|
| 2322 |
+
"loss": 0.0029,
|
| 2323 |
+
"step": 2700
|
| 2324 |
+
},
|
| 2325 |
+
{
|
| 2326 |
+
"epoch": 4.1450777202072535,
|
| 2327 |
+
"eval_loss": 0.041070595383644104,
|
| 2328 |
+
"eval_runtime": 195.9711,
|
| 2329 |
+
"eval_samples_per_second": 5.914,
|
| 2330 |
+
"eval_steps_per_second": 5.914,
|
| 2331 |
+
"step": 2700
|
| 2332 |
+
},
|
| 2333 |
+
{
|
| 2334 |
+
"epoch": 4.1604298599117255,
|
| 2335 |
+
"grad_norm": 0.6059698462486267,
|
| 2336 |
+
"learning_rate": 8.302173697173226e-06,
|
| 2337 |
+
"loss": 0.0081,
|
| 2338 |
+
"step": 2710
|
| 2339 |
+
},
|
| 2340 |
+
{
|
| 2341 |
+
"epoch": 4.175781999616197,
|
| 2342 |
+
"grad_norm": 0.09381125122308731,
|
| 2343 |
+
"learning_rate": 8.008636334645631e-06,
|
| 2344 |
+
"loss": 0.0042,
|
| 2345 |
+
"step": 2720
|
| 2346 |
+
},
|
| 2347 |
+
{
|
| 2348 |
+
"epoch": 4.191134139320668,
|
| 2349 |
+
"grad_norm": 0.08234836161136627,
|
| 2350 |
+
"learning_rate": 7.71992974719491e-06,
|
| 2351 |
+
"loss": 0.0038,
|
| 2352 |
+
"step": 2730
|
| 2353 |
+
},
|
| 2354 |
+
{
|
| 2355 |
+
"epoch": 4.206486279025139,
|
| 2356 |
+
"grad_norm": 0.05469619482755661,
|
| 2357 |
+
"learning_rate": 7.436087148236054e-06,
|
| 2358 |
+
"loss": 0.0026,
|
| 2359 |
+
"step": 2740
|
| 2360 |
+
},
|
| 2361 |
+
{
|
| 2362 |
+
"epoch": 4.22183841872961,
|
| 2363 |
+
"grad_norm": 0.08177103847265244,
|
| 2364 |
+
"learning_rate": 7.157141191620548e-06,
|
| 2365 |
+
"loss": 0.0044,
|
| 2366 |
+
"step": 2750
|
| 2367 |
+
},
|
| 2368 |
+
{
|
| 2369 |
+
"epoch": 4.22183841872961,
|
| 2370 |
+
"eval_loss": 0.04218851029872894,
|
| 2371 |
+
"eval_runtime": 196.6612,
|
| 2372 |
+
"eval_samples_per_second": 5.893,
|
| 2373 |
+
"eval_steps_per_second": 5.893,
|
| 2374 |
+
"step": 2750
|
| 2375 |
+
},
|
| 2376 |
+
{
|
| 2377 |
+
"epoch": 4.237190558434082,
|
| 2378 |
+
"grad_norm": 0.119593545794487,
|
| 2379 |
+
"learning_rate": 6.883123967879796e-06,
|
| 2380 |
+
"loss": 0.004,
|
| 2381 |
+
"step": 2760
|
| 2382 |
+
},
|
| 2383 |
+
{
|
| 2384 |
+
"epoch": 4.252542698138553,
|
| 2385 |
+
"grad_norm": 0.1299428641796112,
|
| 2386 |
+
"learning_rate": 6.6140670005334136e-06,
|
| 2387 |
+
"loss": 0.0035,
|
| 2388 |
+
"step": 2770
|
| 2389 |
+
},
|
| 2390 |
+
{
|
| 2391 |
+
"epoch": 4.267894837843024,
|
| 2392 |
+
"grad_norm": 0.19888031482696533,
|
| 2393 |
+
"learning_rate": 6.350001242462617e-06,
|
| 2394 |
+
"loss": 0.0056,
|
| 2395 |
+
"step": 2780
|
| 2396 |
+
},
|
| 2397 |
+
{
|
| 2398 |
+
"epoch": 4.2832469775474955,
|
| 2399 |
+
"grad_norm": 0.17250359058380127,
|
| 2400 |
+
"learning_rate": 6.090957072349385e-06,
|
| 2401 |
+
"loss": 0.0042,
|
| 2402 |
+
"step": 2790
|
| 2403 |
+
},
|
| 2404 |
+
{
|
| 2405 |
+
"epoch": 4.298599117251967,
|
| 2406 |
+
"grad_norm": 0.1387476772069931,
|
| 2407 |
+
"learning_rate": 5.836964291181624e-06,
|
| 2408 |
+
"loss": 0.0035,
|
| 2409 |
+
"step": 2800
|
| 2410 |
+
},
|
| 2411 |
+
{
|
| 2412 |
+
"epoch": 4.298599117251967,
|
| 2413 |
+
"eval_loss": 0.04301900044083595,
|
| 2414 |
+
"eval_runtime": 196.9142,
|
| 2415 |
+
"eval_samples_per_second": 5.886,
|
| 2416 |
+
"eval_steps_per_second": 5.886,
|
| 2417 |
+
"step": 2800
|
| 2418 |
+
},
|
| 2419 |
+
{
|
| 2420 |
+
"epoch": 4.313951256956439,
|
| 2421 |
+
"grad_norm": 0.16032198071479797,
|
| 2422 |
+
"learning_rate": 5.588052118824804e-06,
|
| 2423 |
+
"loss": 0.0054,
|
| 2424 |
+
"step": 2810
|
| 2425 |
+
},
|
| 2426 |
+
{
|
| 2427 |
+
"epoch": 4.32930339666091,
|
| 2428 |
+
"grad_norm": 0.10156236588954926,
|
| 2429 |
+
"learning_rate": 5.344249190660428e-06,
|
| 2430 |
+
"loss": 0.0027,
|
| 2431 |
+
"step": 2820
|
| 2432 |
+
},
|
| 2433 |
+
{
|
| 2434 |
+
"epoch": 4.344655536365381,
|
| 2435 |
+
"grad_norm": 0.2252165526151657,
|
| 2436 |
+
"learning_rate": 5.105583554291765e-06,
|
| 2437 |
+
"loss": 0.0044,
|
| 2438 |
+
"step": 2830
|
| 2439 |
+
},
|
| 2440 |
+
{
|
| 2441 |
+
"epoch": 4.360007676069852,
|
| 2442 |
+
"grad_norm": 0.016164055094122887,
|
| 2443 |
+
"learning_rate": 4.872082666317207e-06,
|
| 2444 |
+
"loss": 0.0023,
|
| 2445 |
+
"step": 2840
|
| 2446 |
+
},
|
| 2447 |
+
{
|
| 2448 |
+
"epoch": 4.375359815774323,
|
| 2449 |
+
"grad_norm": 0.14848938584327698,
|
| 2450 |
+
"learning_rate": 4.6437733891715905e-06,
|
| 2451 |
+
"loss": 0.0031,
|
| 2452 |
+
"step": 2850
|
| 2453 |
+
},
|
| 2454 |
+
{
|
| 2455 |
+
"epoch": 4.375359815774323,
|
| 2456 |
+
"eval_loss": 0.044064514338970184,
|
| 2457 |
+
"eval_runtime": 197.355,
|
| 2458 |
+
"eval_samples_per_second": 5.873,
|
| 2459 |
+
"eval_steps_per_second": 5.873,
|
| 2460 |
+
"step": 2850
|
| 2461 |
+
},
|
| 2462 |
+
{
|
| 2463 |
+
"epoch": 4.390711955478795,
|
| 2464 |
+
"grad_norm": 0.0969080924987793,
|
| 2465 |
+
"learning_rate": 4.420681988035891e-06,
|
| 2466 |
+
"loss": 0.0036,
|
| 2467 |
+
"step": 2860
|
| 2468 |
+
},
|
| 2469 |
+
{
|
| 2470 |
+
"epoch": 4.406064095183266,
|
| 2471 |
+
"grad_norm": 0.1187598705291748,
|
| 2472 |
+
"learning_rate": 4.2028341278156026e-06,
|
| 2473 |
+
"loss": 0.0063,
|
| 2474 |
+
"step": 2870
|
| 2475 |
+
},
|
| 2476 |
+
{
|
| 2477 |
+
"epoch": 4.4214162348877375,
|
| 2478 |
+
"grad_norm": 0.07551722228527069,
|
| 2479 |
+
"learning_rate": 3.990254870188221e-06,
|
| 2480 |
+
"loss": 0.0055,
|
| 2481 |
+
"step": 2880
|
| 2482 |
+
},
|
| 2483 |
+
{
|
| 2484 |
+
"epoch": 4.436768374592209,
|
| 2485 |
+
"grad_norm": 0.09198061376810074,
|
| 2486 |
+
"learning_rate": 3.7829686707200827e-06,
|
| 2487 |
+
"loss": 0.0041,
|
| 2488 |
+
"step": 2890
|
| 2489 |
+
},
|
| 2490 |
+
{
|
| 2491 |
+
"epoch": 4.45212051429668,
|
| 2492 |
+
"grad_norm": 0.09851433336734772,
|
| 2493 |
+
"learning_rate": 3.580999376052946e-06,
|
| 2494 |
+
"loss": 0.004,
|
| 2495 |
+
"step": 2900
|
| 2496 |
+
},
|
| 2497 |
+
{
|
| 2498 |
+
"epoch": 4.45212051429668,
|
| 2499 |
+
"eval_loss": 0.044529687613248825,
|
| 2500 |
+
"eval_runtime": 197.1011,
|
| 2501 |
+
"eval_samples_per_second": 5.88,
|
| 2502 |
+
"eval_steps_per_second": 5.88,
|
| 2503 |
+
"step": 2900
|
| 2504 |
+
},
|
| 2505 |
+
{
|
| 2506 |
+
"epoch": 4.467472654001152,
|
| 2507 |
+
"grad_norm": 0.3292700946331024,
|
| 2508 |
+
"learning_rate": 3.3843702211606153e-06,
|
| 2509 |
+
"loss": 0.0045,
|
| 2510 |
+
"step": 2910
|
| 2511 |
+
},
|
| 2512 |
+
{
|
| 2513 |
+
"epoch": 4.482824793705623,
|
| 2514 |
+
"grad_norm": 0.028438210487365723,
|
| 2515 |
+
"learning_rate": 3.193103826675947e-06,
|
| 2516 |
+
"loss": 0.0052,
|
| 2517 |
+
"step": 2920
|
| 2518 |
+
},
|
| 2519 |
+
{
|
| 2520 |
+
"epoch": 4.498176933410094,
|
| 2521 |
+
"grad_norm": 0.2217072695493698,
|
| 2522 |
+
"learning_rate": 3.007222196288545e-06,
|
| 2523 |
+
"loss": 0.0032,
|
| 2524 |
+
"step": 2930
|
| 2525 |
+
},
|
| 2526 |
+
{
|
| 2527 |
+
"epoch": 4.513529073114565,
|
| 2528 |
+
"grad_norm": 0.06991735100746155,
|
| 2529 |
+
"learning_rate": 2.8267467142133687e-06,
|
| 2530 |
+
"loss": 0.0031,
|
| 2531 |
+
"step": 2940
|
| 2532 |
+
},
|
| 2533 |
+
{
|
| 2534 |
+
"epoch": 4.528881212819036,
|
| 2535 |
+
"grad_norm": 0.12266076356172562,
|
| 2536 |
+
"learning_rate": 2.651698142730674e-06,
|
| 2537 |
+
"loss": 0.0035,
|
| 2538 |
+
"step": 2950
|
| 2539 |
+
},
|
| 2540 |
+
{
|
| 2541 |
+
"epoch": 4.528881212819036,
|
| 2542 |
+
"eval_loss": 0.04456078261137009,
|
| 2543 |
+
"eval_runtime": 197.006,
|
| 2544 |
+
"eval_samples_per_second": 5.883,
|
| 2545 |
+
"eval_steps_per_second": 5.883,
|
| 2546 |
+
"step": 2950
|
| 2547 |
+
},
|
| 2548 |
+
{
|
| 2549 |
+
"epoch": 4.544233352523508,
|
| 2550 |
+
"grad_norm": 0.013956692069768906,
|
| 2551 |
+
"learning_rate": 2.4820966197974748e-06,
|
| 2552 |
+
"loss": 0.0036,
|
| 2553 |
+
"step": 2960
|
| 2554 |
+
},
|
| 2555 |
+
{
|
| 2556 |
+
"epoch": 4.5595854922279795,
|
| 2557 |
+
"grad_norm": 0.03814029321074486,
|
| 2558 |
+
"learning_rate": 2.3179616567308216e-06,
|
| 2559 |
+
"loss": 0.0029,
|
| 2560 |
+
"step": 2970
|
| 2561 |
+
},
|
| 2562 |
+
{
|
| 2563 |
+
"epoch": 4.574937631932451,
|
| 2564 |
+
"grad_norm": 0.02994069829583168,
|
| 2565 |
+
"learning_rate": 2.1593121359631873e-06,
|
| 2566 |
+
"loss": 0.0025,
|
| 2567 |
+
"step": 2980
|
| 2568 |
+
},
|
| 2569 |
+
{
|
| 2570 |
+
"epoch": 4.590289771636922,
|
| 2571 |
+
"grad_norm": 0.08673041313886642,
|
| 2572 |
+
"learning_rate": 2.006166308870189e-06,
|
| 2573 |
+
"loss": 0.0037,
|
| 2574 |
+
"step": 2990
|
| 2575 |
+
},
|
| 2576 |
+
{
|
| 2577 |
+
"epoch": 4.605641911341393,
|
| 2578 |
+
"grad_norm": 0.027347547933459282,
|
| 2579 |
+
"learning_rate": 1.8585417936709038e-06,
|
| 2580 |
+
"loss": 0.0021,
|
| 2581 |
+
"step": 3000
|
| 2582 |
+
},
|
| 2583 |
+
{
|
| 2584 |
+
"epoch": 4.605641911341393,
|
| 2585 |
+
"eval_loss": 0.045384086668491364,
|
| 2586 |
+
"eval_runtime": 197.3936,
|
| 2587 |
+
"eval_samples_per_second": 5.872,
|
| 2588 |
+
"eval_steps_per_second": 5.872,
|
| 2589 |
+
"step": 3000
|
| 2590 |
+
},
|
| 2591 |
+
{
|
| 2592 |
+
"epoch": 4.620994051045864,
|
| 2593 |
+
"grad_norm": 0.10114165395498276,
|
| 2594 |
+
"learning_rate": 1.7164555734010545e-06,
|
| 2595 |
+
"loss": 0.0085,
|
| 2596 |
+
"step": 3010
|
| 2597 |
+
},
|
| 2598 |
+
{
|
| 2599 |
+
"epoch": 4.636346190750336,
|
| 2600 |
+
"grad_norm": 0.1202889010310173,
|
| 2601 |
+
"learning_rate": 1.5799239939592204e-06,
|
| 2602 |
+
"loss": 0.0051,
|
| 2603 |
+
"step": 3020
|
| 2604 |
+
},
|
| 2605 |
+
{
|
| 2606 |
+
"epoch": 4.651698330454807,
|
| 2607 |
+
"grad_norm": 0.07459268718957901,
|
| 2608 |
+
"learning_rate": 1.4489627622263747e-06,
|
| 2609 |
+
"loss": 0.0052,
|
| 2610 |
+
"step": 3030
|
| 2611 |
+
},
|
| 2612 |
+
{
|
| 2613 |
+
"epoch": 4.667050470159278,
|
| 2614 |
+
"grad_norm": 0.22482830286026,
|
| 2615 |
+
"learning_rate": 1.3235869442589255e-06,
|
| 2616 |
+
"loss": 0.0086,
|
| 2617 |
+
"step": 3040
|
| 2618 |
+
},
|
| 2619 |
+
{
|
| 2620 |
+
"epoch": 4.6824026098637495,
|
| 2621 |
+
"grad_norm": 0.17435556650161743,
|
| 2622 |
+
"learning_rate": 1.2038109635555406e-06,
|
| 2623 |
+
"loss": 0.0041,
|
| 2624 |
+
"step": 3050
|
| 2625 |
+
},
|
| 2626 |
+
{
|
| 2627 |
+
"epoch": 4.6824026098637495,
|
| 2628 |
+
"eval_loss": 0.04585828632116318,
|
| 2629 |
+
"eval_runtime": 197.2718,
|
| 2630 |
+
"eval_samples_per_second": 5.875,
|
| 2631 |
+
"eval_steps_per_second": 5.875,
|
| 2632 |
+
"step": 3050
|
| 2633 |
+
},
|
| 2634 |
+
{
|
| 2635 |
+
"epoch": 4.697754749568221,
|
| 2636 |
+
"grad_norm": 0.034258339554071426,
|
| 2637 |
+
"learning_rate": 1.0896485993977467e-06,
|
| 2638 |
+
"loss": 0.0047,
|
| 2639 |
+
"step": 3060
|
| 2640 |
+
},
|
| 2641 |
+
{
|
| 2642 |
+
"epoch": 4.713106889272693,
|
| 2643 |
+
"grad_norm": 0.058088961988687515,
|
| 2644 |
+
"learning_rate": 9.811129852647982e-07,
|
| 2645 |
+
"loss": 0.0022,
|
| 2646 |
+
"step": 3070
|
| 2647 |
+
},
|
| 2648 |
+
{
|
| 2649 |
+
"epoch": 4.728459028977164,
|
| 2650 |
+
"grad_norm": 0.37871047854423523,
|
| 2651 |
+
"learning_rate": 8.782166073227515e-07,
|
| 2652 |
+
"loss": 0.0066,
|
| 2653 |
+
"step": 3080
|
| 2654 |
+
},
|
| 2655 |
+
{
|
| 2656 |
+
"epoch": 4.743811168681635,
|
| 2657 |
+
"grad_norm": 0.023409299552440643,
|
| 2658 |
+
"learning_rate": 7.809713029880428e-07,
|
| 2659 |
+
"loss": 0.0038,
|
| 2660 |
+
"step": 3090
|
| 2661 |
+
},
|
| 2662 |
+
{
|
| 2663 |
+
"epoch": 4.759163308386106,
|
| 2664 |
+
"grad_norm": 0.10357584804296494,
|
| 2665 |
+
"learning_rate": 6.893882595656598e-07,
|
| 2666 |
+
"loss": 0.006,
|
| 2667 |
+
"step": 3100
|
| 2668 |
+
},
|
| 2669 |
+
{
|
| 2670 |
+
"epoch": 4.759163308386106,
|
| 2671 |
+
"eval_loss": 0.04564342275261879,
|
| 2672 |
+
"eval_runtime": 197.4178,
|
| 2673 |
+
"eval_samples_per_second": 5.871,
|
| 2674 |
+
"eval_steps_per_second": 5.871,
|
| 2675 |
+
"step": 3100
|
| 2676 |
+
},
|
| 2677 |
+
{
|
| 2678 |
+
"epoch": 4.774515448090577,
|
| 2679 |
+
"grad_norm": 0.20756296813488007,
|
| 2680 |
+
"learning_rate": 6.034780129621664e-07,
|
| 2681 |
+
"loss": 0.0067,
|
| 2682 |
+
"step": 3110
|
| 2683 |
+
},
|
| 2684 |
+
{
|
| 2685 |
+
"epoch": 4.789867587795049,
|
| 2686 |
+
"grad_norm": 0.32027533650398254,
|
| 2687 |
+
"learning_rate": 5.232504464735833e-07,
|
| 2688 |
+
"loss": 0.0039,
|
| 2689 |
+
"step": 3120
|
| 2690 |
+
},
|
| 2691 |
+
{
|
| 2692 |
+
"epoch": 4.80521972749952,
|
| 2693 |
+
"grad_norm": 0.3811037838459015,
|
| 2694 |
+
"learning_rate": 4.487147896484523e-07,
|
| 2695 |
+
"loss": 0.007,
|
| 2696 |
+
"step": 3130
|
| 2697 |
+
},
|
| 2698 |
+
{
|
| 2699 |
+
"epoch": 4.8205718672039914,
|
| 2700 |
+
"grad_norm": 0.0528700053691864,
|
| 2701 |
+
"learning_rate": 3.7987961722599773e-07,
|
| 2702 |
+
"loss": 0.0063,
|
| 2703 |
+
"step": 3140
|
| 2704 |
+
},
|
| 2705 |
+
{
|
| 2706 |
+
"epoch": 4.835924006908463,
|
| 2707 |
+
"grad_norm": 0.03822799026966095,
|
| 2708 |
+
"learning_rate": 3.167528481496984e-07,
|
| 2709 |
+
"loss": 0.0043,
|
| 2710 |
+
"step": 3150
|
| 2711 |
+
},
|
| 2712 |
+
{
|
| 2713 |
+
"epoch": 4.835924006908463,
|
| 2714 |
+
"eval_loss": 0.04553144425153732,
|
| 2715 |
+
"eval_runtime": 197.6602,
|
| 2716 |
+
"eval_samples_per_second": 5.864,
|
| 2717 |
+
"eval_steps_per_second": 5.864,
|
| 2718 |
+
"step": 3150
|
| 2719 |
+
},
|
| 2720 |
+
{
|
| 2721 |
+
"epoch": 4.851276146612934,
|
| 2722 |
+
"grad_norm": 0.2740378975868225,
|
| 2723 |
+
"learning_rate": 2.593417446562607e-07,
|
| 2724 |
+
"loss": 0.0083,
|
| 2725 |
+
"step": 3160
|
| 2726 |
+
},
|
| 2727 |
+
{
|
| 2728 |
+
"epoch": 4.866628286317406,
|
| 2729 |
+
"grad_norm": 0.2202148288488388,
|
| 2730 |
+
"learning_rate": 2.0765291144016486e-07,
|
| 2731 |
+
"loss": 0.0043,
|
| 2732 |
+
"step": 3170
|
| 2733 |
+
},
|
| 2734 |
+
{
|
| 2735 |
+
"epoch": 4.881980426021877,
|
| 2736 |
+
"grad_norm": 0.12892213463783264,
|
| 2737 |
+
"learning_rate": 1.6169229489385595e-07,
|
| 2738 |
+
"loss": 0.0042,
|
| 2739 |
+
"step": 3180
|
| 2740 |
+
},
|
| 2741 |
+
{
|
| 2742 |
+
"epoch": 4.897332565726348,
|
| 2743 |
+
"grad_norm": 0.006994785740971565,
|
| 2744 |
+
"learning_rate": 1.2146518242363014e-07,
|
| 2745 |
+
"loss": 0.0027,
|
| 2746 |
+
"step": 3190
|
| 2747 |
+
},
|
| 2748 |
+
{
|
| 2749 |
+
"epoch": 4.912684705430819,
|
| 2750 |
+
"grad_norm": 0.1142883449792862,
|
| 2751 |
+
"learning_rate": 8.697620184138222e-08,
|
| 2752 |
+
"loss": 0.0031,
|
| 2753 |
+
"step": 3200
|
| 2754 |
+
},
|
| 2755 |
+
{
|
| 2756 |
+
"epoch": 4.912684705430819,
|
| 2757 |
+
"eval_loss": 0.04559043049812317,
|
| 2758 |
+
"eval_runtime": 197.5618,
|
| 2759 |
+
"eval_samples_per_second": 5.867,
|
| 2760 |
+
"eval_steps_per_second": 5.867,
|
| 2761 |
+
"step": 3200
|
| 2762 |
+
},
|
| 2763 |
+
{
|
| 2764 |
+
"epoch": 4.92803684513529,
|
| 2765 |
+
"grad_norm": 0.10935990512371063,
|
| 2766 |
+
"learning_rate": 5.822932083221488e-08,
|
| 2767 |
+
"loss": 0.0054,
|
| 2768 |
+
"step": 3210
|
| 2769 |
+
},
|
| 2770 |
+
{
|
| 2771 |
+
"epoch": 4.943388984839762,
|
| 2772 |
+
"grad_norm": 0.2467465102672577,
|
| 2773 |
+
"learning_rate": 3.5227846497970504e-08,
|
| 2774 |
+
"loss": 0.0037,
|
| 2775 |
+
"step": 3220
|
| 2776 |
+
},
|
| 2777 |
+
{
|
| 2778 |
+
"epoch": 4.958741124544233,
|
| 2779 |
+
"grad_norm": 0.15151652693748474,
|
| 2780 |
+
"learning_rate": 1.7974424976796577e-08,
|
| 2781 |
+
"loss": 0.0031,
|
| 2782 |
+
"step": 3230
|
| 2783 |
+
},
|
| 2784 |
+
{
|
| 2785 |
+
"epoch": 4.974093264248705,
|
| 2786 |
+
"grad_norm": 0.09973540902137756,
|
| 2787 |
+
"learning_rate": 6.47104113870034e-09,
|
| 2788 |
+
"loss": 0.0045,
|
| 2789 |
+
"step": 3240
|
| 2790 |
+
},
|
| 2791 |
+
{
|
| 2792 |
+
"epoch": 4.989445403953176,
|
| 2793 |
+
"grad_norm": 0.1923939734697342,
|
| 2794 |
+
"learning_rate": 7.190183572314269e-10,
|
| 2795 |
+
"loss": 0.0073,
|
| 2796 |
+
"step": 3250
|
| 2797 |
+
},
|
| 2798 |
+
{
|
| 2799 |
+
"epoch": 4.989445403953176,
|
| 2800 |
+
"eval_loss": 0.04562755674123764,
|
| 2801 |
+
"eval_runtime": 197.9034,
|
| 2802 |
+
"eval_samples_per_second": 5.856,
|
| 2803 |
+
"eval_steps_per_second": 5.856,
|
| 2804 |
+
"step": 3250
|
| 2805 |
+
},
|
| 2806 |
+
{
|
| 2807 |
+
"epoch": 4.997121473805412,
|
| 2808 |
+
"step": 3255,
|
| 2809 |
+
"total_flos": 8.80470547326763e+17,
|
| 2810 |
+
"train_loss": 0.04897872828713943,
|
| 2811 |
+
"train_runtime": 41700.3864,
|
| 2812 |
+
"train_samples_per_second": 1.25,
|
| 2813 |
+
"train_steps_per_second": 0.078
|
| 2814 |
+
}
|
| 2815 |
+
],
|
| 2816 |
+
"logging_steps": 10,
|
| 2817 |
+
"max_steps": 3255,
|
| 2818 |
+
"num_input_tokens_seen": 0,
|
| 2819 |
+
"num_train_epochs": 5,
|
| 2820 |
+
"save_steps": 50,
|
| 2821 |
+
"stateful_callbacks": {
|
| 2822 |
+
"TrainerControl": {
|
| 2823 |
+
"args": {
|
| 2824 |
+
"should_epoch_stop": false,
|
| 2825 |
+
"should_evaluate": false,
|
| 2826 |
+
"should_log": false,
|
| 2827 |
+
"should_save": true,
|
| 2828 |
+
"should_training_stop": true
|
| 2829 |
+
},
|
| 2830 |
+
"attributes": {}
|
| 2831 |
+
}
|
| 2832 |
+
},
|
| 2833 |
+
"total_flos": 8.80470547326763e+17,
|
| 2834 |
+
"train_batch_size": 1,
|
| 2835 |
+
"trial_name": null,
|
| 2836 |
+
"trial_params": null
|
| 2837 |
+
}
|
training_eval_loss.png
ADDED
|
training_loss.png
ADDED
|