End of training
Browse files- README.md +2 -2
- all_results.json +13 -0
- eval_results.json +8 -0
- train_results.json +8 -0
- trainer_state.json +1170 -0
README.md
CHANGED
|
@@ -4,7 +4,7 @@ base_model: google/long-t5-tglobal-xl
|
|
| 4 |
tags:
|
| 5 |
- generated_from_trainer
|
| 6 |
datasets:
|
| 7 |
-
- scrolls
|
| 8 |
model-index:
|
| 9 |
- name: longt5_xl_sfd_20
|
| 10 |
results: []
|
|
@@ -15,7 +15,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 15 |
|
| 16 |
# longt5_xl_sfd_20
|
| 17 |
|
| 18 |
-
This model is a fine-tuned version of [google/long-t5-tglobal-xl](https://huggingface.co/google/long-t5-tglobal-xl) on the scrolls dataset.
|
| 19 |
It achieves the following results on the evaluation set:
|
| 20 |
- Loss: 4.8167
|
| 21 |
|
|
|
|
| 4 |
tags:
|
| 5 |
- generated_from_trainer
|
| 6 |
datasets:
|
| 7 |
+
- tau/scrolls
|
| 8 |
model-index:
|
| 9 |
- name: longt5_xl_sfd_20
|
| 10 |
results: []
|
|
|
|
| 15 |
|
| 16 |
# longt5_xl_sfd_20
|
| 17 |
|
| 18 |
+
This model is a fine-tuned version of [google/long-t5-tglobal-xl](https://huggingface.co/google/long-t5-tglobal-xl) on the tau/scrolls summ_screen_fd dataset.
|
| 19 |
It achieves the following results on the evaluation set:
|
| 20 |
- Loss: 4.8167
|
| 21 |
|
all_results.json
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 19.48,
|
| 3 |
+
"eval_loss": 4.81671667098999,
|
| 4 |
+
"eval_runtime": 80.254,
|
| 5 |
+
"eval_samples": 338,
|
| 6 |
+
"eval_samples_per_second": 4.212,
|
| 7 |
+
"eval_steps_per_second": 0.536,
|
| 8 |
+
"train_loss": 0.8494854368801628,
|
| 9 |
+
"train_runtime": 68771.7044,
|
| 10 |
+
"train_samples": 3673,
|
| 11 |
+
"train_samples_per_second": 1.068,
|
| 12 |
+
"train_steps_per_second": 0.004
|
| 13 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 19.48,
|
| 3 |
+
"eval_loss": 4.81671667098999,
|
| 4 |
+
"eval_runtime": 80.254,
|
| 5 |
+
"eval_samples": 338,
|
| 6 |
+
"eval_samples_per_second": 4.212,
|
| 7 |
+
"eval_steps_per_second": 0.536
|
| 8 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 19.48,
|
| 3 |
+
"train_loss": 0.8494854368801628,
|
| 4 |
+
"train_runtime": 68771.7044,
|
| 5 |
+
"train_samples": 3673,
|
| 6 |
+
"train_samples_per_second": 1.068,
|
| 7 |
+
"train_steps_per_second": 0.004
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1170 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 2.2994935512542725,
|
| 3 |
+
"best_model_checkpoint": "/exports/eddie/scratch/s1970716/models/longt5_xl_sfd_20/checkpoint-28",
|
| 4 |
+
"epoch": 19.47826086956522,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 280,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.14,
|
| 13 |
+
"grad_norm": 8.068708419799805,
|
| 14 |
+
"learning_rate": 0.001,
|
| 15 |
+
"loss": 3.274,
|
| 16 |
+
"step": 2
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.28,
|
| 20 |
+
"grad_norm": 1.4994572401046753,
|
| 21 |
+
"learning_rate": 0.001,
|
| 22 |
+
"loss": 3.2963,
|
| 23 |
+
"step": 4
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.42,
|
| 27 |
+
"grad_norm": 1.0570803880691528,
|
| 28 |
+
"learning_rate": 0.001,
|
| 29 |
+
"loss": 3.3164,
|
| 30 |
+
"step": 6
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.56,
|
| 34 |
+
"grad_norm": 1.2446849346160889,
|
| 35 |
+
"learning_rate": 0.001,
|
| 36 |
+
"loss": 3.0866,
|
| 37 |
+
"step": 8
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.7,
|
| 41 |
+
"grad_norm": 0.721084713935852,
|
| 42 |
+
"learning_rate": 0.001,
|
| 43 |
+
"loss": 2.8976,
|
| 44 |
+
"step": 10
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.83,
|
| 48 |
+
"grad_norm": 1.2132383584976196,
|
| 49 |
+
"learning_rate": 0.001,
|
| 50 |
+
"loss": 2.8298,
|
| 51 |
+
"step": 12
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.97,
|
| 55 |
+
"grad_norm": 0.4689762592315674,
|
| 56 |
+
"learning_rate": 0.001,
|
| 57 |
+
"loss": 2.9377,
|
| 58 |
+
"step": 14
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.97,
|
| 62 |
+
"eval_loss": 2.7965147495269775,
|
| 63 |
+
"eval_runtime": 81.4763,
|
| 64 |
+
"eval_samples_per_second": 4.148,
|
| 65 |
+
"eval_steps_per_second": 0.528,
|
| 66 |
+
"step": 14
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 1.11,
|
| 70 |
+
"grad_norm": 0.42892181873321533,
|
| 71 |
+
"learning_rate": 0.001,
|
| 72 |
+
"loss": 2.741,
|
| 73 |
+
"step": 16
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 1.25,
|
| 77 |
+
"grad_norm": 0.4487678110599518,
|
| 78 |
+
"learning_rate": 0.001,
|
| 79 |
+
"loss": 2.4441,
|
| 80 |
+
"step": 18
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 1.39,
|
| 84 |
+
"grad_norm": 0.4653552770614624,
|
| 85 |
+
"learning_rate": 0.001,
|
| 86 |
+
"loss": 2.432,
|
| 87 |
+
"step": 20
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 1.53,
|
| 91 |
+
"grad_norm": 0.35275548696517944,
|
| 92 |
+
"learning_rate": 0.001,
|
| 93 |
+
"loss": 2.4016,
|
| 94 |
+
"step": 22
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 1.67,
|
| 98 |
+
"grad_norm": 0.43277695775032043,
|
| 99 |
+
"learning_rate": 0.001,
|
| 100 |
+
"loss": 2.391,
|
| 101 |
+
"step": 24
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 1.81,
|
| 105 |
+
"grad_norm": 0.3408297300338745,
|
| 106 |
+
"learning_rate": 0.001,
|
| 107 |
+
"loss": 2.3911,
|
| 108 |
+
"step": 26
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 1.95,
|
| 112 |
+
"grad_norm": 0.3205319344997406,
|
| 113 |
+
"learning_rate": 0.001,
|
| 114 |
+
"loss": 2.3247,
|
| 115 |
+
"step": 28
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 1.95,
|
| 119 |
+
"eval_loss": 2.2994935512542725,
|
| 120 |
+
"eval_runtime": 81.4693,
|
| 121 |
+
"eval_samples_per_second": 4.149,
|
| 122 |
+
"eval_steps_per_second": 0.528,
|
| 123 |
+
"step": 28
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"epoch": 2.09,
|
| 127 |
+
"grad_norm": 0.4033512771129608,
|
| 128 |
+
"learning_rate": 0.001,
|
| 129 |
+
"loss": 2.0701,
|
| 130 |
+
"step": 30
|
| 131 |
+
},
|
| 132 |
+
{
|
| 133 |
+
"epoch": 2.23,
|
| 134 |
+
"grad_norm": 0.36825311183929443,
|
| 135 |
+
"learning_rate": 0.001,
|
| 136 |
+
"loss": 2.0968,
|
| 137 |
+
"step": 32
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 2.37,
|
| 141 |
+
"grad_norm": 0.5080482363700867,
|
| 142 |
+
"learning_rate": 0.001,
|
| 143 |
+
"loss": 2.0681,
|
| 144 |
+
"step": 34
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 2.5,
|
| 148 |
+
"grad_norm": 0.4196927845478058,
|
| 149 |
+
"learning_rate": 0.001,
|
| 150 |
+
"loss": 2.0914,
|
| 151 |
+
"step": 36
|
| 152 |
+
},
|
| 153 |
+
{
|
| 154 |
+
"epoch": 2.64,
|
| 155 |
+
"grad_norm": 0.3230506479740143,
|
| 156 |
+
"learning_rate": 0.001,
|
| 157 |
+
"loss": 2.0317,
|
| 158 |
+
"step": 38
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 2.78,
|
| 162 |
+
"grad_norm": 0.2733004689216614,
|
| 163 |
+
"learning_rate": 0.001,
|
| 164 |
+
"loss": 1.9723,
|
| 165 |
+
"step": 40
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 2.92,
|
| 169 |
+
"grad_norm": 0.2709517776966095,
|
| 170 |
+
"learning_rate": 0.001,
|
| 171 |
+
"loss": 1.9943,
|
| 172 |
+
"step": 42
|
| 173 |
+
},
|
| 174 |
+
{
|
| 175 |
+
"epoch": 2.99,
|
| 176 |
+
"eval_loss": 2.3308048248291016,
|
| 177 |
+
"eval_runtime": 81.5083,
|
| 178 |
+
"eval_samples_per_second": 4.147,
|
| 179 |
+
"eval_steps_per_second": 0.528,
|
| 180 |
+
"step": 43
|
| 181 |
+
},
|
| 182 |
+
{
|
| 183 |
+
"epoch": 3.06,
|
| 184 |
+
"grad_norm": 0.3230663537979126,
|
| 185 |
+
"learning_rate": 0.001,
|
| 186 |
+
"loss": 1.9093,
|
| 187 |
+
"step": 44
|
| 188 |
+
},
|
| 189 |
+
{
|
| 190 |
+
"epoch": 3.2,
|
| 191 |
+
"grad_norm": 0.3976946175098419,
|
| 192 |
+
"learning_rate": 0.001,
|
| 193 |
+
"loss": 1.7682,
|
| 194 |
+
"step": 46
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 3.34,
|
| 198 |
+
"grad_norm": 0.42008209228515625,
|
| 199 |
+
"learning_rate": 0.001,
|
| 200 |
+
"loss": 1.7119,
|
| 201 |
+
"step": 48
|
| 202 |
+
},
|
| 203 |
+
{
|
| 204 |
+
"epoch": 3.48,
|
| 205 |
+
"grad_norm": 0.31828513741493225,
|
| 206 |
+
"learning_rate": 0.001,
|
| 207 |
+
"loss": 1.7283,
|
| 208 |
+
"step": 50
|
| 209 |
+
},
|
| 210 |
+
{
|
| 211 |
+
"epoch": 3.62,
|
| 212 |
+
"grad_norm": 0.2448839396238327,
|
| 213 |
+
"learning_rate": 0.001,
|
| 214 |
+
"loss": 1.6905,
|
| 215 |
+
"step": 52
|
| 216 |
+
},
|
| 217 |
+
{
|
| 218 |
+
"epoch": 3.76,
|
| 219 |
+
"grad_norm": 0.25552132725715637,
|
| 220 |
+
"learning_rate": 0.001,
|
| 221 |
+
"loss": 1.6645,
|
| 222 |
+
"step": 54
|
| 223 |
+
},
|
| 224 |
+
{
|
| 225 |
+
"epoch": 3.9,
|
| 226 |
+
"grad_norm": 15.679224014282227,
|
| 227 |
+
"learning_rate": 0.001,
|
| 228 |
+
"loss": 1.7056,
|
| 229 |
+
"step": 56
|
| 230 |
+
},
|
| 231 |
+
{
|
| 232 |
+
"epoch": 3.97,
|
| 233 |
+
"eval_loss": 2.3368992805480957,
|
| 234 |
+
"eval_runtime": 81.4742,
|
| 235 |
+
"eval_samples_per_second": 4.149,
|
| 236 |
+
"eval_steps_per_second": 0.528,
|
| 237 |
+
"step": 57
|
| 238 |
+
},
|
| 239 |
+
{
|
| 240 |
+
"epoch": 4.03,
|
| 241 |
+
"grad_norm": 0.29547178745269775,
|
| 242 |
+
"learning_rate": 0.001,
|
| 243 |
+
"loss": 1.564,
|
| 244 |
+
"step": 58
|
| 245 |
+
},
|
| 246 |
+
{
|
| 247 |
+
"epoch": 4.17,
|
| 248 |
+
"grad_norm": 0.31610924005508423,
|
| 249 |
+
"learning_rate": 0.001,
|
| 250 |
+
"loss": 1.3607,
|
| 251 |
+
"step": 60
|
| 252 |
+
},
|
| 253 |
+
{
|
| 254 |
+
"epoch": 4.31,
|
| 255 |
+
"grad_norm": 0.32351407408714294,
|
| 256 |
+
"learning_rate": 0.001,
|
| 257 |
+
"loss": 1.4158,
|
| 258 |
+
"step": 62
|
| 259 |
+
},
|
| 260 |
+
{
|
| 261 |
+
"epoch": 4.45,
|
| 262 |
+
"grad_norm": 0.5101042985916138,
|
| 263 |
+
"learning_rate": 0.001,
|
| 264 |
+
"loss": 1.4694,
|
| 265 |
+
"step": 64
|
| 266 |
+
},
|
| 267 |
+
{
|
| 268 |
+
"epoch": 4.59,
|
| 269 |
+
"grad_norm": 0.41575145721435547,
|
| 270 |
+
"learning_rate": 0.001,
|
| 271 |
+
"loss": 1.4755,
|
| 272 |
+
"step": 66
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 4.73,
|
| 276 |
+
"grad_norm": 0.3269899785518646,
|
| 277 |
+
"learning_rate": 0.001,
|
| 278 |
+
"loss": 1.4268,
|
| 279 |
+
"step": 68
|
| 280 |
+
},
|
| 281 |
+
{
|
| 282 |
+
"epoch": 4.87,
|
| 283 |
+
"grad_norm": 0.4077276587486267,
|
| 284 |
+
"learning_rate": 0.001,
|
| 285 |
+
"loss": 1.4471,
|
| 286 |
+
"step": 70
|
| 287 |
+
},
|
| 288 |
+
{
|
| 289 |
+
"epoch": 4.94,
|
| 290 |
+
"eval_loss": 2.553175926208496,
|
| 291 |
+
"eval_runtime": 81.5149,
|
| 292 |
+
"eval_samples_per_second": 4.146,
|
| 293 |
+
"eval_steps_per_second": 0.528,
|
| 294 |
+
"step": 71
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"epoch": 5.01,
|
| 298 |
+
"grad_norm": 0.37493908405303955,
|
| 299 |
+
"learning_rate": 0.001,
|
| 300 |
+
"loss": 1.4436,
|
| 301 |
+
"step": 72
|
| 302 |
+
},
|
| 303 |
+
{
|
| 304 |
+
"epoch": 5.15,
|
| 305 |
+
"grad_norm": 0.8398223519325256,
|
| 306 |
+
"learning_rate": 0.001,
|
| 307 |
+
"loss": 1.1776,
|
| 308 |
+
"step": 74
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 5.29,
|
| 312 |
+
"grad_norm": 0.621316134929657,
|
| 313 |
+
"learning_rate": 0.001,
|
| 314 |
+
"loss": 1.192,
|
| 315 |
+
"step": 76
|
| 316 |
+
},
|
| 317 |
+
{
|
| 318 |
+
"epoch": 5.43,
|
| 319 |
+
"grad_norm": 0.5988876819610596,
|
| 320 |
+
"learning_rate": 0.001,
|
| 321 |
+
"loss": 1.1561,
|
| 322 |
+
"step": 78
|
| 323 |
+
},
|
| 324 |
+
{
|
| 325 |
+
"epoch": 5.57,
|
| 326 |
+
"grad_norm": 0.561390221118927,
|
| 327 |
+
"learning_rate": 0.001,
|
| 328 |
+
"loss": 1.2129,
|
| 329 |
+
"step": 80
|
| 330 |
+
},
|
| 331 |
+
{
|
| 332 |
+
"epoch": 5.7,
|
| 333 |
+
"grad_norm": 0.32573097944259644,
|
| 334 |
+
"learning_rate": 0.001,
|
| 335 |
+
"loss": 1.19,
|
| 336 |
+
"step": 82
|
| 337 |
+
},
|
| 338 |
+
{
|
| 339 |
+
"epoch": 5.84,
|
| 340 |
+
"grad_norm": 0.3272527754306793,
|
| 341 |
+
"learning_rate": 0.001,
|
| 342 |
+
"loss": 1.1933,
|
| 343 |
+
"step": 84
|
| 344 |
+
},
|
| 345 |
+
{
|
| 346 |
+
"epoch": 5.98,
|
| 347 |
+
"grad_norm": 0.36107558012008667,
|
| 348 |
+
"learning_rate": 0.001,
|
| 349 |
+
"loss": 1.1932,
|
| 350 |
+
"step": 86
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 5.98,
|
| 354 |
+
"eval_loss": 2.696089744567871,
|
| 355 |
+
"eval_runtime": 81.5294,
|
| 356 |
+
"eval_samples_per_second": 4.146,
|
| 357 |
+
"eval_steps_per_second": 0.527,
|
| 358 |
+
"step": 86
|
| 359 |
+
},
|
| 360 |
+
{
|
| 361 |
+
"epoch": 6.12,
|
| 362 |
+
"grad_norm": 0.4167131781578064,
|
| 363 |
+
"learning_rate": 0.001,
|
| 364 |
+
"loss": 0.9285,
|
| 365 |
+
"step": 88
|
| 366 |
+
},
|
| 367 |
+
{
|
| 368 |
+
"epoch": 6.26,
|
| 369 |
+
"grad_norm": 0.38736867904663086,
|
| 370 |
+
"learning_rate": 0.001,
|
| 371 |
+
"loss": 0.9568,
|
| 372 |
+
"step": 90
|
| 373 |
+
},
|
| 374 |
+
{
|
| 375 |
+
"epoch": 6.4,
|
| 376 |
+
"grad_norm": 0.3212537169456482,
|
| 377 |
+
"learning_rate": 0.001,
|
| 378 |
+
"loss": 0.9538,
|
| 379 |
+
"step": 92
|
| 380 |
+
},
|
| 381 |
+
{
|
| 382 |
+
"epoch": 6.54,
|
| 383 |
+
"grad_norm": 0.2966512143611908,
|
| 384 |
+
"learning_rate": 0.001,
|
| 385 |
+
"loss": 0.9133,
|
| 386 |
+
"step": 94
|
| 387 |
+
},
|
| 388 |
+
{
|
| 389 |
+
"epoch": 6.68,
|
| 390 |
+
"grad_norm": 0.3149372935295105,
|
| 391 |
+
"learning_rate": 0.001,
|
| 392 |
+
"loss": 0.9374,
|
| 393 |
+
"step": 96
|
| 394 |
+
},
|
| 395 |
+
{
|
| 396 |
+
"epoch": 6.82,
|
| 397 |
+
"grad_norm": 0.3140605092048645,
|
| 398 |
+
"learning_rate": 0.001,
|
| 399 |
+
"loss": 0.9585,
|
| 400 |
+
"step": 98
|
| 401 |
+
},
|
| 402 |
+
{
|
| 403 |
+
"epoch": 6.96,
|
| 404 |
+
"grad_norm": 0.33559679985046387,
|
| 405 |
+
"learning_rate": 0.001,
|
| 406 |
+
"loss": 0.9199,
|
| 407 |
+
"step": 100
|
| 408 |
+
},
|
| 409 |
+
{
|
| 410 |
+
"epoch": 6.96,
|
| 411 |
+
"eval_loss": 2.645321846008301,
|
| 412 |
+
"eval_runtime": 81.5044,
|
| 413 |
+
"eval_samples_per_second": 4.147,
|
| 414 |
+
"eval_steps_per_second": 0.528,
|
| 415 |
+
"step": 100
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 7.1,
|
| 419 |
+
"grad_norm": 0.3616858720779419,
|
| 420 |
+
"learning_rate": 0.001,
|
| 421 |
+
"loss": 0.7517,
|
| 422 |
+
"step": 102
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 7.23,
|
| 426 |
+
"grad_norm": 0.4970415234565735,
|
| 427 |
+
"learning_rate": 0.001,
|
| 428 |
+
"loss": 0.7378,
|
| 429 |
+
"step": 104
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 7.37,
|
| 433 |
+
"grad_norm": 0.6654688119888306,
|
| 434 |
+
"learning_rate": 0.001,
|
| 435 |
+
"loss": 0.7864,
|
| 436 |
+
"step": 106
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 7.51,
|
| 440 |
+
"grad_norm": 0.51229327917099,
|
| 441 |
+
"learning_rate": 0.001,
|
| 442 |
+
"loss": 0.762,
|
| 443 |
+
"step": 108
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 7.65,
|
| 447 |
+
"grad_norm": 0.4524416923522949,
|
| 448 |
+
"learning_rate": 0.001,
|
| 449 |
+
"loss": 0.7342,
|
| 450 |
+
"step": 110
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 7.79,
|
| 454 |
+
"grad_norm": 0.48206427693367004,
|
| 455 |
+
"learning_rate": 0.001,
|
| 456 |
+
"loss": 0.7706,
|
| 457 |
+
"step": 112
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 7.93,
|
| 461 |
+
"grad_norm": 0.4534417688846588,
|
| 462 |
+
"learning_rate": 0.001,
|
| 463 |
+
"loss": 0.7571,
|
| 464 |
+
"step": 114
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 8.0,
|
| 468 |
+
"eval_loss": 3.0977730751037598,
|
| 469 |
+
"eval_runtime": 81.5778,
|
| 470 |
+
"eval_samples_per_second": 4.143,
|
| 471 |
+
"eval_steps_per_second": 0.527,
|
| 472 |
+
"step": 115
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 8.07,
|
| 476 |
+
"grad_norm": 0.306815505027771,
|
| 477 |
+
"learning_rate": 0.001,
|
| 478 |
+
"loss": 0.6809,
|
| 479 |
+
"step": 116
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 8.21,
|
| 483 |
+
"grad_norm": 0.34183812141418457,
|
| 484 |
+
"learning_rate": 0.001,
|
| 485 |
+
"loss": 0.5853,
|
| 486 |
+
"step": 118
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 8.35,
|
| 490 |
+
"grad_norm": 0.3781261444091797,
|
| 491 |
+
"learning_rate": 0.001,
|
| 492 |
+
"loss": 0.5819,
|
| 493 |
+
"step": 120
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 8.49,
|
| 497 |
+
"grad_norm": 0.36344149708747864,
|
| 498 |
+
"learning_rate": 0.001,
|
| 499 |
+
"loss": 0.6059,
|
| 500 |
+
"step": 122
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 8.63,
|
| 504 |
+
"grad_norm": 0.38990476727485657,
|
| 505 |
+
"learning_rate": 0.001,
|
| 506 |
+
"loss": 0.5929,
|
| 507 |
+
"step": 124
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 8.77,
|
| 511 |
+
"grad_norm": 0.34000781178474426,
|
| 512 |
+
"learning_rate": 0.001,
|
| 513 |
+
"loss": 0.5887,
|
| 514 |
+
"step": 126
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 8.9,
|
| 518 |
+
"grad_norm": 0.32895970344543457,
|
| 519 |
+
"learning_rate": 0.001,
|
| 520 |
+
"loss": 0.6287,
|
| 521 |
+
"step": 128
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 8.97,
|
| 525 |
+
"eval_loss": 3.145782709121704,
|
| 526 |
+
"eval_runtime": 81.5735,
|
| 527 |
+
"eval_samples_per_second": 4.144,
|
| 528 |
+
"eval_steps_per_second": 0.527,
|
| 529 |
+
"step": 129
|
| 530 |
+
},
|
| 531 |
+
{
|
| 532 |
+
"epoch": 9.04,
|
| 533 |
+
"grad_norm": 0.36275872588157654,
|
| 534 |
+
"learning_rate": 0.001,
|
| 535 |
+
"loss": 0.5983,
|
| 536 |
+
"step": 130
|
| 537 |
+
},
|
| 538 |
+
{
|
| 539 |
+
"epoch": 9.18,
|
| 540 |
+
"grad_norm": 0.3596336245536804,
|
| 541 |
+
"learning_rate": 0.001,
|
| 542 |
+
"loss": 0.4615,
|
| 543 |
+
"step": 132
|
| 544 |
+
},
|
| 545 |
+
{
|
| 546 |
+
"epoch": 9.32,
|
| 547 |
+
"grad_norm": 0.37557095289230347,
|
| 548 |
+
"learning_rate": 0.001,
|
| 549 |
+
"loss": 0.4756,
|
| 550 |
+
"step": 134
|
| 551 |
+
},
|
| 552 |
+
{
|
| 553 |
+
"epoch": 9.46,
|
| 554 |
+
"grad_norm": 0.39249515533447266,
|
| 555 |
+
"learning_rate": 0.001,
|
| 556 |
+
"loss": 0.4546,
|
| 557 |
+
"step": 136
|
| 558 |
+
},
|
| 559 |
+
{
|
| 560 |
+
"epoch": 9.6,
|
| 561 |
+
"grad_norm": 0.3760348856449127,
|
| 562 |
+
"learning_rate": 0.001,
|
| 563 |
+
"loss": 0.4792,
|
| 564 |
+
"step": 138
|
| 565 |
+
},
|
| 566 |
+
{
|
| 567 |
+
"epoch": 9.74,
|
| 568 |
+
"grad_norm": 0.3137217164039612,
|
| 569 |
+
"learning_rate": 0.001,
|
| 570 |
+
"loss": 0.4674,
|
| 571 |
+
"step": 140
|
| 572 |
+
},
|
| 573 |
+
{
|
| 574 |
+
"epoch": 9.88,
|
| 575 |
+
"grad_norm": 0.40549594163894653,
|
| 576 |
+
"learning_rate": 0.001,
|
| 577 |
+
"loss": 0.4939,
|
| 578 |
+
"step": 142
|
| 579 |
+
},
|
| 580 |
+
{
|
| 581 |
+
"epoch": 9.95,
|
| 582 |
+
"eval_loss": 3.5685999393463135,
|
| 583 |
+
"eval_runtime": 81.5958,
|
| 584 |
+
"eval_samples_per_second": 4.142,
|
| 585 |
+
"eval_steps_per_second": 0.527,
|
| 586 |
+
"step": 143
|
| 587 |
+
},
|
| 588 |
+
{
|
| 589 |
+
"epoch": 10.02,
|
| 590 |
+
"grad_norm": 0.4173819422721863,
|
| 591 |
+
"learning_rate": 0.001,
|
| 592 |
+
"loss": 0.5055,
|
| 593 |
+
"step": 144
|
| 594 |
+
},
|
| 595 |
+
{
|
| 596 |
+
"epoch": 10.16,
|
| 597 |
+
"grad_norm": 0.280066579580307,
|
| 598 |
+
"learning_rate": 0.001,
|
| 599 |
+
"loss": 0.3353,
|
| 600 |
+
"step": 146
|
| 601 |
+
},
|
| 602 |
+
{
|
| 603 |
+
"epoch": 10.3,
|
| 604 |
+
"grad_norm": 0.30166783928871155,
|
| 605 |
+
"learning_rate": 0.001,
|
| 606 |
+
"loss": 0.351,
|
| 607 |
+
"step": 148
|
| 608 |
+
},
|
| 609 |
+
{
|
| 610 |
+
"epoch": 10.43,
|
| 611 |
+
"grad_norm": 0.28606531023979187,
|
| 612 |
+
"learning_rate": 0.001,
|
| 613 |
+
"loss": 0.3834,
|
| 614 |
+
"step": 150
|
| 615 |
+
},
|
| 616 |
+
{
|
| 617 |
+
"epoch": 10.57,
|
| 618 |
+
"grad_norm": 0.2835221588611603,
|
| 619 |
+
"learning_rate": 0.001,
|
| 620 |
+
"loss": 0.3718,
|
| 621 |
+
"step": 152
|
| 622 |
+
},
|
| 623 |
+
{
|
| 624 |
+
"epoch": 10.71,
|
| 625 |
+
"grad_norm": 0.3148328959941864,
|
| 626 |
+
"learning_rate": 0.001,
|
| 627 |
+
"loss": 0.3692,
|
| 628 |
+
"step": 154
|
| 629 |
+
},
|
| 630 |
+
{
|
| 631 |
+
"epoch": 10.85,
|
| 632 |
+
"grad_norm": 0.3502219021320343,
|
| 633 |
+
"learning_rate": 0.001,
|
| 634 |
+
"loss": 0.38,
|
| 635 |
+
"step": 156
|
| 636 |
+
},
|
| 637 |
+
{
|
| 638 |
+
"epoch": 10.99,
|
| 639 |
+
"grad_norm": 0.3344653844833374,
|
| 640 |
+
"learning_rate": 0.001,
|
| 641 |
+
"loss": 0.376,
|
| 642 |
+
"step": 158
|
| 643 |
+
},
|
| 644 |
+
{
|
| 645 |
+
"epoch": 10.99,
|
| 646 |
+
"eval_loss": 3.425977945327759,
|
| 647 |
+
"eval_runtime": 81.532,
|
| 648 |
+
"eval_samples_per_second": 4.146,
|
| 649 |
+
"eval_steps_per_second": 0.527,
|
| 650 |
+
"step": 158
|
| 651 |
+
},
|
| 652 |
+
{
|
| 653 |
+
"epoch": 11.13,
|
| 654 |
+
"grad_norm": 0.32332998514175415,
|
| 655 |
+
"learning_rate": 0.001,
|
| 656 |
+
"loss": 0.2827,
|
| 657 |
+
"step": 160
|
| 658 |
+
},
|
| 659 |
+
{
|
| 660 |
+
"epoch": 11.27,
|
| 661 |
+
"grad_norm": 0.35432103276252747,
|
| 662 |
+
"learning_rate": 0.001,
|
| 663 |
+
"loss": 0.2966,
|
| 664 |
+
"step": 162
|
| 665 |
+
},
|
| 666 |
+
{
|
| 667 |
+
"epoch": 11.41,
|
| 668 |
+
"grad_norm": 0.29032111167907715,
|
| 669 |
+
"learning_rate": 0.001,
|
| 670 |
+
"loss": 0.2954,
|
| 671 |
+
"step": 164
|
| 672 |
+
},
|
| 673 |
+
{
|
| 674 |
+
"epoch": 11.55,
|
| 675 |
+
"grad_norm": 0.3170696198940277,
|
| 676 |
+
"learning_rate": 0.001,
|
| 677 |
+
"loss": 0.2738,
|
| 678 |
+
"step": 166
|
| 679 |
+
},
|
| 680 |
+
{
|
| 681 |
+
"epoch": 11.69,
|
| 682 |
+
"grad_norm": 0.3339516520500183,
|
| 683 |
+
"learning_rate": 0.001,
|
| 684 |
+
"loss": 0.2786,
|
| 685 |
+
"step": 168
|
| 686 |
+
},
|
| 687 |
+
{
|
| 688 |
+
"epoch": 11.83,
|
| 689 |
+
"grad_norm": 0.3187398910522461,
|
| 690 |
+
"learning_rate": 0.001,
|
| 691 |
+
"loss": 0.315,
|
| 692 |
+
"step": 170
|
| 693 |
+
},
|
| 694 |
+
{
|
| 695 |
+
"epoch": 11.97,
|
| 696 |
+
"grad_norm": 0.2842791974544525,
|
| 697 |
+
"learning_rate": 0.001,
|
| 698 |
+
"loss": 0.313,
|
| 699 |
+
"step": 172
|
| 700 |
+
},
|
| 701 |
+
{
|
| 702 |
+
"epoch": 11.97,
|
| 703 |
+
"eval_loss": 3.9301607608795166,
|
| 704 |
+
"eval_runtime": 81.5908,
|
| 705 |
+
"eval_samples_per_second": 4.143,
|
| 706 |
+
"eval_steps_per_second": 0.527,
|
| 707 |
+
"step": 172
|
| 708 |
+
},
|
| 709 |
+
{
|
| 710 |
+
"epoch": 12.1,
|
| 711 |
+
"grad_norm": 0.2522130012512207,
|
| 712 |
+
"learning_rate": 0.001,
|
| 713 |
+
"loss": 0.2504,
|
| 714 |
+
"step": 174
|
| 715 |
+
},
|
| 716 |
+
{
|
| 717 |
+
"epoch": 12.24,
|
| 718 |
+
"grad_norm": 0.23560765385627747,
|
| 719 |
+
"learning_rate": 0.001,
|
| 720 |
+
"loss": 0.212,
|
| 721 |
+
"step": 176
|
| 722 |
+
},
|
| 723 |
+
{
|
| 724 |
+
"epoch": 12.38,
|
| 725 |
+
"grad_norm": 0.24140460789203644,
|
| 726 |
+
"learning_rate": 0.001,
|
| 727 |
+
"loss": 0.2156,
|
| 728 |
+
"step": 178
|
| 729 |
+
},
|
| 730 |
+
{
|
| 731 |
+
"epoch": 12.52,
|
| 732 |
+
"grad_norm": 0.2790488302707672,
|
| 733 |
+
"learning_rate": 0.001,
|
| 734 |
+
"loss": 0.2474,
|
| 735 |
+
"step": 180
|
| 736 |
+
},
|
| 737 |
+
{
|
| 738 |
+
"epoch": 12.66,
|
| 739 |
+
"grad_norm": 0.2879179120063782,
|
| 740 |
+
"learning_rate": 0.001,
|
| 741 |
+
"loss": 0.2486,
|
| 742 |
+
"step": 182
|
| 743 |
+
},
|
| 744 |
+
{
|
| 745 |
+
"epoch": 12.8,
|
| 746 |
+
"grad_norm": 0.3126004934310913,
|
| 747 |
+
"learning_rate": 0.001,
|
| 748 |
+
"loss": 0.2499,
|
| 749 |
+
"step": 184
|
| 750 |
+
},
|
| 751 |
+
{
|
| 752 |
+
"epoch": 12.94,
|
| 753 |
+
"grad_norm": 0.3011338412761688,
|
| 754 |
+
"learning_rate": 0.001,
|
| 755 |
+
"loss": 0.2562,
|
| 756 |
+
"step": 186
|
| 757 |
+
},
|
| 758 |
+
{
|
| 759 |
+
"epoch": 12.94,
|
| 760 |
+
"eval_loss": 3.743312120437622,
|
| 761 |
+
"eval_runtime": 81.5885,
|
| 762 |
+
"eval_samples_per_second": 4.143,
|
| 763 |
+
"eval_steps_per_second": 0.527,
|
| 764 |
+
"step": 186
|
| 765 |
+
},
|
| 766 |
+
{
|
| 767 |
+
"epoch": 13.08,
|
| 768 |
+
"grad_norm": 0.24417123198509216,
|
| 769 |
+
"learning_rate": 0.001,
|
| 770 |
+
"loss": 0.2166,
|
| 771 |
+
"step": 188
|
| 772 |
+
},
|
| 773 |
+
{
|
| 774 |
+
"epoch": 13.22,
|
| 775 |
+
"grad_norm": 0.21955759823322296,
|
| 776 |
+
"learning_rate": 0.001,
|
| 777 |
+
"loss": 0.1767,
|
| 778 |
+
"step": 190
|
| 779 |
+
},
|
| 780 |
+
{
|
| 781 |
+
"epoch": 13.36,
|
| 782 |
+
"grad_norm": 0.20537225902080536,
|
| 783 |
+
"learning_rate": 0.001,
|
| 784 |
+
"loss": 0.1715,
|
| 785 |
+
"step": 192
|
| 786 |
+
},
|
| 787 |
+
{
|
| 788 |
+
"epoch": 13.5,
|
| 789 |
+
"grad_norm": 0.21406413614749908,
|
| 790 |
+
"learning_rate": 0.001,
|
| 791 |
+
"loss": 0.1857,
|
| 792 |
+
"step": 194
|
| 793 |
+
},
|
| 794 |
+
{
|
| 795 |
+
"epoch": 13.63,
|
| 796 |
+
"grad_norm": 0.21677067875862122,
|
| 797 |
+
"learning_rate": 0.001,
|
| 798 |
+
"loss": 0.1881,
|
| 799 |
+
"step": 196
|
| 800 |
+
},
|
| 801 |
+
{
|
| 802 |
+
"epoch": 13.77,
|
| 803 |
+
"grad_norm": 0.2592070996761322,
|
| 804 |
+
"learning_rate": 0.001,
|
| 805 |
+
"loss": 0.2022,
|
| 806 |
+
"step": 198
|
| 807 |
+
},
|
| 808 |
+
{
|
| 809 |
+
"epoch": 13.91,
|
| 810 |
+
"grad_norm": 0.23913638293743134,
|
| 811 |
+
"learning_rate": 0.001,
|
| 812 |
+
"loss": 0.2051,
|
| 813 |
+
"step": 200
|
| 814 |
+
},
|
| 815 |
+
{
|
| 816 |
+
"epoch": 13.98,
|
| 817 |
+
"eval_loss": 3.911346197128296,
|
| 818 |
+
"eval_runtime": 81.5425,
|
| 819 |
+
"eval_samples_per_second": 4.145,
|
| 820 |
+
"eval_steps_per_second": 0.527,
|
| 821 |
+
"step": 201
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 14.05,
|
| 825 |
+
"grad_norm": 0.19888806343078613,
|
| 826 |
+
"learning_rate": 0.001,
|
| 827 |
+
"loss": 0.1774,
|
| 828 |
+
"step": 202
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 14.19,
|
| 832 |
+
"grad_norm": 0.17841410636901855,
|
| 833 |
+
"learning_rate": 0.001,
|
| 834 |
+
"loss": 0.1409,
|
| 835 |
+
"step": 204
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 14.33,
|
| 839 |
+
"grad_norm": 0.22502601146697998,
|
| 840 |
+
"learning_rate": 0.001,
|
| 841 |
+
"loss": 0.1432,
|
| 842 |
+
"step": 206
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 14.47,
|
| 846 |
+
"grad_norm": 0.21947847306728363,
|
| 847 |
+
"learning_rate": 0.001,
|
| 848 |
+
"loss": 0.1487,
|
| 849 |
+
"step": 208
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 14.61,
|
| 853 |
+
"grad_norm": 0.20319664478302002,
|
| 854 |
+
"learning_rate": 0.001,
|
| 855 |
+
"loss": 0.1753,
|
| 856 |
+
"step": 210
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 14.75,
|
| 860 |
+
"grad_norm": 0.20484566688537598,
|
| 861 |
+
"learning_rate": 0.001,
|
| 862 |
+
"loss": 0.1627,
|
| 863 |
+
"step": 212
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 14.89,
|
| 867 |
+
"grad_norm": 0.24411869049072266,
|
| 868 |
+
"learning_rate": 0.001,
|
| 869 |
+
"loss": 0.1802,
|
| 870 |
+
"step": 214
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 14.96,
|
| 874 |
+
"eval_loss": 4.0449538230896,
|
| 875 |
+
"eval_runtime": 81.5583,
|
| 876 |
+
"eval_samples_per_second": 4.144,
|
| 877 |
+
"eval_steps_per_second": 0.527,
|
| 878 |
+
"step": 215
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 15.03,
|
| 882 |
+
"grad_norm": 0.23610645532608032,
|
| 883 |
+
"learning_rate": 0.001,
|
| 884 |
+
"loss": 0.1881,
|
| 885 |
+
"step": 216
|
| 886 |
+
},
|
| 887 |
+
{
|
| 888 |
+
"epoch": 15.17,
|
| 889 |
+
"grad_norm": 0.17829175293445587,
|
| 890 |
+
"learning_rate": 0.001,
|
| 891 |
+
"loss": 0.123,
|
| 892 |
+
"step": 218
|
| 893 |
+
},
|
| 894 |
+
{
|
| 895 |
+
"epoch": 15.3,
|
| 896 |
+
"grad_norm": 0.178519606590271,
|
| 897 |
+
"learning_rate": 0.001,
|
| 898 |
+
"loss": 0.1166,
|
| 899 |
+
"step": 220
|
| 900 |
+
},
|
| 901 |
+
{
|
| 902 |
+
"epoch": 15.44,
|
| 903 |
+
"grad_norm": 0.19595706462860107,
|
| 904 |
+
"learning_rate": 0.001,
|
| 905 |
+
"loss": 0.135,
|
| 906 |
+
"step": 222
|
| 907 |
+
},
|
| 908 |
+
{
|
| 909 |
+
"epoch": 15.58,
|
| 910 |
+
"grad_norm": 0.20790521800518036,
|
| 911 |
+
"learning_rate": 0.001,
|
| 912 |
+
"loss": 0.1494,
|
| 913 |
+
"step": 224
|
| 914 |
+
},
|
| 915 |
+
{
|
| 916 |
+
"epoch": 15.72,
|
| 917 |
+
"grad_norm": 0.1832074671983719,
|
| 918 |
+
"learning_rate": 0.001,
|
| 919 |
+
"loss": 0.1488,
|
| 920 |
+
"step": 226
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 15.86,
|
| 924 |
+
"grad_norm": 0.17795896530151367,
|
| 925 |
+
"learning_rate": 0.001,
|
| 926 |
+
"loss": 0.1448,
|
| 927 |
+
"step": 228
|
| 928 |
+
},
|
| 929 |
+
{
|
| 930 |
+
"epoch": 16.0,
|
| 931 |
+
"grad_norm": 0.20039702951908112,
|
| 932 |
+
"learning_rate": 0.001,
|
| 933 |
+
"loss": 0.1378,
|
| 934 |
+
"step": 230
|
| 935 |
+
},
|
| 936 |
+
{
|
| 937 |
+
"epoch": 16.0,
|
| 938 |
+
"eval_loss": 3.939739227294922,
|
| 939 |
+
"eval_runtime": 81.6032,
|
| 940 |
+
"eval_samples_per_second": 4.142,
|
| 941 |
+
"eval_steps_per_second": 0.527,
|
| 942 |
+
"step": 230
|
| 943 |
+
},
|
| 944 |
+
{
|
| 945 |
+
"epoch": 16.14,
|
| 946 |
+
"grad_norm": 0.19622142612934113,
|
| 947 |
+
"learning_rate": 0.001,
|
| 948 |
+
"loss": 0.3001,
|
| 949 |
+
"step": 232
|
| 950 |
+
},
|
| 951 |
+
{
|
| 952 |
+
"epoch": 16.28,
|
| 953 |
+
"grad_norm": 19.05455207824707,
|
| 954 |
+
"learning_rate": 0.001,
|
| 955 |
+
"loss": 0.2708,
|
| 956 |
+
"step": 234
|
| 957 |
+
},
|
| 958 |
+
{
|
| 959 |
+
"epoch": 16.42,
|
| 960 |
+
"grad_norm": 29.798582077026367,
|
| 961 |
+
"learning_rate": 0.001,
|
| 962 |
+
"loss": 0.2154,
|
| 963 |
+
"step": 236
|
| 964 |
+
},
|
| 965 |
+
{
|
| 966 |
+
"epoch": 16.56,
|
| 967 |
+
"grad_norm": 8.835821151733398,
|
| 968 |
+
"learning_rate": 0.001,
|
| 969 |
+
"loss": 0.1348,
|
| 970 |
+
"step": 238
|
| 971 |
+
},
|
| 972 |
+
{
|
| 973 |
+
"epoch": 16.7,
|
| 974 |
+
"grad_norm": 0.3760863244533539,
|
| 975 |
+
"learning_rate": 0.001,
|
| 976 |
+
"loss": 0.6235,
|
| 977 |
+
"step": 240
|
| 978 |
+
},
|
| 979 |
+
{
|
| 980 |
+
"epoch": 16.83,
|
| 981 |
+
"grad_norm": 0.3473583459854126,
|
| 982 |
+
"learning_rate": 0.001,
|
| 983 |
+
"loss": 0.1445,
|
| 984 |
+
"step": 242
|
| 985 |
+
},
|
| 986 |
+
{
|
| 987 |
+
"epoch": 16.97,
|
| 988 |
+
"grad_norm": 0.4041793942451477,
|
| 989 |
+
"learning_rate": 0.001,
|
| 990 |
+
"loss": 0.1546,
|
| 991 |
+
"step": 244
|
| 992 |
+
},
|
| 993 |
+
{
|
| 994 |
+
"epoch": 16.97,
|
| 995 |
+
"eval_loss": 4.307888984680176,
|
| 996 |
+
"eval_runtime": 81.6566,
|
| 997 |
+
"eval_samples_per_second": 4.139,
|
| 998 |
+
"eval_steps_per_second": 0.527,
|
| 999 |
+
"step": 244
|
| 1000 |
+
},
|
| 1001 |
+
{
|
| 1002 |
+
"epoch": 17.11,
|
| 1003 |
+
"grad_norm": 0.2586219906806946,
|
| 1004 |
+
"learning_rate": 0.001,
|
| 1005 |
+
"loss": 0.1188,
|
| 1006 |
+
"step": 246
|
| 1007 |
+
},
|
| 1008 |
+
{
|
| 1009 |
+
"epoch": 17.25,
|
| 1010 |
+
"grad_norm": 0.4334220886230469,
|
| 1011 |
+
"learning_rate": 0.001,
|
| 1012 |
+
"loss": 0.1041,
|
| 1013 |
+
"step": 248
|
| 1014 |
+
},
|
| 1015 |
+
{
|
| 1016 |
+
"epoch": 17.39,
|
| 1017 |
+
"grad_norm": 17.520734786987305,
|
| 1018 |
+
"learning_rate": 0.001,
|
| 1019 |
+
"loss": 0.1108,
|
| 1020 |
+
"step": 250
|
| 1021 |
+
},
|
| 1022 |
+
{
|
| 1023 |
+
"epoch": 17.53,
|
| 1024 |
+
"grad_norm": 0.5943770408630371,
|
| 1025 |
+
"learning_rate": 0.001,
|
| 1026 |
+
"loss": 0.1146,
|
| 1027 |
+
"step": 252
|
| 1028 |
+
},
|
| 1029 |
+
{
|
| 1030 |
+
"epoch": 17.67,
|
| 1031 |
+
"grad_norm": 0.4325353503227234,
|
| 1032 |
+
"learning_rate": 0.001,
|
| 1033 |
+
"loss": 0.1325,
|
| 1034 |
+
"step": 254
|
| 1035 |
+
},
|
| 1036 |
+
{
|
| 1037 |
+
"epoch": 17.81,
|
| 1038 |
+
"grad_norm": 0.41412413120269775,
|
| 1039 |
+
"learning_rate": 0.001,
|
| 1040 |
+
"loss": 0.1491,
|
| 1041 |
+
"step": 256
|
| 1042 |
+
},
|
| 1043 |
+
{
|
| 1044 |
+
"epoch": 17.95,
|
| 1045 |
+
"grad_norm": 0.19986829161643982,
|
| 1046 |
+
"learning_rate": 0.001,
|
| 1047 |
+
"loss": 0.1375,
|
| 1048 |
+
"step": 258
|
| 1049 |
+
},
|
| 1050 |
+
{
|
| 1051 |
+
"epoch": 17.95,
|
| 1052 |
+
"eval_loss": 4.552526950836182,
|
| 1053 |
+
"eval_runtime": 81.6054,
|
| 1054 |
+
"eval_samples_per_second": 4.142,
|
| 1055 |
+
"eval_steps_per_second": 0.527,
|
| 1056 |
+
"step": 258
|
| 1057 |
+
},
|
| 1058 |
+
{
|
| 1059 |
+
"epoch": 18.09,
|
| 1060 |
+
"grad_norm": 0.7999384999275208,
|
| 1061 |
+
"learning_rate": 0.001,
|
| 1062 |
+
"loss": 0.1155,
|
| 1063 |
+
"step": 260
|
| 1064 |
+
},
|
| 1065 |
+
{
|
| 1066 |
+
"epoch": 18.23,
|
| 1067 |
+
"grad_norm": 0.17563021183013916,
|
| 1068 |
+
"learning_rate": 0.001,
|
| 1069 |
+
"loss": 0.1006,
|
| 1070 |
+
"step": 262
|
| 1071 |
+
},
|
| 1072 |
+
{
|
| 1073 |
+
"epoch": 18.37,
|
| 1074 |
+
"grad_norm": 0.17661228775978088,
|
| 1075 |
+
"learning_rate": 0.001,
|
| 1076 |
+
"loss": 0.1062,
|
| 1077 |
+
"step": 264
|
| 1078 |
+
},
|
| 1079 |
+
{
|
| 1080 |
+
"epoch": 18.5,
|
| 1081 |
+
"grad_norm": 0.17768113315105438,
|
| 1082 |
+
"learning_rate": 0.001,
|
| 1083 |
+
"loss": 0.1059,
|
| 1084 |
+
"step": 266
|
| 1085 |
+
},
|
| 1086 |
+
{
|
| 1087 |
+
"epoch": 18.64,
|
| 1088 |
+
"grad_norm": 0.15412819385528564,
|
| 1089 |
+
"learning_rate": 0.001,
|
| 1090 |
+
"loss": 0.0981,
|
| 1091 |
+
"step": 268
|
| 1092 |
+
},
|
| 1093 |
+
{
|
| 1094 |
+
"epoch": 18.78,
|
| 1095 |
+
"grad_norm": 0.1754271388053894,
|
| 1096 |
+
"learning_rate": 0.001,
|
| 1097 |
+
"loss": 0.0988,
|
| 1098 |
+
"step": 270
|
| 1099 |
+
},
|
| 1100 |
+
{
|
| 1101 |
+
"epoch": 18.92,
|
| 1102 |
+
"grad_norm": 0.15736614167690277,
|
| 1103 |
+
"learning_rate": 0.001,
|
| 1104 |
+
"loss": 0.1005,
|
| 1105 |
+
"step": 272
|
| 1106 |
+
},
|
| 1107 |
+
{
|
| 1108 |
+
"epoch": 18.99,
|
| 1109 |
+
"eval_loss": 4.900540828704834,
|
| 1110 |
+
"eval_runtime": 81.5789,
|
| 1111 |
+
"eval_samples_per_second": 4.143,
|
| 1112 |
+
"eval_steps_per_second": 0.527,
|
| 1113 |
+
"step": 273
|
| 1114 |
+
},
|
| 1115 |
+
{
|
| 1116 |
+
"epoch": 19.06,
|
| 1117 |
+
"grad_norm": 0.1531495302915573,
|
| 1118 |
+
"learning_rate": 0.001,
|
| 1119 |
+
"loss": 0.0844,
|
| 1120 |
+
"step": 274
|
| 1121 |
+
},
|
| 1122 |
+
{
|
| 1123 |
+
"epoch": 19.2,
|
| 1124 |
+
"grad_norm": 0.15237411856651306,
|
| 1125 |
+
"learning_rate": 0.001,
|
| 1126 |
+
"loss": 0.0752,
|
| 1127 |
+
"step": 276
|
| 1128 |
+
},
|
| 1129 |
+
{
|
| 1130 |
+
"epoch": 19.34,
|
| 1131 |
+
"grad_norm": 0.1433786153793335,
|
| 1132 |
+
"learning_rate": 0.001,
|
| 1133 |
+
"loss": 0.0782,
|
| 1134 |
+
"step": 278
|
| 1135 |
+
},
|
| 1136 |
+
{
|
| 1137 |
+
"epoch": 19.48,
|
| 1138 |
+
"grad_norm": 0.1296713650226593,
|
| 1139 |
+
"learning_rate": 0.001,
|
| 1140 |
+
"loss": 0.0808,
|
| 1141 |
+
"step": 280
|
| 1142 |
+
},
|
| 1143 |
+
{
|
| 1144 |
+
"epoch": 19.48,
|
| 1145 |
+
"eval_loss": 4.81671667098999,
|
| 1146 |
+
"eval_runtime": 81.4692,
|
| 1147 |
+
"eval_samples_per_second": 4.149,
|
| 1148 |
+
"eval_steps_per_second": 0.528,
|
| 1149 |
+
"step": 280
|
| 1150 |
+
},
|
| 1151 |
+
{
|
| 1152 |
+
"epoch": 19.48,
|
| 1153 |
+
"step": 280,
|
| 1154 |
+
"total_flos": 4.895208054457934e+18,
|
| 1155 |
+
"train_loss": 0.8494854368801628,
|
| 1156 |
+
"train_runtime": 68771.7044,
|
| 1157 |
+
"train_samples_per_second": 1.068,
|
| 1158 |
+
"train_steps_per_second": 0.004
|
| 1159 |
+
}
|
| 1160 |
+
],
|
| 1161 |
+
"logging_steps": 2,
|
| 1162 |
+
"max_steps": 280,
|
| 1163 |
+
"num_input_tokens_seen": 0,
|
| 1164 |
+
"num_train_epochs": 20,
|
| 1165 |
+
"save_steps": 500,
|
| 1166 |
+
"total_flos": 4.895208054457934e+18,
|
| 1167 |
+
"train_batch_size": 8,
|
| 1168 |
+
"trial_name": null,
|
| 1169 |
+
"trial_params": null
|
| 1170 |
+
}
|