🍻 cheers
Browse files- README.md +6 -5
- all_results.json +13 -0
- eval_results.json +8 -0
- runs/Apr22_19-46-52_25f23db457d4/events.out.tfevents.1713817857.25f23db457d4.34.1 +3 -0
- train_results.json +8 -0
- trainer_state.json +1772 -0
README.md
CHANGED
|
@@ -2,6 +2,7 @@
|
|
| 2 |
license: apache-2.0
|
| 3 |
base_model: microsoft/swinv2-tiny-patch4-window16-256
|
| 4 |
tags:
|
|
|
|
| 5 |
- generated_from_trainer
|
| 6 |
datasets:
|
| 7 |
- imagefolder
|
|
@@ -14,7 +15,7 @@ model-index:
|
|
| 14 |
name: Image Classification
|
| 15 |
type: image-classification
|
| 16 |
dataset:
|
| 17 |
-
name:
|
| 18 |
type: imagefolder
|
| 19 |
config: default
|
| 20 |
split: train
|
|
@@ -22,7 +23,7 @@ model-index:
|
|
| 22 |
metrics:
|
| 23 |
- name: Accuracy
|
| 24 |
type: accuracy
|
| 25 |
-
value: 0.
|
| 26 |
---
|
| 27 |
|
| 28 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
@@ -30,10 +31,10 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 30 |
|
| 31 |
# Action_agent
|
| 32 |
|
| 33 |
-
This model is a fine-tuned version of [microsoft/swinv2-tiny-patch4-window16-256](https://huggingface.co/microsoft/swinv2-tiny-patch4-window16-256) on the
|
| 34 |
It achieves the following results on the evaluation set:
|
| 35 |
-
- Loss: 0.
|
| 36 |
-
- Accuracy: 0.
|
| 37 |
|
| 38 |
## Model description
|
| 39 |
|
|
|
|
| 2 |
license: apache-2.0
|
| 3 |
base_model: microsoft/swinv2-tiny-patch4-window16-256
|
| 4 |
tags:
|
| 5 |
+
- image-classification
|
| 6 |
- generated_from_trainer
|
| 7 |
datasets:
|
| 8 |
- imagefolder
|
|
|
|
| 15 |
name: Image Classification
|
| 16 |
type: image-classification
|
| 17 |
dataset:
|
| 18 |
+
name: agent_action_class
|
| 19 |
type: imagefolder
|
| 20 |
config: default
|
| 21 |
split: train
|
|
|
|
| 23 |
metrics:
|
| 24 |
- name: Accuracy
|
| 25 |
type: accuracy
|
| 26 |
+
value: 0.7895238095238095
|
| 27 |
---
|
| 28 |
|
| 29 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
|
|
| 31 |
|
| 32 |
# Action_agent
|
| 33 |
|
| 34 |
+
This model is a fine-tuned version of [microsoft/swinv2-tiny-patch4-window16-256](https://huggingface.co/microsoft/swinv2-tiny-patch4-window16-256) on the agent_action_class dataset.
|
| 35 |
It achieves the following results on the evaluation set:
|
| 36 |
+
- Loss: 0.6309
|
| 37 |
+
- Accuracy: 0.7895
|
| 38 |
|
| 39 |
## Model description
|
| 40 |
|
all_results.json
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"eval_accuracy": 0.7895238095238095,
|
| 4 |
+
"eval_loss": 0.6309406161308289,
|
| 5 |
+
"eval_runtime": 13.6501,
|
| 6 |
+
"eval_samples_per_second": 76.922,
|
| 7 |
+
"eval_steps_per_second": 9.67,
|
| 8 |
+
"total_flos": 3.265548125287219e+18,
|
| 9 |
+
"train_loss": 0.7155859537944672,
|
| 10 |
+
"train_runtime": 2626.1064,
|
| 11 |
+
"train_samples_per_second": 38.212,
|
| 12 |
+
"train_steps_per_second": 1.196
|
| 13 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"eval_accuracy": 0.7895238095238095,
|
| 4 |
+
"eval_loss": 0.6309406161308289,
|
| 5 |
+
"eval_runtime": 13.6501,
|
| 6 |
+
"eval_samples_per_second": 76.922,
|
| 7 |
+
"eval_steps_per_second": 9.67
|
| 8 |
+
}
|
runs/Apr22_19-46-52_25f23db457d4/events.out.tfevents.1713817857.25f23db457d4.34.1
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9cf74ea96e9b43de8ad834f5d54784b30c7dcdb390678146c2a4d45ccec9fc69
|
| 3 |
+
size 411
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"total_flos": 3.265548125287219e+18,
|
| 4 |
+
"train_loss": 0.7155859537944672,
|
| 5 |
+
"train_runtime": 2626.1064,
|
| 6 |
+
"train_samples_per_second": 38.212,
|
| 7 |
+
"train_steps_per_second": 1.196
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1772 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.6309406161308289,
|
| 3 |
+
"best_model_checkpoint": "Action_agent/checkpoint-2600",
|
| 4 |
+
"epoch": 10.0,
|
| 5 |
+
"eval_steps": 100,
|
| 6 |
+
"global_step": 3140,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.05,
|
| 13 |
+
"grad_norm": 6.1902642250061035,
|
| 14 |
+
"learning_rate": 9.952229299363057e-06,
|
| 15 |
+
"loss": 2.3093,
|
| 16 |
+
"step": 15
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.1,
|
| 20 |
+
"grad_norm": 8.609545707702637,
|
| 21 |
+
"learning_rate": 9.904458598726116e-06,
|
| 22 |
+
"loss": 2.2798,
|
| 23 |
+
"step": 30
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.14,
|
| 27 |
+
"grad_norm": 8.198623657226562,
|
| 28 |
+
"learning_rate": 9.856687898089172e-06,
|
| 29 |
+
"loss": 2.2163,
|
| 30 |
+
"step": 45
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.19,
|
| 34 |
+
"grad_norm": 7.1882829666137695,
|
| 35 |
+
"learning_rate": 9.80891719745223e-06,
|
| 36 |
+
"loss": 2.1529,
|
| 37 |
+
"step": 60
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.24,
|
| 41 |
+
"grad_norm": 8.259012222290039,
|
| 42 |
+
"learning_rate": 9.761146496815288e-06,
|
| 43 |
+
"loss": 2.114,
|
| 44 |
+
"step": 75
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.29,
|
| 48 |
+
"grad_norm": 9.213942527770996,
|
| 49 |
+
"learning_rate": 9.713375796178345e-06,
|
| 50 |
+
"loss": 2.039,
|
| 51 |
+
"step": 90
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.32,
|
| 55 |
+
"eval_accuracy": 0.4847619047619048,
|
| 56 |
+
"eval_loss": 1.7706103324890137,
|
| 57 |
+
"eval_runtime": 17.9171,
|
| 58 |
+
"eval_samples_per_second": 58.603,
|
| 59 |
+
"eval_steps_per_second": 7.367,
|
| 60 |
+
"step": 100
|
| 61 |
+
},
|
| 62 |
+
{
|
| 63 |
+
"epoch": 0.33,
|
| 64 |
+
"grad_norm": 8.229316711425781,
|
| 65 |
+
"learning_rate": 9.665605095541401e-06,
|
| 66 |
+
"loss": 1.9431,
|
| 67 |
+
"step": 105
|
| 68 |
+
},
|
| 69 |
+
{
|
| 70 |
+
"epoch": 0.38,
|
| 71 |
+
"grad_norm": 9.092002868652344,
|
| 72 |
+
"learning_rate": 9.617834394904459e-06,
|
| 73 |
+
"loss": 1.7759,
|
| 74 |
+
"step": 120
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 0.43,
|
| 78 |
+
"grad_norm": 12.155420303344727,
|
| 79 |
+
"learning_rate": 9.570063694267517e-06,
|
| 80 |
+
"loss": 1.7233,
|
| 81 |
+
"step": 135
|
| 82 |
+
},
|
| 83 |
+
{
|
| 84 |
+
"epoch": 0.48,
|
| 85 |
+
"grad_norm": 15.371952056884766,
|
| 86 |
+
"learning_rate": 9.522292993630574e-06,
|
| 87 |
+
"loss": 1.6441,
|
| 88 |
+
"step": 150
|
| 89 |
+
},
|
| 90 |
+
{
|
| 91 |
+
"epoch": 0.53,
|
| 92 |
+
"grad_norm": 15.928441047668457,
|
| 93 |
+
"learning_rate": 9.47452229299363e-06,
|
| 94 |
+
"loss": 1.5518,
|
| 95 |
+
"step": 165
|
| 96 |
+
},
|
| 97 |
+
{
|
| 98 |
+
"epoch": 0.57,
|
| 99 |
+
"grad_norm": 14.13963794708252,
|
| 100 |
+
"learning_rate": 9.426751592356688e-06,
|
| 101 |
+
"loss": 1.4362,
|
| 102 |
+
"step": 180
|
| 103 |
+
},
|
| 104 |
+
{
|
| 105 |
+
"epoch": 0.62,
|
| 106 |
+
"grad_norm": 22.86189842224121,
|
| 107 |
+
"learning_rate": 9.378980891719746e-06,
|
| 108 |
+
"loss": 1.3695,
|
| 109 |
+
"step": 195
|
| 110 |
+
},
|
| 111 |
+
{
|
| 112 |
+
"epoch": 0.64,
|
| 113 |
+
"eval_accuracy": 0.6457142857142857,
|
| 114 |
+
"eval_loss": 1.0885976552963257,
|
| 115 |
+
"eval_runtime": 13.391,
|
| 116 |
+
"eval_samples_per_second": 78.411,
|
| 117 |
+
"eval_steps_per_second": 9.857,
|
| 118 |
+
"step": 200
|
| 119 |
+
},
|
| 120 |
+
{
|
| 121 |
+
"epoch": 0.67,
|
| 122 |
+
"grad_norm": 18.267860412597656,
|
| 123 |
+
"learning_rate": 9.331210191082803e-06,
|
| 124 |
+
"loss": 1.2723,
|
| 125 |
+
"step": 210
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"epoch": 0.72,
|
| 129 |
+
"grad_norm": 17.083765029907227,
|
| 130 |
+
"learning_rate": 9.283439490445861e-06,
|
| 131 |
+
"loss": 1.2886,
|
| 132 |
+
"step": 225
|
| 133 |
+
},
|
| 134 |
+
{
|
| 135 |
+
"epoch": 0.76,
|
| 136 |
+
"grad_norm": 13.116181373596191,
|
| 137 |
+
"learning_rate": 9.235668789808919e-06,
|
| 138 |
+
"loss": 1.1825,
|
| 139 |
+
"step": 240
|
| 140 |
+
},
|
| 141 |
+
{
|
| 142 |
+
"epoch": 0.81,
|
| 143 |
+
"grad_norm": 15.386600494384766,
|
| 144 |
+
"learning_rate": 9.187898089171975e-06,
|
| 145 |
+
"loss": 1.2512,
|
| 146 |
+
"step": 255
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 0.86,
|
| 150 |
+
"grad_norm": 36.78362274169922,
|
| 151 |
+
"learning_rate": 9.140127388535032e-06,
|
| 152 |
+
"loss": 1.1629,
|
| 153 |
+
"step": 270
|
| 154 |
+
},
|
| 155 |
+
{
|
| 156 |
+
"epoch": 0.91,
|
| 157 |
+
"grad_norm": 16.107894897460938,
|
| 158 |
+
"learning_rate": 9.09235668789809e-06,
|
| 159 |
+
"loss": 1.0824,
|
| 160 |
+
"step": 285
|
| 161 |
+
},
|
| 162 |
+
{
|
| 163 |
+
"epoch": 0.96,
|
| 164 |
+
"grad_norm": 15.064929962158203,
|
| 165 |
+
"learning_rate": 9.044585987261148e-06,
|
| 166 |
+
"loss": 1.099,
|
| 167 |
+
"step": 300
|
| 168 |
+
},
|
| 169 |
+
{
|
| 170 |
+
"epoch": 0.96,
|
| 171 |
+
"eval_accuracy": 0.68,
|
| 172 |
+
"eval_loss": 0.9092212915420532,
|
| 173 |
+
"eval_runtime": 14.0444,
|
| 174 |
+
"eval_samples_per_second": 74.763,
|
| 175 |
+
"eval_steps_per_second": 9.399,
|
| 176 |
+
"step": 300
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 1.0,
|
| 180 |
+
"grad_norm": 18.469070434570312,
|
| 181 |
+
"learning_rate": 8.996815286624204e-06,
|
| 182 |
+
"loss": 1.0649,
|
| 183 |
+
"step": 315
|
| 184 |
+
},
|
| 185 |
+
{
|
| 186 |
+
"epoch": 1.05,
|
| 187 |
+
"grad_norm": 17.13619613647461,
|
| 188 |
+
"learning_rate": 8.949044585987261e-06,
|
| 189 |
+
"loss": 1.1011,
|
| 190 |
+
"step": 330
|
| 191 |
+
},
|
| 192 |
+
{
|
| 193 |
+
"epoch": 1.1,
|
| 194 |
+
"grad_norm": 18.0528621673584,
|
| 195 |
+
"learning_rate": 8.901273885350319e-06,
|
| 196 |
+
"loss": 1.0787,
|
| 197 |
+
"step": 345
|
| 198 |
+
},
|
| 199 |
+
{
|
| 200 |
+
"epoch": 1.15,
|
| 201 |
+
"grad_norm": 14.162128448486328,
|
| 202 |
+
"learning_rate": 8.853503184713377e-06,
|
| 203 |
+
"loss": 1.0927,
|
| 204 |
+
"step": 360
|
| 205 |
+
},
|
| 206 |
+
{
|
| 207 |
+
"epoch": 1.19,
|
| 208 |
+
"grad_norm": 21.933330535888672,
|
| 209 |
+
"learning_rate": 8.805732484076433e-06,
|
| 210 |
+
"loss": 1.087,
|
| 211 |
+
"step": 375
|
| 212 |
+
},
|
| 213 |
+
{
|
| 214 |
+
"epoch": 1.24,
|
| 215 |
+
"grad_norm": 16.677528381347656,
|
| 216 |
+
"learning_rate": 8.757961783439492e-06,
|
| 217 |
+
"loss": 1.0011,
|
| 218 |
+
"step": 390
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 1.27,
|
| 222 |
+
"eval_accuracy": 0.7171428571428572,
|
| 223 |
+
"eval_loss": 0.8182899951934814,
|
| 224 |
+
"eval_runtime": 13.5557,
|
| 225 |
+
"eval_samples_per_second": 77.458,
|
| 226 |
+
"eval_steps_per_second": 9.738,
|
| 227 |
+
"step": 400
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 1.29,
|
| 231 |
+
"grad_norm": 13.870976448059082,
|
| 232 |
+
"learning_rate": 8.710191082802548e-06,
|
| 233 |
+
"loss": 0.9318,
|
| 234 |
+
"step": 405
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 1.34,
|
| 238 |
+
"grad_norm": 14.700624465942383,
|
| 239 |
+
"learning_rate": 8.662420382165606e-06,
|
| 240 |
+
"loss": 0.8846,
|
| 241 |
+
"step": 420
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 1.39,
|
| 245 |
+
"grad_norm": 17.18898582458496,
|
| 246 |
+
"learning_rate": 8.614649681528664e-06,
|
| 247 |
+
"loss": 1.0089,
|
| 248 |
+
"step": 435
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 1.43,
|
| 252 |
+
"grad_norm": 18.89067840576172,
|
| 253 |
+
"learning_rate": 8.566878980891721e-06,
|
| 254 |
+
"loss": 0.9356,
|
| 255 |
+
"step": 450
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 1.48,
|
| 259 |
+
"grad_norm": 15.696223258972168,
|
| 260 |
+
"learning_rate": 8.519108280254777e-06,
|
| 261 |
+
"loss": 0.8215,
|
| 262 |
+
"step": 465
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 1.53,
|
| 266 |
+
"grad_norm": 36.806602478027344,
|
| 267 |
+
"learning_rate": 8.471337579617835e-06,
|
| 268 |
+
"loss": 0.9668,
|
| 269 |
+
"step": 480
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 1.58,
|
| 273 |
+
"grad_norm": 13.624329566955566,
|
| 274 |
+
"learning_rate": 8.423566878980893e-06,
|
| 275 |
+
"loss": 0.8437,
|
| 276 |
+
"step": 495
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 1.59,
|
| 280 |
+
"eval_accuracy": 0.719047619047619,
|
| 281 |
+
"eval_loss": 0.7674332857131958,
|
| 282 |
+
"eval_runtime": 13.5992,
|
| 283 |
+
"eval_samples_per_second": 77.211,
|
| 284 |
+
"eval_steps_per_second": 9.706,
|
| 285 |
+
"step": 500
|
| 286 |
+
},
|
| 287 |
+
{
|
| 288 |
+
"epoch": 1.62,
|
| 289 |
+
"grad_norm": 15.652780532836914,
|
| 290 |
+
"learning_rate": 8.37579617834395e-06,
|
| 291 |
+
"loss": 0.8169,
|
| 292 |
+
"step": 510
|
| 293 |
+
},
|
| 294 |
+
{
|
| 295 |
+
"epoch": 1.67,
|
| 296 |
+
"grad_norm": 13.216598510742188,
|
| 297 |
+
"learning_rate": 8.328025477707006e-06,
|
| 298 |
+
"loss": 0.9283,
|
| 299 |
+
"step": 525
|
| 300 |
+
},
|
| 301 |
+
{
|
| 302 |
+
"epoch": 1.72,
|
| 303 |
+
"grad_norm": 19.80782127380371,
|
| 304 |
+
"learning_rate": 8.280254777070064e-06,
|
| 305 |
+
"loss": 0.8614,
|
| 306 |
+
"step": 540
|
| 307 |
+
},
|
| 308 |
+
{
|
| 309 |
+
"epoch": 1.77,
|
| 310 |
+
"grad_norm": 18.636619567871094,
|
| 311 |
+
"learning_rate": 8.232484076433122e-06,
|
| 312 |
+
"loss": 0.8656,
|
| 313 |
+
"step": 555
|
| 314 |
+
},
|
| 315 |
+
{
|
| 316 |
+
"epoch": 1.82,
|
| 317 |
+
"grad_norm": 17.433523178100586,
|
| 318 |
+
"learning_rate": 8.18471337579618e-06,
|
| 319 |
+
"loss": 0.8313,
|
| 320 |
+
"step": 570
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 1.86,
|
| 324 |
+
"grad_norm": 14.271307945251465,
|
| 325 |
+
"learning_rate": 8.136942675159237e-06,
|
| 326 |
+
"loss": 0.8857,
|
| 327 |
+
"step": 585
|
| 328 |
+
},
|
| 329 |
+
{
|
| 330 |
+
"epoch": 1.91,
|
| 331 |
+
"grad_norm": 16.59528923034668,
|
| 332 |
+
"learning_rate": 8.089171974522295e-06,
|
| 333 |
+
"loss": 0.8613,
|
| 334 |
+
"step": 600
|
| 335 |
+
},
|
| 336 |
+
{
|
| 337 |
+
"epoch": 1.91,
|
| 338 |
+
"eval_accuracy": 0.7409523809523809,
|
| 339 |
+
"eval_loss": 0.7168479561805725,
|
| 340 |
+
"eval_runtime": 13.4058,
|
| 341 |
+
"eval_samples_per_second": 78.324,
|
| 342 |
+
"eval_steps_per_second": 9.846,
|
| 343 |
+
"step": 600
|
| 344 |
+
},
|
| 345 |
+
{
|
| 346 |
+
"epoch": 1.96,
|
| 347 |
+
"grad_norm": 13.598006248474121,
|
| 348 |
+
"learning_rate": 8.04140127388535e-06,
|
| 349 |
+
"loss": 0.8694,
|
| 350 |
+
"step": 615
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 2.01,
|
| 354 |
+
"grad_norm": 13.958739280700684,
|
| 355 |
+
"learning_rate": 7.993630573248408e-06,
|
| 356 |
+
"loss": 0.9166,
|
| 357 |
+
"step": 630
|
| 358 |
+
},
|
| 359 |
+
{
|
| 360 |
+
"epoch": 2.05,
|
| 361 |
+
"grad_norm": 13.80545425415039,
|
| 362 |
+
"learning_rate": 7.945859872611466e-06,
|
| 363 |
+
"loss": 0.8781,
|
| 364 |
+
"step": 645
|
| 365 |
+
},
|
| 366 |
+
{
|
| 367 |
+
"epoch": 2.1,
|
| 368 |
+
"grad_norm": 14.67716121673584,
|
| 369 |
+
"learning_rate": 7.898089171974524e-06,
|
| 370 |
+
"loss": 0.7684,
|
| 371 |
+
"step": 660
|
| 372 |
+
},
|
| 373 |
+
{
|
| 374 |
+
"epoch": 2.15,
|
| 375 |
+
"grad_norm": 18.161645889282227,
|
| 376 |
+
"learning_rate": 7.85031847133758e-06,
|
| 377 |
+
"loss": 0.7923,
|
| 378 |
+
"step": 675
|
| 379 |
+
},
|
| 380 |
+
{
|
| 381 |
+
"epoch": 2.2,
|
| 382 |
+
"grad_norm": 11.349693298339844,
|
| 383 |
+
"learning_rate": 7.802547770700637e-06,
|
| 384 |
+
"loss": 0.7427,
|
| 385 |
+
"step": 690
|
| 386 |
+
},
|
| 387 |
+
{
|
| 388 |
+
"epoch": 2.23,
|
| 389 |
+
"eval_accuracy": 0.7352380952380952,
|
| 390 |
+
"eval_loss": 0.7270055413246155,
|
| 391 |
+
"eval_runtime": 13.6582,
|
| 392 |
+
"eval_samples_per_second": 76.877,
|
| 393 |
+
"eval_steps_per_second": 9.665,
|
| 394 |
+
"step": 700
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 2.25,
|
| 398 |
+
"grad_norm": 16.357065200805664,
|
| 399 |
+
"learning_rate": 7.754777070063695e-06,
|
| 400 |
+
"loss": 0.7645,
|
| 401 |
+
"step": 705
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 2.29,
|
| 405 |
+
"grad_norm": 15.481508255004883,
|
| 406 |
+
"learning_rate": 7.707006369426753e-06,
|
| 407 |
+
"loss": 0.7973,
|
| 408 |
+
"step": 720
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 2.34,
|
| 412 |
+
"grad_norm": 16.786632537841797,
|
| 413 |
+
"learning_rate": 7.659235668789809e-06,
|
| 414 |
+
"loss": 0.8256,
|
| 415 |
+
"step": 735
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 2.39,
|
| 419 |
+
"grad_norm": 11.738802909851074,
|
| 420 |
+
"learning_rate": 7.611464968152867e-06,
|
| 421 |
+
"loss": 0.8307,
|
| 422 |
+
"step": 750
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 2.44,
|
| 426 |
+
"grad_norm": 13.264825820922852,
|
| 427 |
+
"learning_rate": 7.563694267515924e-06,
|
| 428 |
+
"loss": 0.7431,
|
| 429 |
+
"step": 765
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 2.48,
|
| 433 |
+
"grad_norm": 15.430547714233398,
|
| 434 |
+
"learning_rate": 7.515923566878982e-06,
|
| 435 |
+
"loss": 0.7867,
|
| 436 |
+
"step": 780
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 2.53,
|
| 440 |
+
"grad_norm": 15.555388450622559,
|
| 441 |
+
"learning_rate": 7.468152866242039e-06,
|
| 442 |
+
"loss": 0.693,
|
| 443 |
+
"step": 795
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 2.55,
|
| 447 |
+
"eval_accuracy": 0.7676190476190476,
|
| 448 |
+
"eval_loss": 0.6801217198371887,
|
| 449 |
+
"eval_runtime": 13.4462,
|
| 450 |
+
"eval_samples_per_second": 78.089,
|
| 451 |
+
"eval_steps_per_second": 9.817,
|
| 452 |
+
"step": 800
|
| 453 |
+
},
|
| 454 |
+
{
|
| 455 |
+
"epoch": 2.58,
|
| 456 |
+
"grad_norm": 16.23971176147461,
|
| 457 |
+
"learning_rate": 7.4203821656050955e-06,
|
| 458 |
+
"loss": 0.7791,
|
| 459 |
+
"step": 810
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 2.63,
|
| 463 |
+
"grad_norm": 22.803543090820312,
|
| 464 |
+
"learning_rate": 7.372611464968153e-06,
|
| 465 |
+
"loss": 0.7564,
|
| 466 |
+
"step": 825
|
| 467 |
+
},
|
| 468 |
+
{
|
| 469 |
+
"epoch": 2.68,
|
| 470 |
+
"grad_norm": 15.14857292175293,
|
| 471 |
+
"learning_rate": 7.32484076433121e-06,
|
| 472 |
+
"loss": 0.6895,
|
| 473 |
+
"step": 840
|
| 474 |
+
},
|
| 475 |
+
{
|
| 476 |
+
"epoch": 2.72,
|
| 477 |
+
"grad_norm": 18.52122688293457,
|
| 478 |
+
"learning_rate": 7.2770700636942685e-06,
|
| 479 |
+
"loss": 0.7016,
|
| 480 |
+
"step": 855
|
| 481 |
+
},
|
| 482 |
+
{
|
| 483 |
+
"epoch": 2.77,
|
| 484 |
+
"grad_norm": 11.38332748413086,
|
| 485 |
+
"learning_rate": 7.2292993630573245e-06,
|
| 486 |
+
"loss": 0.8174,
|
| 487 |
+
"step": 870
|
| 488 |
+
},
|
| 489 |
+
{
|
| 490 |
+
"epoch": 2.82,
|
| 491 |
+
"grad_norm": 22.539424896240234,
|
| 492 |
+
"learning_rate": 7.181528662420383e-06,
|
| 493 |
+
"loss": 0.7147,
|
| 494 |
+
"step": 885
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 2.87,
|
| 498 |
+
"grad_norm": 15.064950942993164,
|
| 499 |
+
"learning_rate": 7.13375796178344e-06,
|
| 500 |
+
"loss": 0.7789,
|
| 501 |
+
"step": 900
|
| 502 |
+
},
|
| 503 |
+
{
|
| 504 |
+
"epoch": 2.87,
|
| 505 |
+
"eval_accuracy": 0.7590476190476191,
|
| 506 |
+
"eval_loss": 0.6831705570220947,
|
| 507 |
+
"eval_runtime": 13.5379,
|
| 508 |
+
"eval_samples_per_second": 77.56,
|
| 509 |
+
"eval_steps_per_second": 9.75,
|
| 510 |
+
"step": 900
|
| 511 |
+
},
|
| 512 |
+
{
|
| 513 |
+
"epoch": 2.91,
|
| 514 |
+
"grad_norm": 17.286598205566406,
|
| 515 |
+
"learning_rate": 7.085987261146498e-06,
|
| 516 |
+
"loss": 0.6743,
|
| 517 |
+
"step": 915
|
| 518 |
+
},
|
| 519 |
+
{
|
| 520 |
+
"epoch": 2.96,
|
| 521 |
+
"grad_norm": 19.45290756225586,
|
| 522 |
+
"learning_rate": 7.0382165605095544e-06,
|
| 523 |
+
"loss": 0.6263,
|
| 524 |
+
"step": 930
|
| 525 |
+
},
|
| 526 |
+
{
|
| 527 |
+
"epoch": 3.01,
|
| 528 |
+
"grad_norm": 26.688581466674805,
|
| 529 |
+
"learning_rate": 6.990445859872612e-06,
|
| 530 |
+
"loss": 0.7736,
|
| 531 |
+
"step": 945
|
| 532 |
+
},
|
| 533 |
+
{
|
| 534 |
+
"epoch": 3.06,
|
| 535 |
+
"grad_norm": 22.443763732910156,
|
| 536 |
+
"learning_rate": 6.942675159235669e-06,
|
| 537 |
+
"loss": 0.6392,
|
| 538 |
+
"step": 960
|
| 539 |
+
},
|
| 540 |
+
{
|
| 541 |
+
"epoch": 3.11,
|
| 542 |
+
"grad_norm": 18.81976318359375,
|
| 543 |
+
"learning_rate": 6.894904458598727e-06,
|
| 544 |
+
"loss": 0.7653,
|
| 545 |
+
"step": 975
|
| 546 |
+
},
|
| 547 |
+
{
|
| 548 |
+
"epoch": 3.15,
|
| 549 |
+
"grad_norm": 12.312933921813965,
|
| 550 |
+
"learning_rate": 6.8471337579617835e-06,
|
| 551 |
+
"loss": 0.6863,
|
| 552 |
+
"step": 990
|
| 553 |
+
},
|
| 554 |
+
{
|
| 555 |
+
"epoch": 3.18,
|
| 556 |
+
"eval_accuracy": 0.7752380952380953,
|
| 557 |
+
"eval_loss": 0.665543794631958,
|
| 558 |
+
"eval_runtime": 13.4956,
|
| 559 |
+
"eval_samples_per_second": 77.803,
|
| 560 |
+
"eval_steps_per_second": 9.781,
|
| 561 |
+
"step": 1000
|
| 562 |
+
},
|
| 563 |
+
{
|
| 564 |
+
"epoch": 3.2,
|
| 565 |
+
"grad_norm": 15.371318817138672,
|
| 566 |
+
"learning_rate": 6.799363057324841e-06,
|
| 567 |
+
"loss": 0.7106,
|
| 568 |
+
"step": 1005
|
| 569 |
+
},
|
| 570 |
+
{
|
| 571 |
+
"epoch": 3.25,
|
| 572 |
+
"grad_norm": 18.258623123168945,
|
| 573 |
+
"learning_rate": 6.751592356687898e-06,
|
| 574 |
+
"loss": 0.7305,
|
| 575 |
+
"step": 1020
|
| 576 |
+
},
|
| 577 |
+
{
|
| 578 |
+
"epoch": 3.3,
|
| 579 |
+
"grad_norm": 16.52337074279785,
|
| 580 |
+
"learning_rate": 6.7038216560509565e-06,
|
| 581 |
+
"loss": 0.6947,
|
| 582 |
+
"step": 1035
|
| 583 |
+
},
|
| 584 |
+
{
|
| 585 |
+
"epoch": 3.34,
|
| 586 |
+
"grad_norm": 18.67824363708496,
|
| 587 |
+
"learning_rate": 6.6560509554140125e-06,
|
| 588 |
+
"loss": 0.6669,
|
| 589 |
+
"step": 1050
|
| 590 |
+
},
|
| 591 |
+
{
|
| 592 |
+
"epoch": 3.39,
|
| 593 |
+
"grad_norm": 16.26685905456543,
|
| 594 |
+
"learning_rate": 6.608280254777071e-06,
|
| 595 |
+
"loss": 0.6801,
|
| 596 |
+
"step": 1065
|
| 597 |
+
},
|
| 598 |
+
{
|
| 599 |
+
"epoch": 3.44,
|
| 600 |
+
"grad_norm": 13.744972229003906,
|
| 601 |
+
"learning_rate": 6.560509554140128e-06,
|
| 602 |
+
"loss": 0.6035,
|
| 603 |
+
"step": 1080
|
| 604 |
+
},
|
| 605 |
+
{
|
| 606 |
+
"epoch": 3.49,
|
| 607 |
+
"grad_norm": 12.479057312011719,
|
| 608 |
+
"learning_rate": 6.5127388535031856e-06,
|
| 609 |
+
"loss": 0.6437,
|
| 610 |
+
"step": 1095
|
| 611 |
+
},
|
| 612 |
+
{
|
| 613 |
+
"epoch": 3.5,
|
| 614 |
+
"eval_accuracy": 0.7771428571428571,
|
| 615 |
+
"eval_loss": 0.6382023692131042,
|
| 616 |
+
"eval_runtime": 13.3473,
|
| 617 |
+
"eval_samples_per_second": 78.667,
|
| 618 |
+
"eval_steps_per_second": 9.89,
|
| 619 |
+
"step": 1100
|
| 620 |
+
},
|
| 621 |
+
{
|
| 622 |
+
"epoch": 3.54,
|
| 623 |
+
"grad_norm": 14.826581954956055,
|
| 624 |
+
"learning_rate": 6.464968152866242e-06,
|
| 625 |
+
"loss": 0.7309,
|
| 626 |
+
"step": 1110
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 3.58,
|
| 630 |
+
"grad_norm": 12.955341339111328,
|
| 631 |
+
"learning_rate": 6.4171974522293e-06,
|
| 632 |
+
"loss": 0.6864,
|
| 633 |
+
"step": 1125
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 3.63,
|
| 637 |
+
"grad_norm": 14.903204917907715,
|
| 638 |
+
"learning_rate": 6.369426751592357e-06,
|
| 639 |
+
"loss": 0.6711,
|
| 640 |
+
"step": 1140
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 3.68,
|
| 644 |
+
"grad_norm": 15.349693298339844,
|
| 645 |
+
"learning_rate": 6.321656050955415e-06,
|
| 646 |
+
"loss": 0.6362,
|
| 647 |
+
"step": 1155
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 3.73,
|
| 651 |
+
"grad_norm": 25.346343994140625,
|
| 652 |
+
"learning_rate": 6.2738853503184715e-06,
|
| 653 |
+
"loss": 0.6359,
|
| 654 |
+
"step": 1170
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 3.77,
|
| 658 |
+
"grad_norm": 12.536116600036621,
|
| 659 |
+
"learning_rate": 6.226114649681529e-06,
|
| 660 |
+
"loss": 0.6991,
|
| 661 |
+
"step": 1185
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 3.82,
|
| 665 |
+
"grad_norm": 19.788801193237305,
|
| 666 |
+
"learning_rate": 6.178343949044586e-06,
|
| 667 |
+
"loss": 0.6741,
|
| 668 |
+
"step": 1200
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 3.82,
|
| 672 |
+
"eval_accuracy": 0.7790476190476191,
|
| 673 |
+
"eval_loss": 0.6445861458778381,
|
| 674 |
+
"eval_runtime": 13.6114,
|
| 675 |
+
"eval_samples_per_second": 77.141,
|
| 676 |
+
"eval_steps_per_second": 9.698,
|
| 677 |
+
"step": 1200
|
| 678 |
+
},
|
| 679 |
+
{
|
| 680 |
+
"epoch": 3.87,
|
| 681 |
+
"grad_norm": 16.279836654663086,
|
| 682 |
+
"learning_rate": 6.1305732484076445e-06,
|
| 683 |
+
"loss": 0.6977,
|
| 684 |
+
"step": 1215
|
| 685 |
+
},
|
| 686 |
+
{
|
| 687 |
+
"epoch": 3.92,
|
| 688 |
+
"grad_norm": 18.798139572143555,
|
| 689 |
+
"learning_rate": 6.0828025477707005e-06,
|
| 690 |
+
"loss": 0.653,
|
| 691 |
+
"step": 1230
|
| 692 |
+
},
|
| 693 |
+
{
|
| 694 |
+
"epoch": 3.96,
|
| 695 |
+
"grad_norm": 17.142087936401367,
|
| 696 |
+
"learning_rate": 6.035031847133759e-06,
|
| 697 |
+
"loss": 0.6778,
|
| 698 |
+
"step": 1245
|
| 699 |
+
},
|
| 700 |
+
{
|
| 701 |
+
"epoch": 4.01,
|
| 702 |
+
"grad_norm": 17.632762908935547,
|
| 703 |
+
"learning_rate": 5.987261146496816e-06,
|
| 704 |
+
"loss": 0.5343,
|
| 705 |
+
"step": 1260
|
| 706 |
+
},
|
| 707 |
+
{
|
| 708 |
+
"epoch": 4.06,
|
| 709 |
+
"grad_norm": 14.896882057189941,
|
| 710 |
+
"learning_rate": 5.9394904458598736e-06,
|
| 711 |
+
"loss": 0.5694,
|
| 712 |
+
"step": 1275
|
| 713 |
+
},
|
| 714 |
+
{
|
| 715 |
+
"epoch": 4.11,
|
| 716 |
+
"grad_norm": 19.62409019470215,
|
| 717 |
+
"learning_rate": 5.89171974522293e-06,
|
| 718 |
+
"loss": 0.5871,
|
| 719 |
+
"step": 1290
|
| 720 |
+
},
|
| 721 |
+
{
|
| 722 |
+
"epoch": 4.14,
|
| 723 |
+
"eval_accuracy": 0.7838095238095238,
|
| 724 |
+
"eval_loss": 0.6551438570022583,
|
| 725 |
+
"eval_runtime": 13.8432,
|
| 726 |
+
"eval_samples_per_second": 75.849,
|
| 727 |
+
"eval_steps_per_second": 9.535,
|
| 728 |
+
"step": 1300
|
| 729 |
+
},
|
| 730 |
+
{
|
| 731 |
+
"epoch": 4.16,
|
| 732 |
+
"grad_norm": 16.879796981811523,
|
| 733 |
+
"learning_rate": 5.843949044585988e-06,
|
| 734 |
+
"loss": 0.6344,
|
| 735 |
+
"step": 1305
|
| 736 |
+
},
|
| 737 |
+
{
|
| 738 |
+
"epoch": 4.2,
|
| 739 |
+
"grad_norm": 18.603700637817383,
|
| 740 |
+
"learning_rate": 5.796178343949045e-06,
|
| 741 |
+
"loss": 0.5768,
|
| 742 |
+
"step": 1320
|
| 743 |
+
},
|
| 744 |
+
{
|
| 745 |
+
"epoch": 4.25,
|
| 746 |
+
"grad_norm": 16.433502197265625,
|
| 747 |
+
"learning_rate": 5.748407643312103e-06,
|
| 748 |
+
"loss": 0.5884,
|
| 749 |
+
"step": 1335
|
| 750 |
+
},
|
| 751 |
+
{
|
| 752 |
+
"epoch": 4.3,
|
| 753 |
+
"grad_norm": 17.979280471801758,
|
| 754 |
+
"learning_rate": 5.7006369426751594e-06,
|
| 755 |
+
"loss": 0.6167,
|
| 756 |
+
"step": 1350
|
| 757 |
+
},
|
| 758 |
+
{
|
| 759 |
+
"epoch": 4.35,
|
| 760 |
+
"grad_norm": 20.778549194335938,
|
| 761 |
+
"learning_rate": 5.652866242038217e-06,
|
| 762 |
+
"loss": 0.6594,
|
| 763 |
+
"step": 1365
|
| 764 |
+
},
|
| 765 |
+
{
|
| 766 |
+
"epoch": 4.39,
|
| 767 |
+
"grad_norm": 14.834670066833496,
|
| 768 |
+
"learning_rate": 5.605095541401274e-06,
|
| 769 |
+
"loss": 0.6214,
|
| 770 |
+
"step": 1380
|
| 771 |
+
},
|
| 772 |
+
{
|
| 773 |
+
"epoch": 4.44,
|
| 774 |
+
"grad_norm": 19.214466094970703,
|
| 775 |
+
"learning_rate": 5.5573248407643325e-06,
|
| 776 |
+
"loss": 0.6051,
|
| 777 |
+
"step": 1395
|
| 778 |
+
},
|
| 779 |
+
{
|
| 780 |
+
"epoch": 4.46,
|
| 781 |
+
"eval_accuracy": 0.7638095238095238,
|
| 782 |
+
"eval_loss": 0.6970483660697937,
|
| 783 |
+
"eval_runtime": 20.5305,
|
| 784 |
+
"eval_samples_per_second": 51.143,
|
| 785 |
+
"eval_steps_per_second": 6.429,
|
| 786 |
+
"step": 1400
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 4.49,
|
| 790 |
+
"grad_norm": 16.332500457763672,
|
| 791 |
+
"learning_rate": 5.5095541401273885e-06,
|
| 792 |
+
"loss": 0.5996,
|
| 793 |
+
"step": 1410
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 4.54,
|
| 797 |
+
"grad_norm": 16.794343948364258,
|
| 798 |
+
"learning_rate": 5.461783439490447e-06,
|
| 799 |
+
"loss": 0.702,
|
| 800 |
+
"step": 1425
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 4.59,
|
| 804 |
+
"grad_norm": 21.159442901611328,
|
| 805 |
+
"learning_rate": 5.414012738853504e-06,
|
| 806 |
+
"loss": 0.5742,
|
| 807 |
+
"step": 1440
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 4.63,
|
| 811 |
+
"grad_norm": 26.400766372680664,
|
| 812 |
+
"learning_rate": 5.3662420382165615e-06,
|
| 813 |
+
"loss": 0.6288,
|
| 814 |
+
"step": 1455
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 4.68,
|
| 818 |
+
"grad_norm": 19.17631721496582,
|
| 819 |
+
"learning_rate": 5.318471337579618e-06,
|
| 820 |
+
"loss": 0.5819,
|
| 821 |
+
"step": 1470
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 4.73,
|
| 825 |
+
"grad_norm": 18.10342025756836,
|
| 826 |
+
"learning_rate": 5.270700636942676e-06,
|
| 827 |
+
"loss": 0.5842,
|
| 828 |
+
"step": 1485
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 4.78,
|
| 832 |
+
"grad_norm": 21.941911697387695,
|
| 833 |
+
"learning_rate": 5.222929936305733e-06,
|
| 834 |
+
"loss": 0.5175,
|
| 835 |
+
"step": 1500
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 4.78,
|
| 839 |
+
"eval_accuracy": 0.7790476190476191,
|
| 840 |
+
"eval_loss": 0.6552723050117493,
|
| 841 |
+
"eval_runtime": 13.5024,
|
| 842 |
+
"eval_samples_per_second": 77.764,
|
| 843 |
+
"eval_steps_per_second": 9.776,
|
| 844 |
+
"step": 1500
|
| 845 |
+
},
|
| 846 |
+
{
|
| 847 |
+
"epoch": 4.82,
|
| 848 |
+
"grad_norm": 24.317623138427734,
|
| 849 |
+
"learning_rate": 5.175159235668791e-06,
|
| 850 |
+
"loss": 0.5984,
|
| 851 |
+
"step": 1515
|
| 852 |
+
},
|
| 853 |
+
{
|
| 854 |
+
"epoch": 4.87,
|
| 855 |
+
"grad_norm": 14.877484321594238,
|
| 856 |
+
"learning_rate": 5.1273885350318474e-06,
|
| 857 |
+
"loss": 0.6142,
|
| 858 |
+
"step": 1530
|
| 859 |
+
},
|
| 860 |
+
{
|
| 861 |
+
"epoch": 4.92,
|
| 862 |
+
"grad_norm": 20.296701431274414,
|
| 863 |
+
"learning_rate": 5.079617834394905e-06,
|
| 864 |
+
"loss": 0.719,
|
| 865 |
+
"step": 1545
|
| 866 |
+
},
|
| 867 |
+
{
|
| 868 |
+
"epoch": 4.97,
|
| 869 |
+
"grad_norm": 20.335296630859375,
|
| 870 |
+
"learning_rate": 5.031847133757962e-06,
|
| 871 |
+
"loss": 0.5651,
|
| 872 |
+
"step": 1560
|
| 873 |
+
},
|
| 874 |
+
{
|
| 875 |
+
"epoch": 5.02,
|
| 876 |
+
"grad_norm": 17.09543228149414,
|
| 877 |
+
"learning_rate": 4.98407643312102e-06,
|
| 878 |
+
"loss": 0.4632,
|
| 879 |
+
"step": 1575
|
| 880 |
+
},
|
| 881 |
+
{
|
| 882 |
+
"epoch": 5.06,
|
| 883 |
+
"grad_norm": 15.416642189025879,
|
| 884 |
+
"learning_rate": 4.9363057324840765e-06,
|
| 885 |
+
"loss": 0.5795,
|
| 886 |
+
"step": 1590
|
| 887 |
+
},
|
| 888 |
+
{
|
| 889 |
+
"epoch": 5.1,
|
| 890 |
+
"eval_accuracy": 0.7771428571428571,
|
| 891 |
+
"eval_loss": 0.6666560173034668,
|
| 892 |
+
"eval_runtime": 14.1067,
|
| 893 |
+
"eval_samples_per_second": 74.433,
|
| 894 |
+
"eval_steps_per_second": 9.357,
|
| 895 |
+
"step": 1600
|
| 896 |
+
},
|
| 897 |
+
{
|
| 898 |
+
"epoch": 5.11,
|
| 899 |
+
"grad_norm": 12.152099609375,
|
| 900 |
+
"learning_rate": 4.888535031847134e-06,
|
| 901 |
+
"loss": 0.6119,
|
| 902 |
+
"step": 1605
|
| 903 |
+
},
|
| 904 |
+
{
|
| 905 |
+
"epoch": 5.16,
|
| 906 |
+
"grad_norm": 11.709696769714355,
|
| 907 |
+
"learning_rate": 4.840764331210192e-06,
|
| 908 |
+
"loss": 0.5521,
|
| 909 |
+
"step": 1620
|
| 910 |
+
},
|
| 911 |
+
{
|
| 912 |
+
"epoch": 5.21,
|
| 913 |
+
"grad_norm": 12.4248685836792,
|
| 914 |
+
"learning_rate": 4.792993630573249e-06,
|
| 915 |
+
"loss": 0.586,
|
| 916 |
+
"step": 1635
|
| 917 |
+
},
|
| 918 |
+
{
|
| 919 |
+
"epoch": 5.25,
|
| 920 |
+
"grad_norm": 22.69182777404785,
|
| 921 |
+
"learning_rate": 4.745222929936306e-06,
|
| 922 |
+
"loss": 0.5848,
|
| 923 |
+
"step": 1650
|
| 924 |
+
},
|
| 925 |
+
{
|
| 926 |
+
"epoch": 5.3,
|
| 927 |
+
"grad_norm": 15.92928409576416,
|
| 928 |
+
"learning_rate": 4.697452229299363e-06,
|
| 929 |
+
"loss": 0.5922,
|
| 930 |
+
"step": 1665
|
| 931 |
+
},
|
| 932 |
+
{
|
| 933 |
+
"epoch": 5.35,
|
| 934 |
+
"grad_norm": 25.377580642700195,
|
| 935 |
+
"learning_rate": 4.649681528662421e-06,
|
| 936 |
+
"loss": 0.6579,
|
| 937 |
+
"step": 1680
|
| 938 |
+
},
|
| 939 |
+
{
|
| 940 |
+
"epoch": 5.4,
|
| 941 |
+
"grad_norm": 12.89096450805664,
|
| 942 |
+
"learning_rate": 4.601910828025479e-06,
|
| 943 |
+
"loss": 0.4919,
|
| 944 |
+
"step": 1695
|
| 945 |
+
},
|
| 946 |
+
{
|
| 947 |
+
"epoch": 5.41,
|
| 948 |
+
"eval_accuracy": 0.7904761904761904,
|
| 949 |
+
"eval_loss": 0.6316953897476196,
|
| 950 |
+
"eval_runtime": 13.547,
|
| 951 |
+
"eval_samples_per_second": 77.508,
|
| 952 |
+
"eval_steps_per_second": 9.744,
|
| 953 |
+
"step": 1700
|
| 954 |
+
},
|
| 955 |
+
{
|
| 956 |
+
"epoch": 5.45,
|
| 957 |
+
"grad_norm": 13.04831314086914,
|
| 958 |
+
"learning_rate": 4.554140127388535e-06,
|
| 959 |
+
"loss": 0.5459,
|
| 960 |
+
"step": 1710
|
| 961 |
+
},
|
| 962 |
+
{
|
| 963 |
+
"epoch": 5.49,
|
| 964 |
+
"grad_norm": 14.792088508605957,
|
| 965 |
+
"learning_rate": 4.506369426751593e-06,
|
| 966 |
+
"loss": 0.4729,
|
| 967 |
+
"step": 1725
|
| 968 |
+
},
|
| 969 |
+
{
|
| 970 |
+
"epoch": 5.54,
|
| 971 |
+
"grad_norm": 20.434284210205078,
|
| 972 |
+
"learning_rate": 4.45859872611465e-06,
|
| 973 |
+
"loss": 0.5285,
|
| 974 |
+
"step": 1740
|
| 975 |
+
},
|
| 976 |
+
{
|
| 977 |
+
"epoch": 5.59,
|
| 978 |
+
"grad_norm": 16.0216064453125,
|
| 979 |
+
"learning_rate": 4.410828025477708e-06,
|
| 980 |
+
"loss": 0.5891,
|
| 981 |
+
"step": 1755
|
| 982 |
+
},
|
| 983 |
+
{
|
| 984 |
+
"epoch": 5.64,
|
| 985 |
+
"grad_norm": 14.537184715270996,
|
| 986 |
+
"learning_rate": 4.3630573248407645e-06,
|
| 987 |
+
"loss": 0.6203,
|
| 988 |
+
"step": 1770
|
| 989 |
+
},
|
| 990 |
+
{
|
| 991 |
+
"epoch": 5.68,
|
| 992 |
+
"grad_norm": 16.755977630615234,
|
| 993 |
+
"learning_rate": 4.315286624203822e-06,
|
| 994 |
+
"loss": 0.5832,
|
| 995 |
+
"step": 1785
|
| 996 |
+
},
|
| 997 |
+
{
|
| 998 |
+
"epoch": 5.73,
|
| 999 |
+
"grad_norm": 18.05998992919922,
|
| 1000 |
+
"learning_rate": 4.26751592356688e-06,
|
| 1001 |
+
"loss": 0.4986,
|
| 1002 |
+
"step": 1800
|
| 1003 |
+
},
|
| 1004 |
+
{
|
| 1005 |
+
"epoch": 5.73,
|
| 1006 |
+
"eval_accuracy": 0.780952380952381,
|
| 1007 |
+
"eval_loss": 0.6485886573791504,
|
| 1008 |
+
"eval_runtime": 13.712,
|
| 1009 |
+
"eval_samples_per_second": 76.575,
|
| 1010 |
+
"eval_steps_per_second": 9.627,
|
| 1011 |
+
"step": 1800
|
| 1012 |
+
},
|
| 1013 |
+
{
|
| 1014 |
+
"epoch": 5.78,
|
| 1015 |
+
"grad_norm": 13.940254211425781,
|
| 1016 |
+
"learning_rate": 4.219745222929937e-06,
|
| 1017 |
+
"loss": 0.5582,
|
| 1018 |
+
"step": 1815
|
| 1019 |
+
},
|
| 1020 |
+
{
|
| 1021 |
+
"epoch": 5.83,
|
| 1022 |
+
"grad_norm": 13.54953670501709,
|
| 1023 |
+
"learning_rate": 4.171974522292994e-06,
|
| 1024 |
+
"loss": 0.5189,
|
| 1025 |
+
"step": 1830
|
| 1026 |
+
},
|
| 1027 |
+
{
|
| 1028 |
+
"epoch": 5.88,
|
| 1029 |
+
"grad_norm": 19.552183151245117,
|
| 1030 |
+
"learning_rate": 4.124203821656051e-06,
|
| 1031 |
+
"loss": 0.6037,
|
| 1032 |
+
"step": 1845
|
| 1033 |
+
},
|
| 1034 |
+
{
|
| 1035 |
+
"epoch": 5.92,
|
| 1036 |
+
"grad_norm": 13.757224082946777,
|
| 1037 |
+
"learning_rate": 4.076433121019109e-06,
|
| 1038 |
+
"loss": 0.5537,
|
| 1039 |
+
"step": 1860
|
| 1040 |
+
},
|
| 1041 |
+
{
|
| 1042 |
+
"epoch": 5.97,
|
| 1043 |
+
"grad_norm": 24.593406677246094,
|
| 1044 |
+
"learning_rate": 4.0286624203821666e-06,
|
| 1045 |
+
"loss": 0.5527,
|
| 1046 |
+
"step": 1875
|
| 1047 |
+
},
|
| 1048 |
+
{
|
| 1049 |
+
"epoch": 6.02,
|
| 1050 |
+
"grad_norm": 22.236400604248047,
|
| 1051 |
+
"learning_rate": 3.980891719745223e-06,
|
| 1052 |
+
"loss": 0.5104,
|
| 1053 |
+
"step": 1890
|
| 1054 |
+
},
|
| 1055 |
+
{
|
| 1056 |
+
"epoch": 6.05,
|
| 1057 |
+
"eval_accuracy": 0.7742857142857142,
|
| 1058 |
+
"eval_loss": 0.6699539422988892,
|
| 1059 |
+
"eval_runtime": 13.5651,
|
| 1060 |
+
"eval_samples_per_second": 77.405,
|
| 1061 |
+
"eval_steps_per_second": 9.731,
|
| 1062 |
+
"step": 1900
|
| 1063 |
+
},
|
| 1064 |
+
{
|
| 1065 |
+
"epoch": 6.07,
|
| 1066 |
+
"grad_norm": 15.87308120727539,
|
| 1067 |
+
"learning_rate": 3.933121019108281e-06,
|
| 1068 |
+
"loss": 0.5268,
|
| 1069 |
+
"step": 1905
|
| 1070 |
+
},
|
| 1071 |
+
{
|
| 1072 |
+
"epoch": 6.11,
|
| 1073 |
+
"grad_norm": 13.48481273651123,
|
| 1074 |
+
"learning_rate": 3.885350318471338e-06,
|
| 1075 |
+
"loss": 0.5421,
|
| 1076 |
+
"step": 1920
|
| 1077 |
+
},
|
| 1078 |
+
{
|
| 1079 |
+
"epoch": 6.16,
|
| 1080 |
+
"grad_norm": 13.895825386047363,
|
| 1081 |
+
"learning_rate": 3.837579617834396e-06,
|
| 1082 |
+
"loss": 0.6139,
|
| 1083 |
+
"step": 1935
|
| 1084 |
+
},
|
| 1085 |
+
{
|
| 1086 |
+
"epoch": 6.21,
|
| 1087 |
+
"grad_norm": 14.655675888061523,
|
| 1088 |
+
"learning_rate": 3.789808917197453e-06,
|
| 1089 |
+
"loss": 0.495,
|
| 1090 |
+
"step": 1950
|
| 1091 |
+
},
|
| 1092 |
+
{
|
| 1093 |
+
"epoch": 6.26,
|
| 1094 |
+
"grad_norm": 21.782032012939453,
|
| 1095 |
+
"learning_rate": 3.7420382165605097e-06,
|
| 1096 |
+
"loss": 0.513,
|
| 1097 |
+
"step": 1965
|
| 1098 |
+
},
|
| 1099 |
+
{
|
| 1100 |
+
"epoch": 6.31,
|
| 1101 |
+
"grad_norm": 16.350772857666016,
|
| 1102 |
+
"learning_rate": 3.694267515923567e-06,
|
| 1103 |
+
"loss": 0.5182,
|
| 1104 |
+
"step": 1980
|
| 1105 |
+
},
|
| 1106 |
+
{
|
| 1107 |
+
"epoch": 6.35,
|
| 1108 |
+
"grad_norm": 12.87532901763916,
|
| 1109 |
+
"learning_rate": 3.6464968152866242e-06,
|
| 1110 |
+
"loss": 0.4919,
|
| 1111 |
+
"step": 1995
|
| 1112 |
+
},
|
| 1113 |
+
{
|
| 1114 |
+
"epoch": 6.37,
|
| 1115 |
+
"eval_accuracy": 0.7819047619047619,
|
| 1116 |
+
"eval_loss": 0.6527658700942993,
|
| 1117 |
+
"eval_runtime": 13.9166,
|
| 1118 |
+
"eval_samples_per_second": 75.449,
|
| 1119 |
+
"eval_steps_per_second": 9.485,
|
| 1120 |
+
"step": 2000
|
| 1121 |
+
},
|
| 1122 |
+
{
|
| 1123 |
+
"epoch": 6.4,
|
| 1124 |
+
"grad_norm": 12.642027854919434,
|
| 1125 |
+
"learning_rate": 3.5987261146496815e-06,
|
| 1126 |
+
"loss": 0.5212,
|
| 1127 |
+
"step": 2010
|
| 1128 |
+
},
|
| 1129 |
+
{
|
| 1130 |
+
"epoch": 6.45,
|
| 1131 |
+
"grad_norm": 13.786490440368652,
|
| 1132 |
+
"learning_rate": 3.5509554140127388e-06,
|
| 1133 |
+
"loss": 0.5004,
|
| 1134 |
+
"step": 2025
|
| 1135 |
+
},
|
| 1136 |
+
{
|
| 1137 |
+
"epoch": 6.5,
|
| 1138 |
+
"grad_norm": 28.24700927734375,
|
| 1139 |
+
"learning_rate": 3.5031847133757964e-06,
|
| 1140 |
+
"loss": 0.539,
|
| 1141 |
+
"step": 2040
|
| 1142 |
+
},
|
| 1143 |
+
{
|
| 1144 |
+
"epoch": 6.54,
|
| 1145 |
+
"grad_norm": 10.891915321350098,
|
| 1146 |
+
"learning_rate": 3.4554140127388537e-06,
|
| 1147 |
+
"loss": 0.5316,
|
| 1148 |
+
"step": 2055
|
| 1149 |
+
},
|
| 1150 |
+
{
|
| 1151 |
+
"epoch": 6.59,
|
| 1152 |
+
"grad_norm": 21.343164443969727,
|
| 1153 |
+
"learning_rate": 3.407643312101911e-06,
|
| 1154 |
+
"loss": 0.5497,
|
| 1155 |
+
"step": 2070
|
| 1156 |
+
},
|
| 1157 |
+
{
|
| 1158 |
+
"epoch": 6.64,
|
| 1159 |
+
"grad_norm": 15.246662139892578,
|
| 1160 |
+
"learning_rate": 3.3598726114649682e-06,
|
| 1161 |
+
"loss": 0.5212,
|
| 1162 |
+
"step": 2085
|
| 1163 |
+
},
|
| 1164 |
+
{
|
| 1165 |
+
"epoch": 6.69,
|
| 1166 |
+
"grad_norm": 18.424856185913086,
|
| 1167 |
+
"learning_rate": 3.3121019108280255e-06,
|
| 1168 |
+
"loss": 0.5144,
|
| 1169 |
+
"step": 2100
|
| 1170 |
+
},
|
| 1171 |
+
{
|
| 1172 |
+
"epoch": 6.69,
|
| 1173 |
+
"eval_accuracy": 0.7876190476190477,
|
| 1174 |
+
"eval_loss": 0.6354712843894958,
|
| 1175 |
+
"eval_runtime": 13.6508,
|
| 1176 |
+
"eval_samples_per_second": 76.919,
|
| 1177 |
+
"eval_steps_per_second": 9.67,
|
| 1178 |
+
"step": 2100
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 6.74,
|
| 1182 |
+
"grad_norm": 15.570305824279785,
|
| 1183 |
+
"learning_rate": 3.2643312101910827e-06,
|
| 1184 |
+
"loss": 0.5892,
|
| 1185 |
+
"step": 2115
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 6.78,
|
| 1189 |
+
"grad_norm": 16.673995971679688,
|
| 1190 |
+
"learning_rate": 3.2165605095541404e-06,
|
| 1191 |
+
"loss": 0.5079,
|
| 1192 |
+
"step": 2130
|
| 1193 |
+
},
|
| 1194 |
+
{
|
| 1195 |
+
"epoch": 6.83,
|
| 1196 |
+
"grad_norm": 17.703060150146484,
|
| 1197 |
+
"learning_rate": 3.1687898089171977e-06,
|
| 1198 |
+
"loss": 0.496,
|
| 1199 |
+
"step": 2145
|
| 1200 |
+
},
|
| 1201 |
+
{
|
| 1202 |
+
"epoch": 6.88,
|
| 1203 |
+
"grad_norm": 14.203299522399902,
|
| 1204 |
+
"learning_rate": 3.121019108280255e-06,
|
| 1205 |
+
"loss": 0.5223,
|
| 1206 |
+
"step": 2160
|
| 1207 |
+
},
|
| 1208 |
+
{
|
| 1209 |
+
"epoch": 6.93,
|
| 1210 |
+
"grad_norm": 14.10352897644043,
|
| 1211 |
+
"learning_rate": 3.0732484076433122e-06,
|
| 1212 |
+
"loss": 0.521,
|
| 1213 |
+
"step": 2175
|
| 1214 |
+
},
|
| 1215 |
+
{
|
| 1216 |
+
"epoch": 6.97,
|
| 1217 |
+
"grad_norm": 13.882482528686523,
|
| 1218 |
+
"learning_rate": 3.0254777070063695e-06,
|
| 1219 |
+
"loss": 0.5554,
|
| 1220 |
+
"step": 2190
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 7.01,
|
| 1224 |
+
"eval_accuracy": 0.7771428571428571,
|
| 1225 |
+
"eval_loss": 0.6552413702011108,
|
| 1226 |
+
"eval_runtime": 13.0208,
|
| 1227 |
+
"eval_samples_per_second": 80.64,
|
| 1228 |
+
"eval_steps_per_second": 10.138,
|
| 1229 |
+
"step": 2200
|
| 1230 |
+
},
|
| 1231 |
+
{
|
| 1232 |
+
"epoch": 7.02,
|
| 1233 |
+
"grad_norm": 12.480643272399902,
|
| 1234 |
+
"learning_rate": 2.9777070063694267e-06,
|
| 1235 |
+
"loss": 0.4216,
|
| 1236 |
+
"step": 2205
|
| 1237 |
+
},
|
| 1238 |
+
{
|
| 1239 |
+
"epoch": 7.07,
|
| 1240 |
+
"grad_norm": 14.39759349822998,
|
| 1241 |
+
"learning_rate": 2.9299363057324844e-06,
|
| 1242 |
+
"loss": 0.4888,
|
| 1243 |
+
"step": 2220
|
| 1244 |
+
},
|
| 1245 |
+
{
|
| 1246 |
+
"epoch": 7.12,
|
| 1247 |
+
"grad_norm": 17.724123001098633,
|
| 1248 |
+
"learning_rate": 2.8821656050955417e-06,
|
| 1249 |
+
"loss": 0.4579,
|
| 1250 |
+
"step": 2235
|
| 1251 |
+
},
|
| 1252 |
+
{
|
| 1253 |
+
"epoch": 7.17,
|
| 1254 |
+
"grad_norm": 14.149361610412598,
|
| 1255 |
+
"learning_rate": 2.834394904458599e-06,
|
| 1256 |
+
"loss": 0.5295,
|
| 1257 |
+
"step": 2250
|
| 1258 |
+
},
|
| 1259 |
+
{
|
| 1260 |
+
"epoch": 7.21,
|
| 1261 |
+
"grad_norm": 18.39142608642578,
|
| 1262 |
+
"learning_rate": 2.786624203821656e-06,
|
| 1263 |
+
"loss": 0.4918,
|
| 1264 |
+
"step": 2265
|
| 1265 |
+
},
|
| 1266 |
+
{
|
| 1267 |
+
"epoch": 7.26,
|
| 1268 |
+
"grad_norm": 21.38290023803711,
|
| 1269 |
+
"learning_rate": 2.7388535031847135e-06,
|
| 1270 |
+
"loss": 0.5542,
|
| 1271 |
+
"step": 2280
|
| 1272 |
+
},
|
| 1273 |
+
{
|
| 1274 |
+
"epoch": 7.31,
|
| 1275 |
+
"grad_norm": 21.44352912902832,
|
| 1276 |
+
"learning_rate": 2.6910828025477707e-06,
|
| 1277 |
+
"loss": 0.5389,
|
| 1278 |
+
"step": 2295
|
| 1279 |
+
},
|
| 1280 |
+
{
|
| 1281 |
+
"epoch": 7.32,
|
| 1282 |
+
"eval_accuracy": 0.7876190476190477,
|
| 1283 |
+
"eval_loss": 0.6360692381858826,
|
| 1284 |
+
"eval_runtime": 13.6127,
|
| 1285 |
+
"eval_samples_per_second": 77.134,
|
| 1286 |
+
"eval_steps_per_second": 9.697,
|
| 1287 |
+
"step": 2300
|
| 1288 |
+
},
|
| 1289 |
+
{
|
| 1290 |
+
"epoch": 7.36,
|
| 1291 |
+
"grad_norm": 14.326496124267578,
|
| 1292 |
+
"learning_rate": 2.6433121019108284e-06,
|
| 1293 |
+
"loss": 0.4785,
|
| 1294 |
+
"step": 2310
|
| 1295 |
+
},
|
| 1296 |
+
{
|
| 1297 |
+
"epoch": 7.4,
|
| 1298 |
+
"grad_norm": 17.98026466369629,
|
| 1299 |
+
"learning_rate": 2.5955414012738857e-06,
|
| 1300 |
+
"loss": 0.5193,
|
| 1301 |
+
"step": 2325
|
| 1302 |
+
},
|
| 1303 |
+
{
|
| 1304 |
+
"epoch": 7.45,
|
| 1305 |
+
"grad_norm": 11.728538513183594,
|
| 1306 |
+
"learning_rate": 2.547770700636943e-06,
|
| 1307 |
+
"loss": 0.4371,
|
| 1308 |
+
"step": 2340
|
| 1309 |
+
},
|
| 1310 |
+
{
|
| 1311 |
+
"epoch": 7.5,
|
| 1312 |
+
"grad_norm": 17.007251739501953,
|
| 1313 |
+
"learning_rate": 2.5e-06,
|
| 1314 |
+
"loss": 0.4969,
|
| 1315 |
+
"step": 2355
|
| 1316 |
+
},
|
| 1317 |
+
{
|
| 1318 |
+
"epoch": 7.55,
|
| 1319 |
+
"grad_norm": 15.3156099319458,
|
| 1320 |
+
"learning_rate": 2.4522292993630575e-06,
|
| 1321 |
+
"loss": 0.5297,
|
| 1322 |
+
"step": 2370
|
| 1323 |
+
},
|
| 1324 |
+
{
|
| 1325 |
+
"epoch": 7.6,
|
| 1326 |
+
"grad_norm": 13.694135665893555,
|
| 1327 |
+
"learning_rate": 2.4044585987261147e-06,
|
| 1328 |
+
"loss": 0.4651,
|
| 1329 |
+
"step": 2385
|
| 1330 |
+
},
|
| 1331 |
+
{
|
| 1332 |
+
"epoch": 7.64,
|
| 1333 |
+
"grad_norm": 16.395017623901367,
|
| 1334 |
+
"learning_rate": 2.356687898089172e-06,
|
| 1335 |
+
"loss": 0.5751,
|
| 1336 |
+
"step": 2400
|
| 1337 |
+
},
|
| 1338 |
+
{
|
| 1339 |
+
"epoch": 7.64,
|
| 1340 |
+
"eval_accuracy": 0.7904761904761904,
|
| 1341 |
+
"eval_loss": 0.6376513838768005,
|
| 1342 |
+
"eval_runtime": 13.653,
|
| 1343 |
+
"eval_samples_per_second": 76.906,
|
| 1344 |
+
"eval_steps_per_second": 9.668,
|
| 1345 |
+
"step": 2400
|
| 1346 |
+
},
|
| 1347 |
+
{
|
| 1348 |
+
"epoch": 7.69,
|
| 1349 |
+
"grad_norm": 21.47723388671875,
|
| 1350 |
+
"learning_rate": 2.3089171974522297e-06,
|
| 1351 |
+
"loss": 0.4762,
|
| 1352 |
+
"step": 2415
|
| 1353 |
+
},
|
| 1354 |
+
{
|
| 1355 |
+
"epoch": 7.74,
|
| 1356 |
+
"grad_norm": 17.56719970703125,
|
| 1357 |
+
"learning_rate": 2.261146496815287e-06,
|
| 1358 |
+
"loss": 0.6217,
|
| 1359 |
+
"step": 2430
|
| 1360 |
+
},
|
| 1361 |
+
{
|
| 1362 |
+
"epoch": 7.79,
|
| 1363 |
+
"grad_norm": 12.036867141723633,
|
| 1364 |
+
"learning_rate": 2.213375796178344e-06,
|
| 1365 |
+
"loss": 0.4727,
|
| 1366 |
+
"step": 2445
|
| 1367 |
+
},
|
| 1368 |
+
{
|
| 1369 |
+
"epoch": 7.83,
|
| 1370 |
+
"grad_norm": 18.971595764160156,
|
| 1371 |
+
"learning_rate": 2.1656050955414015e-06,
|
| 1372 |
+
"loss": 0.4601,
|
| 1373 |
+
"step": 2460
|
| 1374 |
+
},
|
| 1375 |
+
{
|
| 1376 |
+
"epoch": 7.88,
|
| 1377 |
+
"grad_norm": 18.308382034301758,
|
| 1378 |
+
"learning_rate": 2.1178343949044587e-06,
|
| 1379 |
+
"loss": 0.5086,
|
| 1380 |
+
"step": 2475
|
| 1381 |
+
},
|
| 1382 |
+
{
|
| 1383 |
+
"epoch": 7.93,
|
| 1384 |
+
"grad_norm": 13.486546516418457,
|
| 1385 |
+
"learning_rate": 2.070063694267516e-06,
|
| 1386 |
+
"loss": 0.4743,
|
| 1387 |
+
"step": 2490
|
| 1388 |
+
},
|
| 1389 |
+
{
|
| 1390 |
+
"epoch": 7.96,
|
| 1391 |
+
"eval_accuracy": 0.7866666666666666,
|
| 1392 |
+
"eval_loss": 0.6417487859725952,
|
| 1393 |
+
"eval_runtime": 13.707,
|
| 1394 |
+
"eval_samples_per_second": 76.603,
|
| 1395 |
+
"eval_steps_per_second": 9.63,
|
| 1396 |
+
"step": 2500
|
| 1397 |
+
},
|
| 1398 |
+
{
|
| 1399 |
+
"epoch": 7.98,
|
| 1400 |
+
"grad_norm": 12.4083833694458,
|
| 1401 |
+
"learning_rate": 2.0222929936305737e-06,
|
| 1402 |
+
"loss": 0.453,
|
| 1403 |
+
"step": 2505
|
| 1404 |
+
},
|
| 1405 |
+
{
|
| 1406 |
+
"epoch": 8.03,
|
| 1407 |
+
"grad_norm": 10.967087745666504,
|
| 1408 |
+
"learning_rate": 1.974522292993631e-06,
|
| 1409 |
+
"loss": 0.4937,
|
| 1410 |
+
"step": 2520
|
| 1411 |
+
},
|
| 1412 |
+
{
|
| 1413 |
+
"epoch": 8.07,
|
| 1414 |
+
"grad_norm": 11.663314819335938,
|
| 1415 |
+
"learning_rate": 1.926751592356688e-06,
|
| 1416 |
+
"loss": 0.5115,
|
| 1417 |
+
"step": 2535
|
| 1418 |
+
},
|
| 1419 |
+
{
|
| 1420 |
+
"epoch": 8.12,
|
| 1421 |
+
"grad_norm": 10.820151329040527,
|
| 1422 |
+
"learning_rate": 1.8789808917197455e-06,
|
| 1423 |
+
"loss": 0.4281,
|
| 1424 |
+
"step": 2550
|
| 1425 |
+
},
|
| 1426 |
+
{
|
| 1427 |
+
"epoch": 8.17,
|
| 1428 |
+
"grad_norm": 15.378673553466797,
|
| 1429 |
+
"learning_rate": 1.8312101910828025e-06,
|
| 1430 |
+
"loss": 0.5341,
|
| 1431 |
+
"step": 2565
|
| 1432 |
+
},
|
| 1433 |
+
{
|
| 1434 |
+
"epoch": 8.22,
|
| 1435 |
+
"grad_norm": 18.212982177734375,
|
| 1436 |
+
"learning_rate": 1.78343949044586e-06,
|
| 1437 |
+
"loss": 0.5331,
|
| 1438 |
+
"step": 2580
|
| 1439 |
+
},
|
| 1440 |
+
{
|
| 1441 |
+
"epoch": 8.26,
|
| 1442 |
+
"grad_norm": 25.97978401184082,
|
| 1443 |
+
"learning_rate": 1.7356687898089172e-06,
|
| 1444 |
+
"loss": 0.4519,
|
| 1445 |
+
"step": 2595
|
| 1446 |
+
},
|
| 1447 |
+
{
|
| 1448 |
+
"epoch": 8.28,
|
| 1449 |
+
"eval_accuracy": 0.7895238095238095,
|
| 1450 |
+
"eval_loss": 0.6309406161308289,
|
| 1451 |
+
"eval_runtime": 13.8301,
|
| 1452 |
+
"eval_samples_per_second": 75.921,
|
| 1453 |
+
"eval_steps_per_second": 9.544,
|
| 1454 |
+
"step": 2600
|
| 1455 |
+
},
|
| 1456 |
+
{
|
| 1457 |
+
"epoch": 8.31,
|
| 1458 |
+
"grad_norm": 16.035921096801758,
|
| 1459 |
+
"learning_rate": 1.6878980891719745e-06,
|
| 1460 |
+
"loss": 0.4434,
|
| 1461 |
+
"step": 2610
|
| 1462 |
+
},
|
| 1463 |
+
{
|
| 1464 |
+
"epoch": 8.36,
|
| 1465 |
+
"grad_norm": 33.498626708984375,
|
| 1466 |
+
"learning_rate": 1.640127388535032e-06,
|
| 1467 |
+
"loss": 0.5621,
|
| 1468 |
+
"step": 2625
|
| 1469 |
+
},
|
| 1470 |
+
{
|
| 1471 |
+
"epoch": 8.41,
|
| 1472 |
+
"grad_norm": 22.457271575927734,
|
| 1473 |
+
"learning_rate": 1.5923566878980892e-06,
|
| 1474 |
+
"loss": 0.4843,
|
| 1475 |
+
"step": 2640
|
| 1476 |
+
},
|
| 1477 |
+
{
|
| 1478 |
+
"epoch": 8.46,
|
| 1479 |
+
"grad_norm": 20.040433883666992,
|
| 1480 |
+
"learning_rate": 1.5445859872611465e-06,
|
| 1481 |
+
"loss": 0.5306,
|
| 1482 |
+
"step": 2655
|
| 1483 |
+
},
|
| 1484 |
+
{
|
| 1485 |
+
"epoch": 8.5,
|
| 1486 |
+
"grad_norm": 25.173227310180664,
|
| 1487 |
+
"learning_rate": 1.496815286624204e-06,
|
| 1488 |
+
"loss": 0.5078,
|
| 1489 |
+
"step": 2670
|
| 1490 |
+
},
|
| 1491 |
+
{
|
| 1492 |
+
"epoch": 8.55,
|
| 1493 |
+
"grad_norm": 14.426128387451172,
|
| 1494 |
+
"learning_rate": 1.4490445859872612e-06,
|
| 1495 |
+
"loss": 0.6197,
|
| 1496 |
+
"step": 2685
|
| 1497 |
+
},
|
| 1498 |
+
{
|
| 1499 |
+
"epoch": 8.6,
|
| 1500 |
+
"grad_norm": 21.540132522583008,
|
| 1501 |
+
"learning_rate": 1.4012738853503185e-06,
|
| 1502 |
+
"loss": 0.5058,
|
| 1503 |
+
"step": 2700
|
| 1504 |
+
},
|
| 1505 |
+
{
|
| 1506 |
+
"epoch": 8.6,
|
| 1507 |
+
"eval_accuracy": 0.7866666666666666,
|
| 1508 |
+
"eval_loss": 0.6453108787536621,
|
| 1509 |
+
"eval_runtime": 13.9306,
|
| 1510 |
+
"eval_samples_per_second": 75.373,
|
| 1511 |
+
"eval_steps_per_second": 9.476,
|
| 1512 |
+
"step": 2700
|
| 1513 |
+
},
|
| 1514 |
+
{
|
| 1515 |
+
"epoch": 8.65,
|
| 1516 |
+
"grad_norm": 17.211627960205078,
|
| 1517 |
+
"learning_rate": 1.353503184713376e-06,
|
| 1518 |
+
"loss": 0.6111,
|
| 1519 |
+
"step": 2715
|
| 1520 |
+
},
|
| 1521 |
+
{
|
| 1522 |
+
"epoch": 8.69,
|
| 1523 |
+
"grad_norm": 12.991823196411133,
|
| 1524 |
+
"learning_rate": 1.3057324840764332e-06,
|
| 1525 |
+
"loss": 0.4392,
|
| 1526 |
+
"step": 2730
|
| 1527 |
+
},
|
| 1528 |
+
{
|
| 1529 |
+
"epoch": 8.74,
|
| 1530 |
+
"grad_norm": 10.215910911560059,
|
| 1531 |
+
"learning_rate": 1.2579617834394905e-06,
|
| 1532 |
+
"loss": 0.5505,
|
| 1533 |
+
"step": 2745
|
| 1534 |
+
},
|
| 1535 |
+
{
|
| 1536 |
+
"epoch": 8.79,
|
| 1537 |
+
"grad_norm": 16.872520446777344,
|
| 1538 |
+
"learning_rate": 1.210191082802548e-06,
|
| 1539 |
+
"loss": 0.4304,
|
| 1540 |
+
"step": 2760
|
| 1541 |
+
},
|
| 1542 |
+
{
|
| 1543 |
+
"epoch": 8.84,
|
| 1544 |
+
"grad_norm": 14.673178672790527,
|
| 1545 |
+
"learning_rate": 1.1624203821656052e-06,
|
| 1546 |
+
"loss": 0.4886,
|
| 1547 |
+
"step": 2775
|
| 1548 |
+
},
|
| 1549 |
+
{
|
| 1550 |
+
"epoch": 8.89,
|
| 1551 |
+
"grad_norm": 17.11809730529785,
|
| 1552 |
+
"learning_rate": 1.1146496815286625e-06,
|
| 1553 |
+
"loss": 0.4754,
|
| 1554 |
+
"step": 2790
|
| 1555 |
+
},
|
| 1556 |
+
{
|
| 1557 |
+
"epoch": 8.92,
|
| 1558 |
+
"eval_accuracy": 0.7904761904761904,
|
| 1559 |
+
"eval_loss": 0.6414105296134949,
|
| 1560 |
+
"eval_runtime": 13.6725,
|
| 1561 |
+
"eval_samples_per_second": 76.797,
|
| 1562 |
+
"eval_steps_per_second": 9.654,
|
| 1563 |
+
"step": 2800
|
| 1564 |
+
},
|
| 1565 |
+
{
|
| 1566 |
+
"epoch": 8.93,
|
| 1567 |
+
"grad_norm": 10.194275856018066,
|
| 1568 |
+
"learning_rate": 1.06687898089172e-06,
|
| 1569 |
+
"loss": 0.5041,
|
| 1570 |
+
"step": 2805
|
| 1571 |
+
},
|
| 1572 |
+
{
|
| 1573 |
+
"epoch": 8.98,
|
| 1574 |
+
"grad_norm": 24.478836059570312,
|
| 1575 |
+
"learning_rate": 1.0191082802547772e-06,
|
| 1576 |
+
"loss": 0.5016,
|
| 1577 |
+
"step": 2820
|
| 1578 |
+
},
|
| 1579 |
+
{
|
| 1580 |
+
"epoch": 9.03,
|
| 1581 |
+
"grad_norm": 16.150724411010742,
|
| 1582 |
+
"learning_rate": 9.713375796178345e-07,
|
| 1583 |
+
"loss": 0.486,
|
| 1584 |
+
"step": 2835
|
| 1585 |
+
},
|
| 1586 |
+
{
|
| 1587 |
+
"epoch": 9.08,
|
| 1588 |
+
"grad_norm": 15.390515327453613,
|
| 1589 |
+
"learning_rate": 9.235668789808917e-07,
|
| 1590 |
+
"loss": 0.5024,
|
| 1591 |
+
"step": 2850
|
| 1592 |
+
},
|
| 1593 |
+
{
|
| 1594 |
+
"epoch": 9.12,
|
| 1595 |
+
"grad_norm": 20.307998657226562,
|
| 1596 |
+
"learning_rate": 8.757961783439491e-07,
|
| 1597 |
+
"loss": 0.4982,
|
| 1598 |
+
"step": 2865
|
| 1599 |
+
},
|
| 1600 |
+
{
|
| 1601 |
+
"epoch": 9.17,
|
| 1602 |
+
"grad_norm": 18.18573760986328,
|
| 1603 |
+
"learning_rate": 8.280254777070064e-07,
|
| 1604 |
+
"loss": 0.5054,
|
| 1605 |
+
"step": 2880
|
| 1606 |
+
},
|
| 1607 |
+
{
|
| 1608 |
+
"epoch": 9.22,
|
| 1609 |
+
"grad_norm": 15.57632827758789,
|
| 1610 |
+
"learning_rate": 7.802547770700637e-07,
|
| 1611 |
+
"loss": 0.4637,
|
| 1612 |
+
"step": 2895
|
| 1613 |
+
},
|
| 1614 |
+
{
|
| 1615 |
+
"epoch": 9.24,
|
| 1616 |
+
"eval_accuracy": 0.7904761904761904,
|
| 1617 |
+
"eval_loss": 0.6329751014709473,
|
| 1618 |
+
"eval_runtime": 13.6535,
|
| 1619 |
+
"eval_samples_per_second": 76.904,
|
| 1620 |
+
"eval_steps_per_second": 9.668,
|
| 1621 |
+
"step": 2900
|
| 1622 |
+
},
|
| 1623 |
+
{
|
| 1624 |
+
"epoch": 9.27,
|
| 1625 |
+
"grad_norm": 11.808470726013184,
|
| 1626 |
+
"learning_rate": 7.324840764331211e-07,
|
| 1627 |
+
"loss": 0.5804,
|
| 1628 |
+
"step": 2910
|
| 1629 |
+
},
|
| 1630 |
+
{
|
| 1631 |
+
"epoch": 9.32,
|
| 1632 |
+
"grad_norm": 19.781538009643555,
|
| 1633 |
+
"learning_rate": 6.847133757961784e-07,
|
| 1634 |
+
"loss": 0.5309,
|
| 1635 |
+
"step": 2925
|
| 1636 |
+
},
|
| 1637 |
+
{
|
| 1638 |
+
"epoch": 9.36,
|
| 1639 |
+
"grad_norm": 15.966341972351074,
|
| 1640 |
+
"learning_rate": 6.369426751592357e-07,
|
| 1641 |
+
"loss": 0.536,
|
| 1642 |
+
"step": 2940
|
| 1643 |
+
},
|
| 1644 |
+
{
|
| 1645 |
+
"epoch": 9.41,
|
| 1646 |
+
"grad_norm": 11.989510536193848,
|
| 1647 |
+
"learning_rate": 5.89171974522293e-07,
|
| 1648 |
+
"loss": 0.4474,
|
| 1649 |
+
"step": 2955
|
| 1650 |
+
},
|
| 1651 |
+
{
|
| 1652 |
+
"epoch": 9.46,
|
| 1653 |
+
"grad_norm": 13.803847312927246,
|
| 1654 |
+
"learning_rate": 5.414012738853504e-07,
|
| 1655 |
+
"loss": 0.4868,
|
| 1656 |
+
"step": 2970
|
| 1657 |
+
},
|
| 1658 |
+
{
|
| 1659 |
+
"epoch": 9.51,
|
| 1660 |
+
"grad_norm": 16.266407012939453,
|
| 1661 |
+
"learning_rate": 4.936305732484077e-07,
|
| 1662 |
+
"loss": 0.4453,
|
| 1663 |
+
"step": 2985
|
| 1664 |
+
},
|
| 1665 |
+
{
|
| 1666 |
+
"epoch": 9.55,
|
| 1667 |
+
"grad_norm": 17.96660804748535,
|
| 1668 |
+
"learning_rate": 4.45859872611465e-07,
|
| 1669 |
+
"loss": 0.5028,
|
| 1670 |
+
"step": 3000
|
| 1671 |
+
},
|
| 1672 |
+
{
|
| 1673 |
+
"epoch": 9.55,
|
| 1674 |
+
"eval_accuracy": 0.7857142857142857,
|
| 1675 |
+
"eval_loss": 0.641762375831604,
|
| 1676 |
+
"eval_runtime": 13.8591,
|
| 1677 |
+
"eval_samples_per_second": 75.763,
|
| 1678 |
+
"eval_steps_per_second": 9.524,
|
| 1679 |
+
"step": 3000
|
| 1680 |
+
},
|
| 1681 |
+
{
|
| 1682 |
+
"epoch": 9.6,
|
| 1683 |
+
"grad_norm": 15.89122200012207,
|
| 1684 |
+
"learning_rate": 3.980891719745223e-07,
|
| 1685 |
+
"loss": 0.4376,
|
| 1686 |
+
"step": 3015
|
| 1687 |
+
},
|
| 1688 |
+
{
|
| 1689 |
+
"epoch": 9.65,
|
| 1690 |
+
"grad_norm": 18.61841583251953,
|
| 1691 |
+
"learning_rate": 3.503184713375796e-07,
|
| 1692 |
+
"loss": 0.3695,
|
| 1693 |
+
"step": 3030
|
| 1694 |
+
},
|
| 1695 |
+
{
|
| 1696 |
+
"epoch": 9.7,
|
| 1697 |
+
"grad_norm": 19.915699005126953,
|
| 1698 |
+
"learning_rate": 3.02547770700637e-07,
|
| 1699 |
+
"loss": 0.4777,
|
| 1700 |
+
"step": 3045
|
| 1701 |
+
},
|
| 1702 |
+
{
|
| 1703 |
+
"epoch": 9.75,
|
| 1704 |
+
"grad_norm": 22.503381729125977,
|
| 1705 |
+
"learning_rate": 2.547770700636943e-07,
|
| 1706 |
+
"loss": 0.6298,
|
| 1707 |
+
"step": 3060
|
| 1708 |
+
},
|
| 1709 |
+
{
|
| 1710 |
+
"epoch": 9.79,
|
| 1711 |
+
"grad_norm": 16.984233856201172,
|
| 1712 |
+
"learning_rate": 2.070063694267516e-07,
|
| 1713 |
+
"loss": 0.4051,
|
| 1714 |
+
"step": 3075
|
| 1715 |
+
},
|
| 1716 |
+
{
|
| 1717 |
+
"epoch": 9.84,
|
| 1718 |
+
"grad_norm": 19.879077911376953,
|
| 1719 |
+
"learning_rate": 1.5923566878980893e-07,
|
| 1720 |
+
"loss": 0.4227,
|
| 1721 |
+
"step": 3090
|
| 1722 |
+
},
|
| 1723 |
+
{
|
| 1724 |
+
"epoch": 9.87,
|
| 1725 |
+
"eval_accuracy": 0.7914285714285715,
|
| 1726 |
+
"eval_loss": 0.6412155628204346,
|
| 1727 |
+
"eval_runtime": 13.7603,
|
| 1728 |
+
"eval_samples_per_second": 76.307,
|
| 1729 |
+
"eval_steps_per_second": 9.593,
|
| 1730 |
+
"step": 3100
|
| 1731 |
+
},
|
| 1732 |
+
{
|
| 1733 |
+
"epoch": 9.89,
|
| 1734 |
+
"grad_norm": 18.370866775512695,
|
| 1735 |
+
"learning_rate": 1.1146496815286625e-07,
|
| 1736 |
+
"loss": 0.4079,
|
| 1737 |
+
"step": 3105
|
| 1738 |
+
},
|
| 1739 |
+
{
|
| 1740 |
+
"epoch": 9.94,
|
| 1741 |
+
"grad_norm": 13.279521942138672,
|
| 1742 |
+
"learning_rate": 6.369426751592358e-08,
|
| 1743 |
+
"loss": 0.4035,
|
| 1744 |
+
"step": 3120
|
| 1745 |
+
},
|
| 1746 |
+
{
|
| 1747 |
+
"epoch": 9.98,
|
| 1748 |
+
"grad_norm": 16.93092155456543,
|
| 1749 |
+
"learning_rate": 1.5923566878980894e-08,
|
| 1750 |
+
"loss": 0.4605,
|
| 1751 |
+
"step": 3135
|
| 1752 |
+
},
|
| 1753 |
+
{
|
| 1754 |
+
"epoch": 10.0,
|
| 1755 |
+
"step": 3140,
|
| 1756 |
+
"total_flos": 3.265548125287219e+18,
|
| 1757 |
+
"train_loss": 0.7155859537944672,
|
| 1758 |
+
"train_runtime": 2626.1064,
|
| 1759 |
+
"train_samples_per_second": 38.212,
|
| 1760 |
+
"train_steps_per_second": 1.196
|
| 1761 |
+
}
|
| 1762 |
+
],
|
| 1763 |
+
"logging_steps": 15,
|
| 1764 |
+
"max_steps": 3140,
|
| 1765 |
+
"num_input_tokens_seen": 0,
|
| 1766 |
+
"num_train_epochs": 10,
|
| 1767 |
+
"save_steps": 100,
|
| 1768 |
+
"total_flos": 3.265548125287219e+18,
|
| 1769 |
+
"train_batch_size": 32,
|
| 1770 |
+
"trial_name": null,
|
| 1771 |
+
"trial_params": null
|
| 1772 |
+
}
|