Commit
•
690b705
1
Parent(s):
94d5eaa
End of training
Browse files- README.md +3 -3
- all_results.json +13 -0
- eval_results.json +8 -0
- runs/May20_14-27-57_b94dbfcd7693/events.out.tfevents.1716217090.b94dbfcd7693.1289.8 +3 -0
- train_results.json +8 -0
- trainer_state.json +1023 -0
README.md
CHANGED
|
@@ -22,7 +22,7 @@ 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
|
|
@@ -32,8 +32,8 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 32 |
|
| 33 |
This model is a fine-tuned version of [microsoft/beit-base-patch16-224](https://huggingface.co/microsoft/beit-base-patch16-224) on the imagefolder dataset.
|
| 34 |
It achieves the following results on the evaluation set:
|
| 35 |
-
- Loss: 0.
|
| 36 |
-
- Accuracy: 0.
|
| 37 |
|
| 38 |
## Model description
|
| 39 |
|
|
|
|
| 22 |
metrics:
|
| 23 |
- name: Accuracy
|
| 24 |
type: accuracy
|
| 25 |
+
value: 0.8227848101265823
|
| 26 |
---
|
| 27 |
|
| 28 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
|
|
| 32 |
|
| 33 |
This model is a fine-tuned version of [microsoft/beit-base-patch16-224](https://huggingface.co/microsoft/beit-base-patch16-224) on the imagefolder dataset.
|
| 34 |
It achieves the following results on the evaluation set:
|
| 35 |
+
- Loss: 0.6919
|
| 36 |
+
- Accuracy: 0.8228
|
| 37 |
|
| 38 |
## Model description
|
| 39 |
|
all_results.json
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 85.71428571428571,
|
| 3 |
+
"eval_accuracy": 0.8227848101265823,
|
| 4 |
+
"eval_loss": 0.6919474005699158,
|
| 5 |
+
"eval_runtime": 1.2565,
|
| 6 |
+
"eval_samples_per_second": 62.873,
|
| 7 |
+
"eval_steps_per_second": 2.388,
|
| 8 |
+
"total_flos": 2.9349165326823014e+18,
|
| 9 |
+
"train_loss": 0.2620420479774475,
|
| 10 |
+
"train_runtime": 1806.3529,
|
| 11 |
+
"train_samples_per_second": 24.469,
|
| 12 |
+
"train_steps_per_second": 0.166
|
| 13 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 85.71428571428571,
|
| 3 |
+
"eval_accuracy": 0.8227848101265823,
|
| 4 |
+
"eval_loss": 0.6919474005699158,
|
| 5 |
+
"eval_runtime": 1.2565,
|
| 6 |
+
"eval_samples_per_second": 62.873,
|
| 7 |
+
"eval_steps_per_second": 2.388
|
| 8 |
+
}
|
runs/May20_14-27-57_b94dbfcd7693/events.out.tfevents.1716217090.b94dbfcd7693.1289.8
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9c101d776f341d8fdc9d282e532f3388fb81d51d7c93115c87638ae3eef32fb4
|
| 3 |
+
size 411
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 85.71428571428571,
|
| 3 |
+
"total_flos": 2.9349165326823014e+18,
|
| 4 |
+
"train_loss": 0.2620420479774475,
|
| 5 |
+
"train_runtime": 1806.3529,
|
| 6 |
+
"train_samples_per_second": 24.469,
|
| 7 |
+
"train_steps_per_second": 0.166
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1023 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.8227848101265823,
|
| 3 |
+
"best_model_checkpoint": "beit-base-patch16-224-fold5/checkpoint-126",
|
| 4 |
+
"epoch": 85.71428571428571,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 300,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.8571428571428571,
|
| 13 |
+
"eval_accuracy": 0.4936708860759494,
|
| 14 |
+
"eval_loss": 0.6963750720024109,
|
| 15 |
+
"eval_runtime": 1.3397,
|
| 16 |
+
"eval_samples_per_second": 58.971,
|
| 17 |
+
"eval_steps_per_second": 2.239,
|
| 18 |
+
"step": 3
|
| 19 |
+
},
|
| 20 |
+
{
|
| 21 |
+
"epoch": 2.0,
|
| 22 |
+
"eval_accuracy": 0.6455696202531646,
|
| 23 |
+
"eval_loss": 0.6335641145706177,
|
| 24 |
+
"eval_runtime": 1.2784,
|
| 25 |
+
"eval_samples_per_second": 61.795,
|
| 26 |
+
"eval_steps_per_second": 2.347,
|
| 27 |
+
"step": 7
|
| 28 |
+
},
|
| 29 |
+
{
|
| 30 |
+
"epoch": 2.857142857142857,
|
| 31 |
+
"grad_norm": 10.374295234680176,
|
| 32 |
+
"learning_rate": 1.6666666666666667e-05,
|
| 33 |
+
"loss": 0.7161,
|
| 34 |
+
"step": 10
|
| 35 |
+
},
|
| 36 |
+
{
|
| 37 |
+
"epoch": 2.857142857142857,
|
| 38 |
+
"eval_accuracy": 0.5063291139240507,
|
| 39 |
+
"eval_loss": 0.6711614727973938,
|
| 40 |
+
"eval_runtime": 1.2465,
|
| 41 |
+
"eval_samples_per_second": 63.375,
|
| 42 |
+
"eval_steps_per_second": 2.407,
|
| 43 |
+
"step": 10
|
| 44 |
+
},
|
| 45 |
+
{
|
| 46 |
+
"epoch": 4.0,
|
| 47 |
+
"eval_accuracy": 0.6835443037974683,
|
| 48 |
+
"eval_loss": 0.5884170532226562,
|
| 49 |
+
"eval_runtime": 1.2485,
|
| 50 |
+
"eval_samples_per_second": 63.276,
|
| 51 |
+
"eval_steps_per_second": 2.403,
|
| 52 |
+
"step": 14
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 4.857142857142857,
|
| 56 |
+
"eval_accuracy": 0.5569620253164557,
|
| 57 |
+
"eval_loss": 0.6325361728668213,
|
| 58 |
+
"eval_runtime": 1.2645,
|
| 59 |
+
"eval_samples_per_second": 62.477,
|
| 60 |
+
"eval_steps_per_second": 2.373,
|
| 61 |
+
"step": 17
|
| 62 |
+
},
|
| 63 |
+
{
|
| 64 |
+
"epoch": 5.714285714285714,
|
| 65 |
+
"grad_norm": 6.489544868469238,
|
| 66 |
+
"learning_rate": 3.3333333333333335e-05,
|
| 67 |
+
"loss": 0.6466,
|
| 68 |
+
"step": 20
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 6.0,
|
| 72 |
+
"eval_accuracy": 0.6075949367088608,
|
| 73 |
+
"eval_loss": 0.6306931376457214,
|
| 74 |
+
"eval_runtime": 1.2522,
|
| 75 |
+
"eval_samples_per_second": 63.087,
|
| 76 |
+
"eval_steps_per_second": 2.396,
|
| 77 |
+
"step": 21
|
| 78 |
+
},
|
| 79 |
+
{
|
| 80 |
+
"epoch": 6.857142857142857,
|
| 81 |
+
"eval_accuracy": 0.6835443037974683,
|
| 82 |
+
"eval_loss": 0.5518103837966919,
|
| 83 |
+
"eval_runtime": 1.2518,
|
| 84 |
+
"eval_samples_per_second": 63.11,
|
| 85 |
+
"eval_steps_per_second": 2.397,
|
| 86 |
+
"step": 24
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 8.0,
|
| 90 |
+
"eval_accuracy": 0.6455696202531646,
|
| 91 |
+
"eval_loss": 0.6855829358100891,
|
| 92 |
+
"eval_runtime": 1.2601,
|
| 93 |
+
"eval_samples_per_second": 62.693,
|
| 94 |
+
"eval_steps_per_second": 2.381,
|
| 95 |
+
"step": 28
|
| 96 |
+
},
|
| 97 |
+
{
|
| 98 |
+
"epoch": 8.571428571428571,
|
| 99 |
+
"grad_norm": 15.54662799835205,
|
| 100 |
+
"learning_rate": 5e-05,
|
| 101 |
+
"loss": 0.5604,
|
| 102 |
+
"step": 30
|
| 103 |
+
},
|
| 104 |
+
{
|
| 105 |
+
"epoch": 8.857142857142858,
|
| 106 |
+
"eval_accuracy": 0.7088607594936709,
|
| 107 |
+
"eval_loss": 0.5678081512451172,
|
| 108 |
+
"eval_runtime": 1.2556,
|
| 109 |
+
"eval_samples_per_second": 62.917,
|
| 110 |
+
"eval_steps_per_second": 2.389,
|
| 111 |
+
"step": 31
|
| 112 |
+
},
|
| 113 |
+
{
|
| 114 |
+
"epoch": 10.0,
|
| 115 |
+
"eval_accuracy": 0.6329113924050633,
|
| 116 |
+
"eval_loss": 0.5764104127883911,
|
| 117 |
+
"eval_runtime": 1.2443,
|
| 118 |
+
"eval_samples_per_second": 63.491,
|
| 119 |
+
"eval_steps_per_second": 2.411,
|
| 120 |
+
"step": 35
|
| 121 |
+
},
|
| 122 |
+
{
|
| 123 |
+
"epoch": 10.857142857142858,
|
| 124 |
+
"eval_accuracy": 0.6455696202531646,
|
| 125 |
+
"eval_loss": 0.6123582124710083,
|
| 126 |
+
"eval_runtime": 1.2354,
|
| 127 |
+
"eval_samples_per_second": 63.945,
|
| 128 |
+
"eval_steps_per_second": 2.428,
|
| 129 |
+
"step": 38
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 11.428571428571429,
|
| 133 |
+
"grad_norm": 7.390775203704834,
|
| 134 |
+
"learning_rate": 4.814814814814815e-05,
|
| 135 |
+
"loss": 0.4941,
|
| 136 |
+
"step": 40
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 12.0,
|
| 140 |
+
"eval_accuracy": 0.5316455696202531,
|
| 141 |
+
"eval_loss": 1.131524920463562,
|
| 142 |
+
"eval_runtime": 1.2833,
|
| 143 |
+
"eval_samples_per_second": 61.562,
|
| 144 |
+
"eval_steps_per_second": 2.338,
|
| 145 |
+
"step": 42
|
| 146 |
+
},
|
| 147 |
+
{
|
| 148 |
+
"epoch": 12.857142857142858,
|
| 149 |
+
"eval_accuracy": 0.6835443037974683,
|
| 150 |
+
"eval_loss": 0.5624261498451233,
|
| 151 |
+
"eval_runtime": 1.2579,
|
| 152 |
+
"eval_samples_per_second": 62.803,
|
| 153 |
+
"eval_steps_per_second": 2.385,
|
| 154 |
+
"step": 45
|
| 155 |
+
},
|
| 156 |
+
{
|
| 157 |
+
"epoch": 14.0,
|
| 158 |
+
"eval_accuracy": 0.6962025316455697,
|
| 159 |
+
"eval_loss": 0.6350968480110168,
|
| 160 |
+
"eval_runtime": 1.2442,
|
| 161 |
+
"eval_samples_per_second": 63.495,
|
| 162 |
+
"eval_steps_per_second": 2.411,
|
| 163 |
+
"step": 49
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 14.285714285714286,
|
| 167 |
+
"grad_norm": 13.477153778076172,
|
| 168 |
+
"learning_rate": 4.62962962962963e-05,
|
| 169 |
+
"loss": 0.4516,
|
| 170 |
+
"step": 50
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 14.857142857142858,
|
| 174 |
+
"eval_accuracy": 0.6835443037974683,
|
| 175 |
+
"eval_loss": 0.6810511350631714,
|
| 176 |
+
"eval_runtime": 1.2437,
|
| 177 |
+
"eval_samples_per_second": 63.519,
|
| 178 |
+
"eval_steps_per_second": 2.412,
|
| 179 |
+
"step": 52
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 16.0,
|
| 183 |
+
"eval_accuracy": 0.6455696202531646,
|
| 184 |
+
"eval_loss": 0.7385033965110779,
|
| 185 |
+
"eval_runtime": 1.2445,
|
| 186 |
+
"eval_samples_per_second": 63.478,
|
| 187 |
+
"eval_steps_per_second": 2.411,
|
| 188 |
+
"step": 56
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 16.857142857142858,
|
| 192 |
+
"eval_accuracy": 0.7215189873417721,
|
| 193 |
+
"eval_loss": 0.6122547388076782,
|
| 194 |
+
"eval_runtime": 1.2613,
|
| 195 |
+
"eval_samples_per_second": 62.632,
|
| 196 |
+
"eval_steps_per_second": 2.378,
|
| 197 |
+
"step": 59
|
| 198 |
+
},
|
| 199 |
+
{
|
| 200 |
+
"epoch": 17.142857142857142,
|
| 201 |
+
"grad_norm": 5.328888893127441,
|
| 202 |
+
"learning_rate": 4.4444444444444447e-05,
|
| 203 |
+
"loss": 0.409,
|
| 204 |
+
"step": 60
|
| 205 |
+
},
|
| 206 |
+
{
|
| 207 |
+
"epoch": 18.0,
|
| 208 |
+
"eval_accuracy": 0.6835443037974683,
|
| 209 |
+
"eval_loss": 0.587742805480957,
|
| 210 |
+
"eval_runtime": 1.2428,
|
| 211 |
+
"eval_samples_per_second": 63.564,
|
| 212 |
+
"eval_steps_per_second": 2.414,
|
| 213 |
+
"step": 63
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 18.857142857142858,
|
| 217 |
+
"eval_accuracy": 0.6962025316455697,
|
| 218 |
+
"eval_loss": 0.6250303387641907,
|
| 219 |
+
"eval_runtime": 1.2444,
|
| 220 |
+
"eval_samples_per_second": 63.486,
|
| 221 |
+
"eval_steps_per_second": 2.411,
|
| 222 |
+
"step": 66
|
| 223 |
+
},
|
| 224 |
+
{
|
| 225 |
+
"epoch": 20.0,
|
| 226 |
+
"grad_norm": 7.51099157333374,
|
| 227 |
+
"learning_rate": 4.259259259259259e-05,
|
| 228 |
+
"loss": 0.3439,
|
| 229 |
+
"step": 70
|
| 230 |
+
},
|
| 231 |
+
{
|
| 232 |
+
"epoch": 20.0,
|
| 233 |
+
"eval_accuracy": 0.7341772151898734,
|
| 234 |
+
"eval_loss": 0.6220384240150452,
|
| 235 |
+
"eval_runtime": 1.2611,
|
| 236 |
+
"eval_samples_per_second": 62.645,
|
| 237 |
+
"eval_steps_per_second": 2.379,
|
| 238 |
+
"step": 70
|
| 239 |
+
},
|
| 240 |
+
{
|
| 241 |
+
"epoch": 20.857142857142858,
|
| 242 |
+
"eval_accuracy": 0.7088607594936709,
|
| 243 |
+
"eval_loss": 0.724984347820282,
|
| 244 |
+
"eval_runtime": 1.237,
|
| 245 |
+
"eval_samples_per_second": 63.863,
|
| 246 |
+
"eval_steps_per_second": 2.425,
|
| 247 |
+
"step": 73
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 22.0,
|
| 251 |
+
"eval_accuracy": 0.7341772151898734,
|
| 252 |
+
"eval_loss": 0.6107321381568909,
|
| 253 |
+
"eval_runtime": 1.2567,
|
| 254 |
+
"eval_samples_per_second": 62.863,
|
| 255 |
+
"eval_steps_per_second": 2.387,
|
| 256 |
+
"step": 77
|
| 257 |
+
},
|
| 258 |
+
{
|
| 259 |
+
"epoch": 22.857142857142858,
|
| 260 |
+
"grad_norm": 6.396337032318115,
|
| 261 |
+
"learning_rate": 4.074074074074074e-05,
|
| 262 |
+
"loss": 0.3268,
|
| 263 |
+
"step": 80
|
| 264 |
+
},
|
| 265 |
+
{
|
| 266 |
+
"epoch": 22.857142857142858,
|
| 267 |
+
"eval_accuracy": 0.7974683544303798,
|
| 268 |
+
"eval_loss": 0.4999098479747772,
|
| 269 |
+
"eval_runtime": 1.2544,
|
| 270 |
+
"eval_samples_per_second": 62.978,
|
| 271 |
+
"eval_steps_per_second": 2.392,
|
| 272 |
+
"step": 80
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 24.0,
|
| 276 |
+
"eval_accuracy": 0.7468354430379747,
|
| 277 |
+
"eval_loss": 0.5325254201889038,
|
| 278 |
+
"eval_runtime": 1.2825,
|
| 279 |
+
"eval_samples_per_second": 61.597,
|
| 280 |
+
"eval_steps_per_second": 2.339,
|
| 281 |
+
"step": 84
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 24.857142857142858,
|
| 285 |
+
"eval_accuracy": 0.7341772151898734,
|
| 286 |
+
"eval_loss": 0.7209397554397583,
|
| 287 |
+
"eval_runtime": 1.2439,
|
| 288 |
+
"eval_samples_per_second": 63.508,
|
| 289 |
+
"eval_steps_per_second": 2.412,
|
| 290 |
+
"step": 87
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 25.714285714285715,
|
| 294 |
+
"grad_norm": 4.736355781555176,
|
| 295 |
+
"learning_rate": 3.888888888888889e-05,
|
| 296 |
+
"loss": 0.2941,
|
| 297 |
+
"step": 90
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 26.0,
|
| 301 |
+
"eval_accuracy": 0.7721518987341772,
|
| 302 |
+
"eval_loss": 0.5556657314300537,
|
| 303 |
+
"eval_runtime": 1.2702,
|
| 304 |
+
"eval_samples_per_second": 62.195,
|
| 305 |
+
"eval_steps_per_second": 2.362,
|
| 306 |
+
"step": 91
|
| 307 |
+
},
|
| 308 |
+
{
|
| 309 |
+
"epoch": 26.857142857142858,
|
| 310 |
+
"eval_accuracy": 0.759493670886076,
|
| 311 |
+
"eval_loss": 0.6655121445655823,
|
| 312 |
+
"eval_runtime": 1.2506,
|
| 313 |
+
"eval_samples_per_second": 63.172,
|
| 314 |
+
"eval_steps_per_second": 2.399,
|
| 315 |
+
"step": 94
|
| 316 |
+
},
|
| 317 |
+
{
|
| 318 |
+
"epoch": 28.0,
|
| 319 |
+
"eval_accuracy": 0.6962025316455697,
|
| 320 |
+
"eval_loss": 1.0775113105773926,
|
| 321 |
+
"eval_runtime": 1.2537,
|
| 322 |
+
"eval_samples_per_second": 63.013,
|
| 323 |
+
"eval_steps_per_second": 2.393,
|
| 324 |
+
"step": 98
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 28.571428571428573,
|
| 328 |
+
"grad_norm": 19.96893310546875,
|
| 329 |
+
"learning_rate": 3.7037037037037037e-05,
|
| 330 |
+
"loss": 0.284,
|
| 331 |
+
"step": 100
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 28.857142857142858,
|
| 335 |
+
"eval_accuracy": 0.759493670886076,
|
| 336 |
+
"eval_loss": 0.6816892027854919,
|
| 337 |
+
"eval_runtime": 1.2538,
|
| 338 |
+
"eval_samples_per_second": 63.006,
|
| 339 |
+
"eval_steps_per_second": 2.393,
|
| 340 |
+
"step": 101
|
| 341 |
+
},
|
| 342 |
+
{
|
| 343 |
+
"epoch": 30.0,
|
| 344 |
+
"eval_accuracy": 0.6835443037974683,
|
| 345 |
+
"eval_loss": 0.9235275983810425,
|
| 346 |
+
"eval_runtime": 1.3158,
|
| 347 |
+
"eval_samples_per_second": 60.038,
|
| 348 |
+
"eval_steps_per_second": 2.28,
|
| 349 |
+
"step": 105
|
| 350 |
+
},
|
| 351 |
+
{
|
| 352 |
+
"epoch": 30.857142857142858,
|
| 353 |
+
"eval_accuracy": 0.759493670886076,
|
| 354 |
+
"eval_loss": 0.6586753129959106,
|
| 355 |
+
"eval_runtime": 1.2811,
|
| 356 |
+
"eval_samples_per_second": 61.664,
|
| 357 |
+
"eval_steps_per_second": 2.342,
|
| 358 |
+
"step": 108
|
| 359 |
+
},
|
| 360 |
+
{
|
| 361 |
+
"epoch": 31.428571428571427,
|
| 362 |
+
"grad_norm": 17.663036346435547,
|
| 363 |
+
"learning_rate": 3.518518518518519e-05,
|
| 364 |
+
"loss": 0.3134,
|
| 365 |
+
"step": 110
|
| 366 |
+
},
|
| 367 |
+
{
|
| 368 |
+
"epoch": 32.0,
|
| 369 |
+
"eval_accuracy": 0.7468354430379747,
|
| 370 |
+
"eval_loss": 0.7085594534873962,
|
| 371 |
+
"eval_runtime": 1.2261,
|
| 372 |
+
"eval_samples_per_second": 64.432,
|
| 373 |
+
"eval_steps_per_second": 2.447,
|
| 374 |
+
"step": 112
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 32.857142857142854,
|
| 378 |
+
"eval_accuracy": 0.7468354430379747,
|
| 379 |
+
"eval_loss": 0.6894867420196533,
|
| 380 |
+
"eval_runtime": 1.2687,
|
| 381 |
+
"eval_samples_per_second": 62.267,
|
| 382 |
+
"eval_steps_per_second": 2.365,
|
| 383 |
+
"step": 115
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"epoch": 34.0,
|
| 387 |
+
"eval_accuracy": 0.7721518987341772,
|
| 388 |
+
"eval_loss": 0.6417700052261353,
|
| 389 |
+
"eval_runtime": 1.2606,
|
| 390 |
+
"eval_samples_per_second": 62.669,
|
| 391 |
+
"eval_steps_per_second": 2.38,
|
| 392 |
+
"step": 119
|
| 393 |
+
},
|
| 394 |
+
{
|
| 395 |
+
"epoch": 34.285714285714285,
|
| 396 |
+
"grad_norm": 5.122420310974121,
|
| 397 |
+
"learning_rate": 3.3333333333333335e-05,
|
| 398 |
+
"loss": 0.2266,
|
| 399 |
+
"step": 120
|
| 400 |
+
},
|
| 401 |
+
{
|
| 402 |
+
"epoch": 34.857142857142854,
|
| 403 |
+
"eval_accuracy": 0.7848101265822784,
|
| 404 |
+
"eval_loss": 0.7007168531417847,
|
| 405 |
+
"eval_runtime": 1.2576,
|
| 406 |
+
"eval_samples_per_second": 62.819,
|
| 407 |
+
"eval_steps_per_second": 2.386,
|
| 408 |
+
"step": 122
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 36.0,
|
| 412 |
+
"eval_accuracy": 0.8227848101265823,
|
| 413 |
+
"eval_loss": 0.6919474005699158,
|
| 414 |
+
"eval_runtime": 1.2393,
|
| 415 |
+
"eval_samples_per_second": 63.744,
|
| 416 |
+
"eval_steps_per_second": 2.421,
|
| 417 |
+
"step": 126
|
| 418 |
+
},
|
| 419 |
+
{
|
| 420 |
+
"epoch": 36.857142857142854,
|
| 421 |
+
"eval_accuracy": 0.7341772151898734,
|
| 422 |
+
"eval_loss": 0.7562373876571655,
|
| 423 |
+
"eval_runtime": 1.249,
|
| 424 |
+
"eval_samples_per_second": 63.253,
|
| 425 |
+
"eval_steps_per_second": 2.402,
|
| 426 |
+
"step": 129
|
| 427 |
+
},
|
| 428 |
+
{
|
| 429 |
+
"epoch": 37.142857142857146,
|
| 430 |
+
"grad_norm": 6.356712341308594,
|
| 431 |
+
"learning_rate": 3.148148148148148e-05,
|
| 432 |
+
"loss": 0.2249,
|
| 433 |
+
"step": 130
|
| 434 |
+
},
|
| 435 |
+
{
|
| 436 |
+
"epoch": 38.0,
|
| 437 |
+
"eval_accuracy": 0.7721518987341772,
|
| 438 |
+
"eval_loss": 0.6774502992630005,
|
| 439 |
+
"eval_runtime": 1.2642,
|
| 440 |
+
"eval_samples_per_second": 62.491,
|
| 441 |
+
"eval_steps_per_second": 2.373,
|
| 442 |
+
"step": 133
|
| 443 |
+
},
|
| 444 |
+
{
|
| 445 |
+
"epoch": 38.857142857142854,
|
| 446 |
+
"eval_accuracy": 0.759493670886076,
|
| 447 |
+
"eval_loss": 0.7787100672721863,
|
| 448 |
+
"eval_runtime": 1.2547,
|
| 449 |
+
"eval_samples_per_second": 62.966,
|
| 450 |
+
"eval_steps_per_second": 2.391,
|
| 451 |
+
"step": 136
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 40.0,
|
| 455 |
+
"grad_norm": 5.0371599197387695,
|
| 456 |
+
"learning_rate": 2.962962962962963e-05,
|
| 457 |
+
"loss": 0.2181,
|
| 458 |
+
"step": 140
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 40.0,
|
| 462 |
+
"eval_accuracy": 0.7721518987341772,
|
| 463 |
+
"eval_loss": 0.7932192087173462,
|
| 464 |
+
"eval_runtime": 1.2659,
|
| 465 |
+
"eval_samples_per_second": 62.404,
|
| 466 |
+
"eval_steps_per_second": 2.37,
|
| 467 |
+
"step": 140
|
| 468 |
+
},
|
| 469 |
+
{
|
| 470 |
+
"epoch": 40.857142857142854,
|
| 471 |
+
"eval_accuracy": 0.759493670886076,
|
| 472 |
+
"eval_loss": 0.933390200138092,
|
| 473 |
+
"eval_runtime": 1.2464,
|
| 474 |
+
"eval_samples_per_second": 63.381,
|
| 475 |
+
"eval_steps_per_second": 2.407,
|
| 476 |
+
"step": 143
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 42.0,
|
| 480 |
+
"eval_accuracy": 0.7468354430379747,
|
| 481 |
+
"eval_loss": 0.822421133518219,
|
| 482 |
+
"eval_runtime": 1.2469,
|
| 483 |
+
"eval_samples_per_second": 63.358,
|
| 484 |
+
"eval_steps_per_second": 2.406,
|
| 485 |
+
"step": 147
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 42.857142857142854,
|
| 489 |
+
"grad_norm": 5.825836181640625,
|
| 490 |
+
"learning_rate": 2.777777777777778e-05,
|
| 491 |
+
"loss": 0.186,
|
| 492 |
+
"step": 150
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 42.857142857142854,
|
| 496 |
+
"eval_accuracy": 0.7721518987341772,
|
| 497 |
+
"eval_loss": 0.8443942666053772,
|
| 498 |
+
"eval_runtime": 1.2708,
|
| 499 |
+
"eval_samples_per_second": 62.166,
|
| 500 |
+
"eval_steps_per_second": 2.361,
|
| 501 |
+
"step": 150
|
| 502 |
+
},
|
| 503 |
+
{
|
| 504 |
+
"epoch": 44.0,
|
| 505 |
+
"eval_accuracy": 0.7721518987341772,
|
| 506 |
+
"eval_loss": 1.0349551439285278,
|
| 507 |
+
"eval_runtime": 1.2484,
|
| 508 |
+
"eval_samples_per_second": 63.281,
|
| 509 |
+
"eval_steps_per_second": 2.403,
|
| 510 |
+
"step": 154
|
| 511 |
+
},
|
| 512 |
+
{
|
| 513 |
+
"epoch": 44.857142857142854,
|
| 514 |
+
"eval_accuracy": 0.7721518987341772,
|
| 515 |
+
"eval_loss": 0.8385674953460693,
|
| 516 |
+
"eval_runtime": 1.253,
|
| 517 |
+
"eval_samples_per_second": 63.049,
|
| 518 |
+
"eval_steps_per_second": 2.394,
|
| 519 |
+
"step": 157
|
| 520 |
+
},
|
| 521 |
+
{
|
| 522 |
+
"epoch": 45.714285714285715,
|
| 523 |
+
"grad_norm": 8.203125,
|
| 524 |
+
"learning_rate": 2.5925925925925925e-05,
|
| 525 |
+
"loss": 0.1882,
|
| 526 |
+
"step": 160
|
| 527 |
+
},
|
| 528 |
+
{
|
| 529 |
+
"epoch": 46.0,
|
| 530 |
+
"eval_accuracy": 0.7721518987341772,
|
| 531 |
+
"eval_loss": 0.8384150266647339,
|
| 532 |
+
"eval_runtime": 1.2347,
|
| 533 |
+
"eval_samples_per_second": 63.983,
|
| 534 |
+
"eval_steps_per_second": 2.43,
|
| 535 |
+
"step": 161
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 46.857142857142854,
|
| 539 |
+
"eval_accuracy": 0.759493670886076,
|
| 540 |
+
"eval_loss": 0.7905216217041016,
|
| 541 |
+
"eval_runtime": 1.2388,
|
| 542 |
+
"eval_samples_per_second": 63.773,
|
| 543 |
+
"eval_steps_per_second": 2.422,
|
| 544 |
+
"step": 164
|
| 545 |
+
},
|
| 546 |
+
{
|
| 547 |
+
"epoch": 48.0,
|
| 548 |
+
"eval_accuracy": 0.7721518987341772,
|
| 549 |
+
"eval_loss": 0.7184091210365295,
|
| 550 |
+
"eval_runtime": 1.2382,
|
| 551 |
+
"eval_samples_per_second": 63.805,
|
| 552 |
+
"eval_steps_per_second": 2.423,
|
| 553 |
+
"step": 168
|
| 554 |
+
},
|
| 555 |
+
{
|
| 556 |
+
"epoch": 48.57142857142857,
|
| 557 |
+
"grad_norm": 6.264881134033203,
|
| 558 |
+
"learning_rate": 2.4074074074074074e-05,
|
| 559 |
+
"loss": 0.1649,
|
| 560 |
+
"step": 170
|
| 561 |
+
},
|
| 562 |
+
{
|
| 563 |
+
"epoch": 48.857142857142854,
|
| 564 |
+
"eval_accuracy": 0.759493670886076,
|
| 565 |
+
"eval_loss": 0.8118771910667419,
|
| 566 |
+
"eval_runtime": 1.2516,
|
| 567 |
+
"eval_samples_per_second": 63.119,
|
| 568 |
+
"eval_steps_per_second": 2.397,
|
| 569 |
+
"step": 171
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 50.0,
|
| 573 |
+
"eval_accuracy": 0.7721518987341772,
|
| 574 |
+
"eval_loss": 0.7171959280967712,
|
| 575 |
+
"eval_runtime": 1.2501,
|
| 576 |
+
"eval_samples_per_second": 63.194,
|
| 577 |
+
"eval_steps_per_second": 2.4,
|
| 578 |
+
"step": 175
|
| 579 |
+
},
|
| 580 |
+
{
|
| 581 |
+
"epoch": 50.857142857142854,
|
| 582 |
+
"eval_accuracy": 0.7721518987341772,
|
| 583 |
+
"eval_loss": 0.9248787760734558,
|
| 584 |
+
"eval_runtime": 1.2392,
|
| 585 |
+
"eval_samples_per_second": 63.752,
|
| 586 |
+
"eval_steps_per_second": 2.421,
|
| 587 |
+
"step": 178
|
| 588 |
+
},
|
| 589 |
+
{
|
| 590 |
+
"epoch": 51.42857142857143,
|
| 591 |
+
"grad_norm": 7.78027868270874,
|
| 592 |
+
"learning_rate": 2.2222222222222223e-05,
|
| 593 |
+
"loss": 0.1847,
|
| 594 |
+
"step": 180
|
| 595 |
+
},
|
| 596 |
+
{
|
| 597 |
+
"epoch": 52.0,
|
| 598 |
+
"eval_accuracy": 0.7721518987341772,
|
| 599 |
+
"eval_loss": 0.8068503737449646,
|
| 600 |
+
"eval_runtime": 1.2411,
|
| 601 |
+
"eval_samples_per_second": 63.654,
|
| 602 |
+
"eval_steps_per_second": 2.417,
|
| 603 |
+
"step": 182
|
| 604 |
+
},
|
| 605 |
+
{
|
| 606 |
+
"epoch": 52.857142857142854,
|
| 607 |
+
"eval_accuracy": 0.7721518987341772,
|
| 608 |
+
"eval_loss": 0.8752752542495728,
|
| 609 |
+
"eval_runtime": 1.2551,
|
| 610 |
+
"eval_samples_per_second": 62.941,
|
| 611 |
+
"eval_steps_per_second": 2.39,
|
| 612 |
+
"step": 185
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 54.0,
|
| 616 |
+
"eval_accuracy": 0.7721518987341772,
|
| 617 |
+
"eval_loss": 0.8822396397590637,
|
| 618 |
+
"eval_runtime": 1.2453,
|
| 619 |
+
"eval_samples_per_second": 63.441,
|
| 620 |
+
"eval_steps_per_second": 2.409,
|
| 621 |
+
"step": 189
|
| 622 |
+
},
|
| 623 |
+
{
|
| 624 |
+
"epoch": 54.285714285714285,
|
| 625 |
+
"grad_norm": 6.302079200744629,
|
| 626 |
+
"learning_rate": 2.037037037037037e-05,
|
| 627 |
+
"loss": 0.1606,
|
| 628 |
+
"step": 190
|
| 629 |
+
},
|
| 630 |
+
{
|
| 631 |
+
"epoch": 54.857142857142854,
|
| 632 |
+
"eval_accuracy": 0.7974683544303798,
|
| 633 |
+
"eval_loss": 0.7480602264404297,
|
| 634 |
+
"eval_runtime": 1.2665,
|
| 635 |
+
"eval_samples_per_second": 62.379,
|
| 636 |
+
"eval_steps_per_second": 2.369,
|
| 637 |
+
"step": 192
|
| 638 |
+
},
|
| 639 |
+
{
|
| 640 |
+
"epoch": 56.0,
|
| 641 |
+
"eval_accuracy": 0.7721518987341772,
|
| 642 |
+
"eval_loss": 0.7299729585647583,
|
| 643 |
+
"eval_runtime": 1.2614,
|
| 644 |
+
"eval_samples_per_second": 62.631,
|
| 645 |
+
"eval_steps_per_second": 2.378,
|
| 646 |
+
"step": 196
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 56.857142857142854,
|
| 650 |
+
"eval_accuracy": 0.810126582278481,
|
| 651 |
+
"eval_loss": 0.732498824596405,
|
| 652 |
+
"eval_runtime": 1.2795,
|
| 653 |
+
"eval_samples_per_second": 61.741,
|
| 654 |
+
"eval_steps_per_second": 2.345,
|
| 655 |
+
"step": 199
|
| 656 |
+
},
|
| 657 |
+
{
|
| 658 |
+
"epoch": 57.142857142857146,
|
| 659 |
+
"grad_norm": 6.085663795471191,
|
| 660 |
+
"learning_rate": 1.8518518518518518e-05,
|
| 661 |
+
"loss": 0.142,
|
| 662 |
+
"step": 200
|
| 663 |
+
},
|
| 664 |
+
{
|
| 665 |
+
"epoch": 58.0,
|
| 666 |
+
"eval_accuracy": 0.8227848101265823,
|
| 667 |
+
"eval_loss": 0.7087646126747131,
|
| 668 |
+
"eval_runtime": 1.2611,
|
| 669 |
+
"eval_samples_per_second": 62.643,
|
| 670 |
+
"eval_steps_per_second": 2.379,
|
| 671 |
+
"step": 203
|
| 672 |
+
},
|
| 673 |
+
{
|
| 674 |
+
"epoch": 58.857142857142854,
|
| 675 |
+
"eval_accuracy": 0.8227848101265823,
|
| 676 |
+
"eval_loss": 0.689956784248352,
|
| 677 |
+
"eval_runtime": 1.2387,
|
| 678 |
+
"eval_samples_per_second": 63.778,
|
| 679 |
+
"eval_steps_per_second": 2.422,
|
| 680 |
+
"step": 206
|
| 681 |
+
},
|
| 682 |
+
{
|
| 683 |
+
"epoch": 60.0,
|
| 684 |
+
"grad_norm": 6.427424430847168,
|
| 685 |
+
"learning_rate": 1.6666666666666667e-05,
|
| 686 |
+
"loss": 0.1546,
|
| 687 |
+
"step": 210
|
| 688 |
+
},
|
| 689 |
+
{
|
| 690 |
+
"epoch": 60.0,
|
| 691 |
+
"eval_accuracy": 0.7721518987341772,
|
| 692 |
+
"eval_loss": 0.8303462266921997,
|
| 693 |
+
"eval_runtime": 1.2444,
|
| 694 |
+
"eval_samples_per_second": 63.486,
|
| 695 |
+
"eval_steps_per_second": 2.411,
|
| 696 |
+
"step": 210
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 60.857142857142854,
|
| 700 |
+
"eval_accuracy": 0.7974683544303798,
|
| 701 |
+
"eval_loss": 0.8218382596969604,
|
| 702 |
+
"eval_runtime": 1.2664,
|
| 703 |
+
"eval_samples_per_second": 62.38,
|
| 704 |
+
"eval_steps_per_second": 2.369,
|
| 705 |
+
"step": 213
|
| 706 |
+
},
|
| 707 |
+
{
|
| 708 |
+
"epoch": 62.0,
|
| 709 |
+
"eval_accuracy": 0.7848101265822784,
|
| 710 |
+
"eval_loss": 0.9628246426582336,
|
| 711 |
+
"eval_runtime": 1.2478,
|
| 712 |
+
"eval_samples_per_second": 63.31,
|
| 713 |
+
"eval_steps_per_second": 2.404,
|
| 714 |
+
"step": 217
|
| 715 |
+
},
|
| 716 |
+
{
|
| 717 |
+
"epoch": 62.857142857142854,
|
| 718 |
+
"grad_norm": 7.047874927520752,
|
| 719 |
+
"learning_rate": 1.4814814814814815e-05,
|
| 720 |
+
"loss": 0.1608,
|
| 721 |
+
"step": 220
|
| 722 |
+
},
|
| 723 |
+
{
|
| 724 |
+
"epoch": 62.857142857142854,
|
| 725 |
+
"eval_accuracy": 0.7848101265822784,
|
| 726 |
+
"eval_loss": 0.9826211929321289,
|
| 727 |
+
"eval_runtime": 1.2439,
|
| 728 |
+
"eval_samples_per_second": 63.51,
|
| 729 |
+
"eval_steps_per_second": 2.412,
|
| 730 |
+
"step": 220
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 64.0,
|
| 734 |
+
"eval_accuracy": 0.759493670886076,
|
| 735 |
+
"eval_loss": 0.7603800892829895,
|
| 736 |
+
"eval_runtime": 1.2499,
|
| 737 |
+
"eval_samples_per_second": 63.204,
|
| 738 |
+
"eval_steps_per_second": 2.4,
|
| 739 |
+
"step": 224
|
| 740 |
+
},
|
| 741 |
+
{
|
| 742 |
+
"epoch": 64.85714285714286,
|
| 743 |
+
"eval_accuracy": 0.7721518987341772,
|
| 744 |
+
"eval_loss": 0.816529393196106,
|
| 745 |
+
"eval_runtime": 1.2553,
|
| 746 |
+
"eval_samples_per_second": 62.935,
|
| 747 |
+
"eval_steps_per_second": 2.39,
|
| 748 |
+
"step": 227
|
| 749 |
+
},
|
| 750 |
+
{
|
| 751 |
+
"epoch": 65.71428571428571,
|
| 752 |
+
"grad_norm": 5.481523513793945,
|
| 753 |
+
"learning_rate": 1.2962962962962962e-05,
|
| 754 |
+
"loss": 0.1418,
|
| 755 |
+
"step": 230
|
| 756 |
+
},
|
| 757 |
+
{
|
| 758 |
+
"epoch": 66.0,
|
| 759 |
+
"eval_accuracy": 0.7721518987341772,
|
| 760 |
+
"eval_loss": 0.8280141353607178,
|
| 761 |
+
"eval_runtime": 1.2471,
|
| 762 |
+
"eval_samples_per_second": 63.346,
|
| 763 |
+
"eval_steps_per_second": 2.406,
|
| 764 |
+
"step": 231
|
| 765 |
+
},
|
| 766 |
+
{
|
| 767 |
+
"epoch": 66.85714285714286,
|
| 768 |
+
"eval_accuracy": 0.7848101265822784,
|
| 769 |
+
"eval_loss": 0.9253725409507751,
|
| 770 |
+
"eval_runtime": 1.2519,
|
| 771 |
+
"eval_samples_per_second": 63.106,
|
| 772 |
+
"eval_steps_per_second": 2.396,
|
| 773 |
+
"step": 234
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 68.0,
|
| 777 |
+
"eval_accuracy": 0.7848101265822784,
|
| 778 |
+
"eval_loss": 0.843929648399353,
|
| 779 |
+
"eval_runtime": 1.2503,
|
| 780 |
+
"eval_samples_per_second": 63.185,
|
| 781 |
+
"eval_steps_per_second": 2.399,
|
| 782 |
+
"step": 238
|
| 783 |
+
},
|
| 784 |
+
{
|
| 785 |
+
"epoch": 68.57142857142857,
|
| 786 |
+
"grad_norm": 7.343584060668945,
|
| 787 |
+
"learning_rate": 1.1111111111111112e-05,
|
| 788 |
+
"loss": 0.1505,
|
| 789 |
+
"step": 240
|
| 790 |
+
},
|
| 791 |
+
{
|
| 792 |
+
"epoch": 68.85714285714286,
|
| 793 |
+
"eval_accuracy": 0.7974683544303798,
|
| 794 |
+
"eval_loss": 0.8095594644546509,
|
| 795 |
+
"eval_runtime": 1.2513,
|
| 796 |
+
"eval_samples_per_second": 63.132,
|
| 797 |
+
"eval_steps_per_second": 2.397,
|
| 798 |
+
"step": 241
|
| 799 |
+
},
|
| 800 |
+
{
|
| 801 |
+
"epoch": 70.0,
|
| 802 |
+
"eval_accuracy": 0.759493670886076,
|
| 803 |
+
"eval_loss": 0.947131872177124,
|
| 804 |
+
"eval_runtime": 1.2591,
|
| 805 |
+
"eval_samples_per_second": 62.745,
|
| 806 |
+
"eval_steps_per_second": 2.383,
|
| 807 |
+
"step": 245
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 70.85714285714286,
|
| 811 |
+
"eval_accuracy": 0.759493670886076,
|
| 812 |
+
"eval_loss": 0.902752161026001,
|
| 813 |
+
"eval_runtime": 1.2601,
|
| 814 |
+
"eval_samples_per_second": 62.695,
|
| 815 |
+
"eval_steps_per_second": 2.381,
|
| 816 |
+
"step": 248
|
| 817 |
+
},
|
| 818 |
+
{
|
| 819 |
+
"epoch": 71.42857142857143,
|
| 820 |
+
"grad_norm": 2.4548463821411133,
|
| 821 |
+
"learning_rate": 9.259259259259259e-06,
|
| 822 |
+
"loss": 0.1211,
|
| 823 |
+
"step": 250
|
| 824 |
+
},
|
| 825 |
+
{
|
| 826 |
+
"epoch": 72.0,
|
| 827 |
+
"eval_accuracy": 0.7974683544303798,
|
| 828 |
+
"eval_loss": 0.7743518352508545,
|
| 829 |
+
"eval_runtime": 1.263,
|
| 830 |
+
"eval_samples_per_second": 62.548,
|
| 831 |
+
"eval_steps_per_second": 2.375,
|
| 832 |
+
"step": 252
|
| 833 |
+
},
|
| 834 |
+
{
|
| 835 |
+
"epoch": 72.85714285714286,
|
| 836 |
+
"eval_accuracy": 0.810126582278481,
|
| 837 |
+
"eval_loss": 0.7755638360977173,
|
| 838 |
+
"eval_runtime": 1.2392,
|
| 839 |
+
"eval_samples_per_second": 63.75,
|
| 840 |
+
"eval_steps_per_second": 2.421,
|
| 841 |
+
"step": 255
|
| 842 |
+
},
|
| 843 |
+
{
|
| 844 |
+
"epoch": 74.0,
|
| 845 |
+
"eval_accuracy": 0.810126582278481,
|
| 846 |
+
"eval_loss": 0.7903711795806885,
|
| 847 |
+
"eval_runtime": 1.2802,
|
| 848 |
+
"eval_samples_per_second": 61.711,
|
| 849 |
+
"eval_steps_per_second": 2.343,
|
| 850 |
+
"step": 259
|
| 851 |
+
},
|
| 852 |
+
{
|
| 853 |
+
"epoch": 74.28571428571429,
|
| 854 |
+
"grad_norm": 4.974315166473389,
|
| 855 |
+
"learning_rate": 7.4074074074074075e-06,
|
| 856 |
+
"loss": 0.1248,
|
| 857 |
+
"step": 260
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 74.85714285714286,
|
| 861 |
+
"eval_accuracy": 0.810126582278481,
|
| 862 |
+
"eval_loss": 0.8036133646965027,
|
| 863 |
+
"eval_runtime": 1.2526,
|
| 864 |
+
"eval_samples_per_second": 63.067,
|
| 865 |
+
"eval_steps_per_second": 2.395,
|
| 866 |
+
"step": 262
|
| 867 |
+
},
|
| 868 |
+
{
|
| 869 |
+
"epoch": 76.0,
|
| 870 |
+
"eval_accuracy": 0.7848101265822784,
|
| 871 |
+
"eval_loss": 0.8591884970664978,
|
| 872 |
+
"eval_runtime": 1.2518,
|
| 873 |
+
"eval_samples_per_second": 63.108,
|
| 874 |
+
"eval_steps_per_second": 2.397,
|
| 875 |
+
"step": 266
|
| 876 |
+
},
|
| 877 |
+
{
|
| 878 |
+
"epoch": 76.85714285714286,
|
| 879 |
+
"eval_accuracy": 0.7848101265822784,
|
| 880 |
+
"eval_loss": 0.8945639133453369,
|
| 881 |
+
"eval_runtime": 1.2541,
|
| 882 |
+
"eval_samples_per_second": 62.996,
|
| 883 |
+
"eval_steps_per_second": 2.392,
|
| 884 |
+
"step": 269
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 77.14285714285714,
|
| 888 |
+
"grad_norm": 5.69161319732666,
|
| 889 |
+
"learning_rate": 5.555555555555556e-06,
|
| 890 |
+
"loss": 0.122,
|
| 891 |
+
"step": 270
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 78.0,
|
| 895 |
+
"eval_accuracy": 0.7848101265822784,
|
| 896 |
+
"eval_loss": 0.8575763702392578,
|
| 897 |
+
"eval_runtime": 1.2492,
|
| 898 |
+
"eval_samples_per_second": 63.241,
|
| 899 |
+
"eval_steps_per_second": 2.402,
|
| 900 |
+
"step": 273
|
| 901 |
+
},
|
| 902 |
+
{
|
| 903 |
+
"epoch": 78.85714285714286,
|
| 904 |
+
"eval_accuracy": 0.7974683544303798,
|
| 905 |
+
"eval_loss": 0.903687596321106,
|
| 906 |
+
"eval_runtime": 1.2474,
|
| 907 |
+
"eval_samples_per_second": 63.333,
|
| 908 |
+
"eval_steps_per_second": 2.405,
|
| 909 |
+
"step": 276
|
| 910 |
+
},
|
| 911 |
+
{
|
| 912 |
+
"epoch": 80.0,
|
| 913 |
+
"grad_norm": 4.918107509613037,
|
| 914 |
+
"learning_rate": 3.7037037037037037e-06,
|
| 915 |
+
"loss": 0.1129,
|
| 916 |
+
"step": 280
|
| 917 |
+
},
|
| 918 |
+
{
|
| 919 |
+
"epoch": 80.0,
|
| 920 |
+
"eval_accuracy": 0.7848101265822784,
|
| 921 |
+
"eval_loss": 0.998208224773407,
|
| 922 |
+
"eval_runtime": 1.2524,
|
| 923 |
+
"eval_samples_per_second": 63.079,
|
| 924 |
+
"eval_steps_per_second": 2.395,
|
| 925 |
+
"step": 280
|
| 926 |
+
},
|
| 927 |
+
{
|
| 928 |
+
"epoch": 80.85714285714286,
|
| 929 |
+
"eval_accuracy": 0.7721518987341772,
|
| 930 |
+
"eval_loss": 1.040386438369751,
|
| 931 |
+
"eval_runtime": 1.2523,
|
| 932 |
+
"eval_samples_per_second": 63.086,
|
| 933 |
+
"eval_steps_per_second": 2.396,
|
| 934 |
+
"step": 283
|
| 935 |
+
},
|
| 936 |
+
{
|
| 937 |
+
"epoch": 82.0,
|
| 938 |
+
"eval_accuracy": 0.7974683544303798,
|
| 939 |
+
"eval_loss": 0.9969496130943298,
|
| 940 |
+
"eval_runtime": 1.2562,
|
| 941 |
+
"eval_samples_per_second": 62.889,
|
| 942 |
+
"eval_steps_per_second": 2.388,
|
| 943 |
+
"step": 287
|
| 944 |
+
},
|
| 945 |
+
{
|
| 946 |
+
"epoch": 82.85714285714286,
|
| 947 |
+
"grad_norm": 3.4930834770202637,
|
| 948 |
+
"learning_rate": 1.8518518518518519e-06,
|
| 949 |
+
"loss": 0.1136,
|
| 950 |
+
"step": 290
|
| 951 |
+
},
|
| 952 |
+
{
|
| 953 |
+
"epoch": 82.85714285714286,
|
| 954 |
+
"eval_accuracy": 0.7974683544303798,
|
| 955 |
+
"eval_loss": 0.9585855007171631,
|
| 956 |
+
"eval_runtime": 1.2777,
|
| 957 |
+
"eval_samples_per_second": 61.829,
|
| 958 |
+
"eval_steps_per_second": 2.348,
|
| 959 |
+
"step": 290
|
| 960 |
+
},
|
| 961 |
+
{
|
| 962 |
+
"epoch": 84.0,
|
| 963 |
+
"eval_accuracy": 0.7848101265822784,
|
| 964 |
+
"eval_loss": 0.919994592666626,
|
| 965 |
+
"eval_runtime": 1.2517,
|
| 966 |
+
"eval_samples_per_second": 63.115,
|
| 967 |
+
"eval_steps_per_second": 2.397,
|
| 968 |
+
"step": 294
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 84.85714285714286,
|
| 972 |
+
"eval_accuracy": 0.7721518987341772,
|
| 973 |
+
"eval_loss": 0.9062978029251099,
|
| 974 |
+
"eval_runtime": 1.2533,
|
| 975 |
+
"eval_samples_per_second": 63.032,
|
| 976 |
+
"eval_steps_per_second": 2.394,
|
| 977 |
+
"step": 297
|
| 978 |
+
},
|
| 979 |
+
{
|
| 980 |
+
"epoch": 85.71428571428571,
|
| 981 |
+
"grad_norm": 5.335267066955566,
|
| 982 |
+
"learning_rate": 0.0,
|
| 983 |
+
"loss": 0.1233,
|
| 984 |
+
"step": 300
|
| 985 |
+
},
|
| 986 |
+
{
|
| 987 |
+
"epoch": 85.71428571428571,
|
| 988 |
+
"eval_accuracy": 0.7721518987341772,
|
| 989 |
+
"eval_loss": 0.9050403833389282,
|
| 990 |
+
"eval_runtime": 1.2508,
|
| 991 |
+
"eval_samples_per_second": 63.158,
|
| 992 |
+
"eval_steps_per_second": 2.398,
|
| 993 |
+
"step": 300
|
| 994 |
+
},
|
| 995 |
+
{
|
| 996 |
+
"epoch": 85.71428571428571,
|
| 997 |
+
"step": 300,
|
| 998 |
+
"total_flos": 2.9349165326823014e+18,
|
| 999 |
+
"train_loss": 0.2620420479774475,
|
| 1000 |
+
"train_runtime": 1806.3529,
|
| 1001 |
+
"train_samples_per_second": 24.469,
|
| 1002 |
+
"train_steps_per_second": 0.166
|
| 1003 |
+
},
|
| 1004 |
+
{
|
| 1005 |
+
"epoch": 85.71428571428571,
|
| 1006 |
+
"eval_accuracy": 0.8227848101265823,
|
| 1007 |
+
"eval_loss": 0.6919474005699158,
|
| 1008 |
+
"eval_runtime": 1.2565,
|
| 1009 |
+
"eval_samples_per_second": 62.873,
|
| 1010 |
+
"eval_steps_per_second": 2.388,
|
| 1011 |
+
"step": 300
|
| 1012 |
+
}
|
| 1013 |
+
],
|
| 1014 |
+
"logging_steps": 10,
|
| 1015 |
+
"max_steps": 300,
|
| 1016 |
+
"num_input_tokens_seen": 0,
|
| 1017 |
+
"num_train_epochs": 100,
|
| 1018 |
+
"save_steps": 500,
|
| 1019 |
+
"total_flos": 2.9349165326823014e+18,
|
| 1020 |
+
"train_batch_size": 32,
|
| 1021 |
+
"trial_name": null,
|
| 1022 |
+
"trial_params": null
|
| 1023 |
+
}
|