JustDoItNow
/

deberta-v3-large-reward-model

Generated from Trainer

Model card Files Files and versions Metrics Training metrics Community

JustDoItNow commited on Aug 19

Commit

0d59c3b

•

1 Parent(s): c637071

Model save

Files changed (1) hide show

README.md +9 -7

README.md CHANGED Viewed

@@ -19,8 +19,8 @@ should probably proofread and complete it, then remove this comment. -->
 This model is a fine-tuned version of [microsoft/deberta-v3-large](https://huggingface.co/microsoft/deberta-v3-large) on an unknown dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.0200
-- Accuracy: 0.9967
 ## Model description
@@ -51,11 +51,13 @@ The following hyperparameters were used during training:
 ### Training results
-| Training Loss | Epoch  | Step | Validation Loss | Accuracy |
-|:-------------:|:------:|:----:|:---------------:|:--------:|
-| 0.0023        | 2.6667 | 100  | 0.0123          | 1.0      |
-| 0.0002        | 5.3333 | 200  | 0.0184          | 0.9967   |
-| 0.0001        | 8.0    | 300  | 0.0200          | 0.9967   |
 ### Framework versions

 This model is a fine-tuned version of [microsoft/deberta-v3-large](https://huggingface.co/microsoft/deberta-v3-large) on an unknown dataset.
 It achieves the following results on the evaluation set:
+- Loss: 0.0128
+- Accuracy: 0.995
 ## Model description
 ### Training results
+| Training Loss | Epoch | Step | Validation Loss | Accuracy |
+|:-------------:|:-----:|:----:|:---------------:|:--------:|
+| 0.0312        | 2.0   | 100  | 0.0212          | 0.9925   |
+| 0.0013        | 4.0   | 200  | 0.0131          | 0.995    |
+| 0.0003        | 6.0   | 300  | 0.0119          | 0.995    |
+| 0.0003        | 8.0   | 400  | 0.0132          | 0.995    |
+| 0.0001        | 10.0  | 500  | 0.0128          | 0.995    |
 ### Framework versions