silx-ai
/

TARS-1B

Text Generation

liquid-neural-network

non-transformer

efficient-model

Model card Files Files and versions

eyad-silx commited on Jul 20

Commit

fc5d1be

·

verified ·

1 Parent(s): e5785ed

Update README.md

Files changed (1) hide show

README.md +1 -0

README.md CHANGED Viewed

@@ -26,6 +26,7 @@ tags:
 | **ARC-Easy**           | 32.95%    |
 | **PIQA**               | 51.36%    |
 | **BoolQ**              | 66.00%    |
 > ⚠️ These results were achieved with **only 300M tokens of pretraining**, making them exceptionally efficient compared to transformer-based models.

 | **ARC-Easy**           | 32.95%    |
 | **PIQA**               | 51.36%    |
 | **BoolQ**              | 66.00%    |
+| **GPQA-Diamond**       | 18.18%    |
 > ⚠️ These results were achieved with **only 300M tokens of pretraining**, making them exceptionally efficient compared to transformer-based models.