Update readme
Browse files
README.md
CHANGED
|
@@ -1,5 +1,20 @@
|
|
| 1 |
---
|
|
|
|
|
|
|
|
|
|
| 2 |
license: mit
|
| 3 |
datasets:
|
| 4 |
- HuggingFaceFW/fineweb-edu
|
| 5 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
---
|
| 2 |
+
library_name: RAT
|
| 3 |
+
language:
|
| 4 |
+
- en
|
| 5 |
license: mit
|
| 6 |
datasets:
|
| 7 |
- HuggingFaceFW/fineweb-edu
|
| 8 |
+
tags:
|
| 9 |
+
- efficient architecture
|
| 10 |
+
- recurrence
|
| 11 |
+
- attention
|
| 12 |
+
- pretraining
|
| 13 |
+
metrics:
|
| 14 |
+
- perplexity
|
| 15 |
+
- accuracy
|
| 16 |
+
---
|
| 17 |
+
|
| 18 |
+
## References
|
| 19 |
+
|
| 20 |
+
* [RAT Paper](https://arxiv.org/abs/2507.04416)
|