Update README.md
Browse files
README.md
CHANGED
@@ -82,3 +82,12 @@ configs:
|
|
82 |
- split: swap_word
|
83 |
path: data/swap_word-*
|
84 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
82 |
- split: swap_word
|
83 |
path: data/swap_word-*
|
84 |
---
|
85 |
+
|
86 |
+
# CUTE
|
87 |
+
|
88 |
+
Here is the CUTE benchmark, a benchmark designed for testing LLM's ability to understand the characters within their tokens.
|
89 |
+
|
90 |
+
To use this dataset as we did, make sure to use ``` tokenizer.apply_chat_template ``` on the prompt, and then add ``` Answer: " ``` afterwards to the resulting string.
|
91 |
+
For smaller LLMs, you may need to do some post-processing on the final answer.
|
92 |
+
|
93 |
+
You can also check out our implementation at: https://github.com/Leukas/cute
|