cyrilzhang
commited on
Commit
•
1e5b7d8
1
Parent(s):
ed23458
Update README.md
Browse files
README.md
CHANGED
@@ -24,4 +24,4 @@ gpt2_tokenizer.decode([10163, 46387]) # '123 pigeon'
|
|
24 |
- This is for my investigations into the arithmetic capabilities of large language models. There is no model here, only a tokenizer.
|
25 |
- [PaLM](https://arxiv.org/abs/2204.02311) does this.
|
26 |
- I think it's very reasonable.
|
27 |
-
- Many models (illustriously, [GPT-3](https://arxiv.org/abs/2005.14165)) use the GPT-2 tokenizer
|
|
|
24 |
- This is for my investigations into the arithmetic capabilities of large language models. There is no model here, only a tokenizer.
|
25 |
- [PaLM](https://arxiv.org/abs/2204.02311) does this.
|
26 |
- I think it's very reasonable.
|
27 |
+
- Many models (illustriously, [GPT-3](https://arxiv.org/abs/2005.14165)) don't do this, because they use the GPT-2 tokenizer.
|