# GPT2TokenizerFast tokenzied output

**URL:** <https://discuss.huggingface.co/t/gpt2tokenizerfast-tokenzied-output/67424>\
**Category:** Beginners\
**Created:** [December 29, 2023, 9:25pm UTC](https://discuss.huggingface.co/t/gpt2tokenizerfast-tokenzied-output/67424 "2023-12-29T21:25:12Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![matthewchung74](https://sea2.discourse-cdn.com/hellohellohello/user_avatar/discuss.huggingface.co/matthewchung74/32/15244_2.png) [@matthewchung74](https://discuss.huggingface.co/u/matthewchung74)\
**Post date:** [December 29, 2023, 9:25pm UTC](https://discuss.huggingface.co/t/gpt2tokenizerfast-tokenzied-output/67424/1 "2023-12-29T21:25:12Z")

</div>

When using the gpt2 tokenizer, I get output where each word starts with the Ġ. I’m not sure what the Ġ is? Any suggestions?

```auto
from transformers import GPT2TokenizerFast
tokenizer = GPT2TokenizerFast.from_pretrained('gpt2', add_prefix_space=True)
inputs = tokenizer("today is a nice day")
inputs.tokens()

```

```auto
['Ġtoday', 'Ġis', 'Ġa', 'Ġnice', 'Ġday']

```