SUMIN/HuggingFace_transformer
1
0
Fork 0
Code Issues Pull Requests Actions 7 Packages Projects Releases Wiki Activity

Fix: typo (#33880)

Browse Source 
Update llm_tutorial.md: typo
This commit is contained in:
TrickEye
2024-10-02 16:12:21 +08:00
committed by GitHub
parent 61ac161a9d
commit 2292be6c1b
1 changed files with 1 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
docs/source/en/llm_tutorial.md
View File

@@ -164,7 +164,7 @@ If not specified in the [`~generation.GenerationConfig`] file, `generate` return
By default, and unless specified in the [`~generation.GenerationConfig`] file, `generate` selects the most likely token at each iteration (greedy decoding). Depending on your task, this may be undesirable; creative tasks like chatbots or writing an essay benefit from sampling. On the other hand, input-grounded tasks like audio transcription or translation benefit from greedy decoding. Enable sampling with `do_sample=True`, and you can learn more about this topic in this [blog post](https://huggingface.co/blog/how-to-generate). By default, and unless specified in the [`~generation.GenerationConfig`] file, `generate` selects the most likely token at each iteration (greedy decoding). Depending on your task, this may be undesirable; creative tasks like chatbots or writing an essay benefit from sampling. On the other hand, input-grounded tasks like audio transcription or translation benefit from greedy decoding. Enable sampling with `do_sample=True`, and you can learn more about this topic in this [blog post](https://huggingface.co/blog/how-to-generate).
```py ```py
>>> # Set seed or reproducibility -- you don't need this unless you want full reproducibility >>> # Set seed for reproducibility -- you don't need this unless you want full reproducibility
>>> from transformers import set_seed >>> from transformers import set_seed
>>> set_seed(42) >>> set_seed(42)