minke.yu/sglang
1
0
Fork 0
Code Issues Pull Requests Actions 12 Packages Projects Releases Wiki Activity

Reject repetition_penalty=0 in SamplingParams.verify() (#24874)

Browse Source 
Co-authored-by: RulinJuice <265952454+RulinJuice@users.noreply.github.com>
This commit is contained in:
RulinJuice
2026-05-12 21:25:23 -07:00
committed by GitHub
co-authored by RulinJuice
parent 622baa17bd
commit 3f048c80b8
3 changed files with 22 additions and 8 deletions
Download Patch File Download Diff File Expand all files Collapse all files
docs/basic_usage/sampling_params.md
+1 -1
View File
@@ -62,7 +62,7 @@ python -m sglang.launch_server --model-path <MODEL> --sampling-defaults openai
|--------------------|------------------------|------------------------------------------------------------------------------------------------------------------------------------------------|
| frequency_penalty | `float = 0.0` | Penalizes tokens based on their frequency in generation so far. Must be between `-2` and `2` where negative numbers encourage repeatment of tokens and positive number encourages sampling of new tokens. The scaling of penalization grows linearly with each appearance of a token. |
| presence_penalty | `float = 0.0` | Penalizes tokens if they appeared in the generation so far. Must be between `-2` and `2` where negative numbers encourage repeatment of tokens and positive number encourages sampling of new tokens. The scaling of the penalization is constant if a token occurred. |
| repetition_penalty | `float = 1.0` | Scales the logits of previously generated tokens to discourage (values > 1) or encourage (values < 1) repetition. Valid range is `[0, 2]`; `1.0` leaves probabilities unchanged. |
| repetition_penalty | `float = 1.0` | Scales the logits of previously generated tokens to discourage (values > 1) or encourage (values < 1) repetition. Valid range is `(0, 2]`; `1.0` leaves probabilities unchanged. |
| min_new_tokens | `int = 0` | Forces the model to generate at least `min_new_tokens` until a stop word or EOS token is sampled. Note that this might lead to unintended behavior, for example, if the distribution is highly skewed towards these tokens. |
### Constrained decoding