LLM-Hallucination-Detection-Script good first issues (Rust)

Probabilities just under the threshold print as equal to it 30 days ago

bug good first issue

Mattbusel/LLM-Hallucination-Detection-Script

17

Rust CLI and library that flags low-confidence spans in LLM answers from token logprobs (offline or live via any OpenAI-compatible API) and renders per-token confidence as terminal, HTML or Markdown.

Rust

#ai-safety#cli#hallucination-detection#llm#llm-evaluation#logprobs#nlp#rust#token-visualization

Add --system to live mode 30 days ago

enhancement good first issue

Mattbusel/LLM-Hallucination-Detection-Script

17

Rust CLI and library that flags low-confidence spans in LLM answers from token logprobs (offline or live via any OpenAI-compatible API) and renders per-token confidence as terminal, HTML or Markdown.

Rust

#ai-safety#cli#hallucination-detection#llm#llm-evaluation#logprobs#nlp#rust#token-visualization

Show the detection summary in HTML and Markdown output 30 days ago

enhancement good first issue

Mattbusel/LLM-Hallucination-Detection-Script

17

Rust CLI and library that flags low-confidence spans in LLM answers from token logprobs (offline or live via any OpenAI-compatible API) and renders per-token confidence as terminal, HTML or Markdown.

Rust

#ai-safety#cli#hallucination-detection#llm#llm-evaluation#logprobs#nlp#rust#token-visualization

Accept the legacy Completions logprobs format 30 days ago

enhancement good first issue

Mattbusel/LLM-Hallucination-Detection-Script

17

Rust CLI and library that flags low-confidence spans in LLM answers from token logprobs (offline or live via any OpenAI-compatible API) and renders per-token confidence as terminal, HTML or Markdown.

Rust

#ai-safety#cli#hallucination-detection#llm#llm-evaluation#logprobs#nlp#rust#token-visualization

detect_issues misses an overconfident streak that runs to the last token 30 days ago

bug good first issue

Mattbusel/LLM-Hallucination-Detection-Script

17

Rust CLI and library that flags low-confidence spans in LLM answers from token logprobs (offline or live via any OpenAI-compatible API) and renders per-token confidence as terminal, HTML or Markdown.

Rust

#ai-safety#cli#hallucination-detection#llm#llm-evaluation#logprobs#nlp#rust#token-visualization

Mattbusel/LLM-Hallucination-Detection-Script

Open Issues Need Help