Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    INDEX
    Explanations

    negation/questioning

    np_max-act · gemini-2.0-flash

    Expressions of epistemic stance and modal/auxiliary constructions that signal knowing, believing, possibility, or negation.

    oai_token-act-pair · gpt-5-miniTriggered by @vetterc0
    New Auto-Interp
    Negative Logits
    ex
    -0.07
    stable
    -0.06
    warf
    -0.06
    Storm
    -0.06
    โรค
    -0.06
    ErrorMsg
    -0.06
    .symbol
    -0.06
    table
    -0.06
    _lang
    -0.06
     профес
    -0.06
    POSITIVE LOGITS
     concerns
    0.07
    .Edit
    0.06
     Reduced
    0.06
     всегда
    0.06
    ریه
    0.06
    ajaran
    0.06
    čan
    0.06
     obce
    0.06
    	line
    0.06
     libido
    0.06
    Act Density 0.311%

    No Known Activations