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

    Neuronpedia

    INDEX
    Explanations

    code and instructions

    np_acts-logits-general · gemini-2.5-flash-lite

    mentions of programming-related identifiers—APIs, libraries, and framework-specific terms—within code or technical instructions.

    oai_token-act-pair · gpt-5Triggered by @vetterc0
    New Auto-Interp
    Negative Logits
     Failed
    0.52
     Favorite
    0.52
     Mechanism
    0.50
     顔
    0.50
     Generated
    0.49
     Faces
    0.49
     Completed
    0.48
     Updated
    0.47
     Billion
    0.47
     Elections
    0.46
    POSITIVE LOGITS
    gel
    0.48
    et
    0.47
    feu
    0.47
    o
    0.46
    al
    0.46
    or
    0.44
    ли
    0.43
    inform
    0.43
    dur
    0.42
    U
    0.42
    Act Density 0.029%

    No Known Activations