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

    Neuronpedia

    INDEX
    Explanations

    literal evaluation ast.literal_eval

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

    present-participle/gerund forms (words in the -ing form) and progressive verb constructions.

    oai_token-act-pair · gpt-5Triggered by @vetterc0
    New Auto-Interp
    Negative Logits
    麻烦
    0.41
    etheless
    0.39
    uiscono
    0.39
     cristiano
    0.39
     bakteri
    0.38
     nyaman
    0.38
     hassles
    0.38
    avljen
    0.38
    儂
    0.38
     organismo
    0.38
    POSITIVE LOGITS
     симво
    0.47
    symbol
    0.44
    Symbol
    0.43
     symbolized
    0.42
     symbolizing
    0.42
     radiating
    0.41
     చిహ్
    0.41
     ада
    0.40
     има
    0.40
     symbolically
    0.39
    Act Density 0.005%

    No Known Activations