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

    Neuronpedia

    INDEX
    Explanations

    uncertainty or possibility

    np_max-act · gemini-2.0-flash

    The neuron fires on conversational “hedging” and discourse‐marker words—i.e. filler or emphasis terms (like Okay, so, maybe, not quite, that, etc.) that modulate or soften statements.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     Encrypt
    -0.07
     koc
    -0.06
    赛
    -0.06
    戦
    -0.06
     kw
    -0.06
     instagram
    -0.06
    (players
    -0.06
     Роб
    -0.06
     HEX
    -0.06
     Mex
    -0.06
    POSITIVE LOGITS
     hlas
    0.08
    ून
    0.07
    ACA
    0.07
    HttpRequest
    0.06
    AFE
    0.06
    */
    0.06
    Torrent
    0.06
    amos
    0.06
    aban
    0.06
     translator
    0.06
    Act Density 0.134%

    No Known Activations