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

    Neuronpedia

    INDEX
    Explanations

    code segments and specific identifiers

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ाइवेट
    0.69
     Sampling
    0.69
     mixes
    0.67
     Sample
    0.66
    闶
    0.65
    湳
    0.64
     blankets
    0.63
    Sample
    0.63
     Circuits
    0.63
     sampling
    0.62
    POSITIVE LOGITS
    reen
    0.70
    chad
    0.64
    chten
    0.64
    proken
    0.60
     শ
    0.59
    祚
    0.58
     ekst
    0.58
    सीपी
    0.58
     forget
    0.58
     받아
    0.58
    Act Density 0.096%

    No Known Activations