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

    Neuronpedia

    INDEX
    Explanations

    instances where a particular "thing" is emphasized within a context

    oai_token-act-pair · gpt-3.5-turbo

    phrases emphasizing important or notable statements

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    inav
    -0.76
    ULT
    -0.72
    cul
    -0.71
    oÄŁ
    -0.70
    lems
    -0.70
    DOS
    -0.67
    bush
    -0.67
    ONSORED
    -0.67
    JV
    -0.65
    vez
    -0.65
    POSITIVE LOGITS
     happens
    0.78
    iverse
    0.76
    rued
    0.74
     happened
    0.70
     separates
    0.68
     Valiant
    0.65
    afort
    0.64
     counts
    0.63
     happening
    0.62
     touches
    0.60
    Act Density 0.032%

    No Known Activations