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

    Neuronpedia

    INDEX
    Explanations

    certain descriptive phrases or qualities within sentences

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Negative Logits
    ayan
    -0.71
    »
    -0.71
    uku
    -0.71
    ional
    -0.70
    Zone
    -0.67
    ione
    -0.67
    Contract
    -0.65
    peat
    -0.65
    vl
    -0.65
    seys
    -0.64
    POSITIVE LOGITS
     distinguishes
    1.27
     inspires
    1.20
     separates
    1.17
     motiv
    1.15
     enables
    1.12
     defines
    1.10
     underpin
    1.09
     attracts
    1.09
     makes
    1.08
     drove
    1.07
    Act Density 0.162%

    No Known Activations