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

    Neuronpedia

    INDEX
    Explanations

    aspects related to specific numerical data or counts within text

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Negative Logits
    kson
    -0.95
    kie
    -0.88
     torches
    -0.80
    stars
    -0.79
    wana
    -0.76
     Hots
    -0.74
     Cosponsors
    -0.73
     flaw
    -0.71
     conflic
    -0.70
    stood
    -0.69
    POSITIVE LOGITS
    stantial
    1.15
    ference
    1.06
    ople
    1.01
    rete
    0.95
    umn
    0.94
    cise
    0.93
    media
    0.92
    ceed
    0.89
    bernatorial
    0.88
    uits
    0.87
    Act Density 0.141%

    No Known Activations