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

    Neuronpedia

    INDEX
    Explanations

    phrases indicating possibility, potential, and considerations for improvement or modification

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     presumably
    -0.94
     obviously
    -0.92
     evidently
    -0.92
     undoubtedly
    -0.91
     basically
    -0.90
    presumably
    -0.85
     usually
    -0.85
     doubtless
    -0.85
     definitely
    -0.82
    basically
    -0.82
    POSITIVE LOGITS
     depending
    1.06
    depending
    0.97
     jopa
    0.96
     someday
    0.93
     algún
    0.92
    anskje
    0.91
     some
    0.89
     might
    0.83
    Depending
    0.82
     algunos
    0.80
    Act Density 0.601%

    No Known Activations