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

    Neuronpedia

    INDEX
    Explanations

    mentions of things being predominantly of a certain type or group

    oai_token-act-pair · gpt-3.5-turbo

    occurrences of the word "mostly."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     Kard
    -0.79
    alid
    -0.78
    ongyang
    -0.72
    anth
    -0.71
     Bray
    -0.69
    ilion
    -0.67
    iry
    -0.67
    yth
    -0.66
    angers
    -0.65
    IDA
    -0.65
    POSITIVE LOGITS
     consisted
    0.91
     consist
    0.85
     consisting
    0.84
     comprised
    0.83
     consists
    0.82
     lacking
    0.82
     scattered
    0.81
     populated
    0.76
     concentrated
    0.76
     composed
    0.75
    Act Density 0.010%

    No Known Activations