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

    Neuronpedia

    INDEX
    Explanations

    remarks on performance or success

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "well" in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    hent
    -0.81
    empt
    -0.76
    leans
    -0.75
    oute
    -0.71
     Alexandria
    -0.71
    ory
    -0.68
    İĭ
    -0.67
    iliary
    -0.66
    leted
    -0.65
    activated
    -0.65
    POSITIVE LOGITS
     enough
    1.08
    enough
    0.98
     Enough
    0.78
     suited
    0.72
     outweigh
    0.70
     liked
    0.69
    espie
    0.68
    Topic
    0.68
     alright
    0.67
     Archdemon
    0.67
    Act Density 0.032%

    No Known Activations