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

    Neuronpedia

    INDEX
    Explanations

    negatively connotated words or phrases, particularly related to mistakes or misdeeds

    oai_token-act-pair · gpt-3.5-turbo

    words related to various forms of mistakes or errors

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ulhu
    -0.71
    minster
    -0.68
     Fres
    -0.67
     Collider
    -0.62
     cooled
    -0.62
    gel
    -0.62
    Lago
    -0.61
    Fra
    -0.60
    retty
    -0.59
    raltar
    -0.58
    POSITIVE LOGITS
    vous
    0.90
    ukong
    0.76
    dule
    0.76
    gotten
    0.75
    ammad
    0.70
    ceived
    0.70
    WARE
    0.70
    rued
    0.68
    Ö
    0.67
    adows
    0.67
    Act Density 0.050%

    No Known Activations