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

    Neuronpedia

    INDEX
    Explanations

    phrases related to not giving up or showing determination

    oai_token-act-pair · gpt-3.5-turbo

    expressions indicating resistance or retreating behavior

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    marks
    -0.72
    usable
    -0.67
    tained
    -0.64
     Fam
    -0.64
    anon
    -0.63
    lain
    -0.61
    burning
    -0.59
    frey
    -0.59
    marked
    -0.59
    oret
    -0.59
    POSITIVE LOGITS
     blindly
    0.81
     timid
    0.80
     inaction
    0.76
     cowardly
    0.75
    olicy
    0.74
     conced
    0.74
     retreat
    0.72
     retreating
    0.71
     apologise
    0.71
     hesitate
    0.71
    Act Density 0.116%

    No Known Activations