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

    Neuronpedia

    INDEX
    Explanations

    phrases related to leaving or departing

    oai_token-act-pair · gpt-3.5-turbo

    phrases related to escape or evasion

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    urally
    -0.81
    umbing
    -0.78
    matically
    -0.75
     underestimate
    -0.74
     challeng
    -0.70
    orescent
    -0.69
    mos
    -0.69
    κ
    -0.68
    immer
    -0.68
     reactive
    -0.68
    POSITIVE LOGITS
    away
    0.90
    aways
    0.83
     Opportun
    0.80
     Wildcats
    0.77
    heid
    0.77
    lane
    0.73
    916
    0.69
    vantage
    0.68
    AU
    0.68
    tery
    0.67
    Act Density 0.017%

    No Known Activations