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

    Neuronpedia

    INDEX
    Explanations

    words related to cardinal or ordinal directions

    oai_token-act-pair · gpt-3.5-turbo

    directional terms related to travel or movement

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    aints
    -0.67
    ellation
    -0.64
    arth
    -0.61
    Owner
    -0.60
    lest
    -0.59
     tie
    -0.59
     Tale
    -0.58
    acea
    -0.58
     Scal
    -0.57
     Return
    -0.57
    POSITIVE LOGITS
    wards
    0.83
    WARD
    0.81
     unnoticed
    0.79
    ModLoader
    0.79
    iless
    0.78
     toward
    0.73
     towards
    0.70
    noticed
    0.70
    stairs
    0.69
     stairs
    0.68
    Act Density 0.113%

    No Known Activations