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

    Neuronpedia

    INDEX
    Explanations

    references related to physical actions, particularly rolling or moving in a specific direction

    oai_token-act-pair · gpt-3.5-turbo

    phrases related to actions involving rolling or physical movement

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    auna
    -0.85
    orage
    -0.74
    hai
    -0.72
    eele
    -0.67
    aline
    -0.65
    uo
    -0.65
     Kut
    -0.64
    ibo
    -0.64
    nih
    -0.63
    oka
    -0.63
    POSITIVE LOGITS
     sleeves
    1.04
     curtains
    0.81
     shoulders
    0.74
     punches
    0.73
     hips
    0.71
     dice
    0.70
     sided
    0.68
     shotgun
    0.67
     corners
    0.66
     boulder
    0.66
    Act Density 0.089%

    No Known Activations