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

    Neuronpedia

    INDEX
    Explanations

    associated with physical barriers or obstacles

    oai_token-act-pair · gpt-3.5-turbo

    references to obstacles or hindrances

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ivery
    -0.84
    iability
    -0.79
    phis
    -0.76
    sch
    -0.72
    inav
    -0.70
    ergy
    -0.69
    orp
    -0.69
    sb
    -0.68
    ovie
    -0.68
    ribution
    -0.68
    POSITIVE LOGITS
     barriers
    1.11
     barrier
    0.98
     erected
    0.96
     separating
    0.88
     walls
    0.87
     Walls
    0.86
     Barrier
    0.81
    buster
    0.79
     wall
    0.78
     fence
    0.73
    Act Density 0.016%

    No Known Activations