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

    Neuronpedia

    INDEX
    Explanations

    The main thing this neuron does is find mentions of the word 'Border'

    oai_token-act-pair · gpt-3.5-turbo

    mentions of "Border Patrol" and "Border" related topics

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    thora
    -0.84
    issance
    -0.78
    IVES
    -0.77
    hots
    -0.76
    ynthesis
    -0.75
    VERTISEMENT
    -0.74
    andum
    -0.74
    æ©Ł
    -0.74
    IGHTS
    -0.73
    ource
    -0.72
    POSITIVE LOGITS
     crossings
    1.16
     Patrol
    0.97
     patrol
    0.91
    lines
    0.91
     crossing
    0.86
     Border
    0.83
     border
    0.82
    line
    0.81
     guards
    0.81
     clearance
    0.80
    Act Density 0.015%

    No Known Activations