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

    Neuronpedia

    INDEX
    Explanations

    references to spirals or spiraling actions

    oai_token-act-pair · gpt-3.5-turbo

    references to spirals and vortexes, often in metaphorical contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    Hispanic
    -0.87
    ourke
    -0.71
    endment
    -0.68
    lean
    -0.67
    ervatives
    -0.66
    utenant
    -0.66
    ropolitan
    -0.66
    ervative
    -0.64
    alty
    -0.64
    essee
    -0.63
    POSITIVE LOGITS
     spiral
    1.16
     staircase
    1.06
     spir
    1.04
    uler
    0.86
     Spiral
    0.82
    rily
    0.77
     vortex
    0.75
     stair
    0.74
     galaxies
    0.72
    iosity
    0.72
    Act Density 0.025%

    No Known Activations