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

    Neuronpedia

    INDEX
    Explanations

    words related to the human body

    oai_token-act-pair · gpt-3.5-turbo

    references to the human body and its physical aspects

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     Hoover
    -0.84
     Clover
    -0.76
     Ans
    -0.73
     Kafka
    -0.72
     Booth
    -0.72
     Mub
    -0.67
     Nex
    -0.66
     Trey
    -0.65
     Doodle
    -0.65
    yrinth
    -0.64
    POSITIVE LOGITS
    guards
    1.30
    building
    1.26
    builders
    1.23
    builder
    1.18
    weight
    1.11
    guard
    1.11
    parts
    1.08
     politic
    0.99
    fat
    0.99
    wash
    0.94
    Act Density 0.028%

    No Known Activations