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

    Neuronpedia

    INDEX
    Explanations

    instances of exposure or being exposed to something

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word 'to' indicating exposure or influence

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    chief
    -0.71
    aturdays
    -0.70
    signed
    -0.68
     meant
    -0.68
    auld
    -0.67
    widget
    -0.67
     largeDownload
    -0.67
    rollers
    -0.66
    å¦
    -0.66
    notation
    -0.65
    POSITIVE LOGITS
    othy
    0.87
    asty
    0.82
     conserve
    0.81
     accommodate
    0.79
     fend
    0.77
    asted
    0.76
     preserve
    0.75
    pload
    0.75
    ggles
    0.74
     withstand
    0.73
    Act Density 0.189%

    No Known Activations