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

    Neuronpedia

    INDEX
    Explanations

    instances of the word "very"

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "very."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    osal
    -0.84
    adelphia
    -0.77
    ensis
    -0.72
    olor
    -0.69
    imus
    -0.67
    ando
    -0.67
    orno
    -0.67
    iture
    -0.65
    igi
    -0.63
    onent
    -0.63
    POSITIVE LOGITS
     seldom
    0.77
     different
    0.76
     readable
    0.76
     rarely
    0.76
     informative
    0.76
     handy
    0.75
     similar
    0.75
     important
    0.75
     expensive
    0.73
     dear
    0.73
    Act Density 0.061%

    No Known Activations