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

    Neuronpedia

    INDEX
    Explanations

    numerical values or codes

    oai_token-act-pair · gpt-3.5-turbo

    currency symbols and related numerical data

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     Shank
    -0.79
    hare
    -0.79
    hene
    -0.78
    sle
    -0.77
    ĭ
    -0.77
    hang
    -0.74
    beck
    -0.74
    oÄŁ
    -0.73
    arers
    -0.73
    hift
    -0.73
    POSITIVE LOGITS
     91
    1.15
     92
    1.08
     88
    0.96
    91
    0.96
     86
    0.96
     96
    0.95
     90
    0.94
     93
    0.93
     94
    0.91
     87
    0.90
    Act Density 0.229%

    No Known Activations