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

    Neuronpedia

    INDEX
    Explanations

    instances where words are followed by punctuation marks

    oai_token-act-pair · gpt-3.5-turbo

    instances of punctuation, particularly commas

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ãĥŁ
    -0.81
    ulla
    -0.73
    ãĤ§
    -0.73
    ahu
    -0.72
    ©¶æ¥µ
    -0.70
    ãĤº
    -0.69
    mite
    -0.68
    ãĤ¶
    -0.68
    atre
    -0.68
    н
    -0.64
    POSITIVE LOGITS
     how
    1.34
     including
    1.11
     whether
    1.09
     namely
    1.05
     why
    1.02
    how
    1.01
     comparing
    0.97
     outlining
    0.95
     noting
    0.95
     detailing
    0.94
    Act Density 0.298%

    No Known Activations