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

    Neuronpedia

    INDEX
    Explanations

    conjunctions and punctuation marks

    oai_token-act-pair · gpt-3.5-turbo

    phrases that express uncertainty or conditionality

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     To
    -0.53
     to
    -0.50
     Demand
    -0.46
    to
    -0.43
     TO
    -0.41
     Wonderful
    -0.41
    ombo
    -0.40
     Seeking
    -0.40
     Quant
    -0.39
     Celebr
    -0.39
    POSITIVE LOGITS
    none
    0.61
    although
    0.60
    only
    0.59
    albeit
    0.59
     only
    0.56
    without
    0.56
    though
    0.56
    even
    0.55
    yond
    0.54
    but
    0.54
    Act Density 1.419%

    No Known Activations