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

    Neuronpedia

    INDEX
    Explanations

    phrases containing punctuation marks

    oai_token-act-pair · gpt-3.5-turbo

    punctuation and pauses in the text

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    erent
    -0.78
    itational
    -0.78
    igham
    -0.78
    emn
    -0.77
    \">
    -0.76
    Additionally
    -0.76
    cerning
    -0.76
    maxwell
    -0.76
    rarily
    -0.74
    mx
    -0.73
    POSITIVE LOGITS
     albeit
    1.09
     culminating
    0.99
     embod
    0.99
     namely
    0.93
     reminding
    0.91
     except
    0.91
     morp
    0.89
     echoing
    0.88
     insofar
    0.87
     minus
    0.86
    Act Density 0.444%

    No Known Activations