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

    Neuronpedia

    INDEX
    Explanations

    punctuation indicating contrast or continuation in a sentence

    oai_token-act-pair · gpt-3.5-turbo

    patterns of repetition or transition in phrases

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    bound
    -0.61
    Krist
    -0.57
     suitcase
    -0.56
    ish
    -0.54
     sedan
    -0.53
    SD
    -0.53
    pak
    -0.52
    cott
    -0.52
    mate
    -0.52
    SM
    -0.52
    POSITIVE LOGITS
    âķIJâķIJ
    0.82
     alas
    0.76
     beware
    0.73
     acknow
    0.73
     //[
    0.71
    ebus
    0.70
     anecd
    0.68
     nonetheless
    0.67
     suffice
    0.66
    ̶
    0.66
    Act Density 0.079%

    No Known Activations