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

    Neuronpedia

    INDEX
    Explanations

    strings starting with specific characters followed by a sequence of other characters

    oai_token-act-pair · gpt-3.5-turbo

    punctuation and formatting related to dialogue or quotations

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    wcsstore
    -0.89
    ģ«
    -0.75
     pairing
    -0.74
    bath
    -0.70
    ixture
    -0.70
    furt
    -0.70
     Marketable
    -0.69
    inventoryQuantity
    -0.69
    aware
    -0.68
    iggurat
    -0.67
    POSITIVE LOGITS
    said
    1.03
     Said
    0.86
     Meh
    0.77
     Verse
    0.76
    Said
    0.73
     said
    0.73
     says
    0.72
     exclaimed
    0.72
     Mil
    0.71
     Lon
    0.70
    Act Density 0.627%

    No Known Activations