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

    Neuronpedia

    INDEX
    Explanations

    words with non-English characters, potentially related to a specific language or encoding system

    oai_token-act-pair · gpt-3.5-turbo

    special characters or sequences in a document

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    geries
    -1.04
     foreground
    -0.83
     distracted
    -0.75
     distracting
    -0.71
     background
    -0.70
     dividing
    -0.68
     intrig
    -0.67
    gery
    -0.67
     Catalyst
    -0.66
    raints
    -0.66
    POSITIVE LOGITS
    и
    1.12
    ļ
    1.06
    Ħ
    1.06
    ij
    1.04
    о
    0.98
    а
    0.95
    à¸
    0.94
    Ñĥ
    0.94
    ski
    0.94
    ł
    0.94
    Act Density 0.006%

    No Known Activations