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

    Neuronpedia

    INDEX
    Explanations

    phrases related to desires and intentions

    oai_token-act-pair · gpt-3.5-turbo

    expressions of desire or reluctance indicated by the word "want"

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    issance
    -0.90
     Enhancement
    -0.72
    ework
    -0.72
    bard
    -0.69
    iste
    -0.66
    unny
    -0.66
    ilian
    -0.64
    utical
    -0.63
     âĨij
    -0.63
    icol
    -0.63
    POSITIVE LOGITS
     anymore
    0.98
     anybody
    0.95
     anything
    0.91
     anyone
    0.90
     any
    0.82
     nor
    0.82
    ANY
    0.78
     ANY
    0.75
     wasting
    0.74
     them
    0.73
    Act Density 0.060%

    No Known Activations