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

    Neuronpedia

    INDEX
    Explanations

    phrases discussing the potential or possibility of action

    oai_token-act-pair · gpt-3.5-turbo

    modal verbs indicating capability or potential

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    furt
    -0.95
    ele
    -0.76
    quartered
    -0.73
    bum
    -0.72
    76561
    -0.70
     Cheong
    -0.69
    phal
    -0.67
    teen
    -0.66
    anamo
    -0.66
    ciating
    -0.64
    POSITIVE LOGITS
    't
    1.31
     improve
    1.15
     safely
    1.08
     overcome
    1.02
     help
    1.01
     mitigate
    1.00
     afford
    0.98
     contribute
    0.96
     capitalize
    0.95
     accomplish
    0.94
    Act Density 0.143%

    No Known Activations