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

    Neuronpedia

    INDEX
    Explanations

    instances of figurative language and tones indicating ambiguity

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     ëĨĢ
    -0.14
    ihil
    -0.13
    atomy
    -0.13
    goto
    -0.13
    hurst
    -0.13
    rin
    -0.13
    angelog
    -0.12
    arine
    -0.12
    anta
    -0.12
    Pow
    -0.12
    POSITIVE LOGITS
     literal
    1.05
     literally
    1.00
     Liter
    0.94
    liter
    0.89
    literal
    0.88
     Literal
    0.84
    Liter
    0.77
    Literal
    0.75
     liter
    0.70
    -liter
    0.70
    Act Density 0.127%

    No Known Activations