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

    Neuronpedia

    INDEX
    Explanations

    names with some special characters such as 'Ļ' and 'âĢ'

    oai_token-act-pair · gpt-3.5-turbo

    instances of a specific character or entity within the text

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     redundancy
    -0.72
     trickle
    -0.69
     relevance
    -0.69
     diffusion
    -0.69
     dwindling
    -0.68
     friction
    -0.67
     taboo
    -0.67
     relegation
    -0.67
     geography
    -0.67
     contingency
    -0.65
    POSITIVE LOGITS
    ves
    1.02
    s
    0.93
    ï¸ı
    0.92
    wrote
    0.92
    sent
    0.91
    felt
    0.91
     knows
    0.89
    t
    0.88
    ishly
    0.88
     owns
    0.87
    Act Density 0.206%

    No Known Activations