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

    Neuronpedia

    INDEX
    Explanations

    phrases starting with "As a" followed by a specific identity or role

    oai_token-act-pair · gpt-3.5-turbo

    statements that begin with a reference to personal identity or status

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    TERN
    -0.72
    illin
    -0.70
    »Ĵ
    -0.67
    arat
    -0.61
    Stars
    -0.61
    Cause
    -0.61
    iously
    -0.60
    UGE
    -0.60
     eruption
    -0.59
    ÅĤ
    -0.59
    POSITIVE LOGITS
     however
    0.85
     naturally
    0.69
     meanwhile
    0.69
     moreover
    0.65
     accustomed
    0.63
     huh
    0.63
     Sue
    0.62
     though
    0.61
     certific
    0.61
     Michele
    0.61
    Act Density 0.127%

    No Known Activations