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

    Neuronpedia

    INDEX
    Explanations

    numeric and alphanumeric patterns

    oai_token-act-pair · gpt-3.5-turbo

    instances of irony or sarcasm

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    nces
    -0.76
    ournals
    -0.70
    endale
    -0.67
     Hannity
    -0.66
    hess
    -0.64
    enhagen
    -0.62
    cair
    -0.61
     Hale
    -0.61
    zynski
    -0.60
    SHARE
    -0.59
    POSITIVE LOGITS
    ãĥ³ãĤ¸
    0.85
    izoph
    0.75
    usterity
    0.73
    leneck
    0.68
    tenance
    0.67
    idepress
    0.66
     distingu
    0.66
    ollah
    0.66
    chest
    0.65
    WithNo
    0.64
    Act Density 1.042%

    No Known Activations