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

    Neuronpedia

    INDEX
    Explanations

    expressions of surprise or emphasis, particularly variations of "oh."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     faſt
    -0.69
     againſt
    -0.62
     Theſe
    -0.60
     abstrait
    -0.59
     يتيمه
    -0.58
     AssemblyTitle
    -0.58
     dezelve
    -0.57
     raiſ
    -0.57
     tričko
    -0.57
     eventName
    -0.56
    POSITIVE LOGITS
    oh
    0.82
     oh
    0.70
     OH
    0.62
    OH
    0.62
    dex
    0.59
     Oh
    0.58
    je
    0.57
     Jiang
    0.55
    ust
    0.54
    Oh
    0.53
    Act Density 0.296%

    No Known Activations