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

    Neuronpedia

    INDEX
    Explanations

    intensity in expressions of emotion, particularly the word "really."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    really + following words

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ocate
    -0.50
     kiosk
    -0.48
     AssemblyCompany
    -0.48
     enrollment
    -0.48
    ContentLoaded
    -0.46
     Enrollment
    -0.46
    Inc
    -0.46
     kios
    -0.45
     enrolment
    -0.45
     HasFactory
    -0.45
    POSITIVE LOGITS
    really
    1.14
     REALLY
    1.03
     really
    1.02
     Really
    0.96
    Really
    0.96
     realy
    0.81
    truly
    0.76
     truly
    0.72
    benar
    0.69
     Truly
    0.68
    Act Density 0.014%

    No Known Activations