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

    Neuronpedia

    INDEX
    Explanations

    sharpness or precision in descriptions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ỡng
    -0.78
    Mule
    -0.72
     Mule
    -0.70
    soa
    -0.70
     Fonda
    -0.68
    Tube
    -0.66
    mule
    -0.66
     Yolanda
    -0.65
    intree
    -0.64
     cloud
    -0.64
    POSITIVE LOGITS
     Sharp
    1.49
     sharp
    1.44
    sharp
    1.39
    Sharp
    1.29
     SHARP
    1.28
     sharper
    1.23
     sharpness
    1.20
     sharpening
    1.11
     sharpen
    1.11
     Sharpe
    1.08
    Act Density 0.078%

    No Known Activations