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

    Neuronpedia

    INDEX
    Explanations

    "so to speak" and "figuratively"

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    FR
    -0.07
    commands
    -0.07
     באות
    -0.06
    สงคราม
    -0.06
     בעיר
    -0.06
    계
    -0.06
    昏
    -0.06
     ihrer
    -0.06
    cow
    -0.06
    這種
    -0.06
    POSITIVE LOGITS
    本质
    0.07
    Storyboard
    0.07
    arie
    0.07
     Vert
    0.07
     Nikol
    0.07
     Sampler
    0.06
    -s
    0.06
     ask
    0.06
    terior
    0.06
     Instant
    0.06
    Act Density 0.008%

    No Known Activations