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

    Neuronpedia

    INDEX
    Explanations

    warm

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    reten
    -0.07
     ctl
    -0.07
    cell
    -0.07
    Cell
    -0.07
     trivia
    -0.07
    viewController
    -0.06
    fuck
    -0.06
    comment
    -0.06
    _bits
    -0.06
     crawler
    -0.06
    POSITIVE LOGITS
     warm
    0.13
    Warm
    0.12
     Warm
    0.11
     warmer
    0.09
     warmth
    0.09
    warm
    0.08
     warmed
    0.08
     Thermal
    0.07
    凉
    0.07
     гар
    0.07
    Act Density 0.010%

    No Known Activations