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

    Neuronpedia

    INDEX
    Explanations

    training

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     Nova
    -0.07
     topic
    -0.07
    poons
    -0.07
    sects
    -0.07
     compounded
    -0.07
    picked
    -0.07
     Looking
    -0.07
     cos
    -0.07
     section
    -0.06
    acky
    -0.06
    POSITIVE LOGITS
     training
    0.12
     Training
    0.11
    training
    0.10
     trained
    0.08
     train
    0.08
    Training
    0.08
     onay
    0.08
     Andrews
    0.07
     Andrew
    0.07
    >Main
    0.07
    Act Density 0.048%

    No Known Activations