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

    Neuronpedia

    INDEX
    Explanations

    Ruby/ActiveRecord models

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    haul
    -0.07
    columns
    -0.07
    _TARGET
    -0.07
    ौद
    -0.07
    GU
    -0.06
     orthogonal
    -0.06
    нення
    -0.06
     Diamonds
    -0.06
    train
    -0.06
    hong
    -0.06
    POSITIVE LOGITS
    assage
    0.06
     curves
    0.06
    雑
    0.06
     计
    0.06
     πέ
    0.06
     ]).
    0.06
     тих
    0.06
     结
    0.05
     chấm
    0.05
    ันต
    0.05
    Act Density 0.007%

    No Known Activations