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

    Neuronpedia

    INDEX
    Explanations

    layers and decant

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    䍜
    0.56
     cuál
    0.47
    <unused77>
    0.46
     quién
    0.45
     दट
    0.45
     trigonometric
    0.45
    𝕥
    0.45
     ř
    0.44
    잍
    0.44
     ва
    0.44
    POSITIVE LOGITS
    h
    0.57
    F
    0.54
    С
    0.52
    在
    0.51
    bound
    0.51
    At
    0.50
    style
    0.50
    Ext
    0.50
    Ç
    0.49
    G
    0.49
    Act Density 0.000%

    No Known Activations