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

    Neuronpedia

    INDEX
    Explanations

    open-weights model

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    顺利
    0.39
    preparing
    0.39
    順
    0.39
     बराम
    0.38
    édéric
    0.37
    volvement
    0.36
    顺
    0.36
    还没
    0.36
    ネン
    0.35
    പാടി
    0.35
    POSITIVE LOGITS
     weights
    0.92
    Weights
    0.81
     Weights
    0.72
     internals
    0.71
    权重
    0.66
    weights
    0.65
     weight
    0.55
     insides
    0.52
     parameters
    0.50
    Weight
    0.49
    Act Density 0.027%

    No Known Activations