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

    Neuronpedia

    INDEX
    Explanations

    transformation

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     reluctantly
    -0.08
     подар
    -0.08
     aega
    -0.08
     rusa
    -0.08
     დარჩ
    -0.08
     novice
    -0.08
    迷
    -0.08
     statues
    -0.08
     turn
    -0.08
    -sav
    -0.08
    POSITIVE LOGITS
    _estim
    0.09
     CUDA
    0.09
    Interfaces
    0.09
    Parameterized
    0.08
     Tensor
    0.08
    .sqlite
    0.08
    Middleware
    0.08
    Reducers
    0.08
    Transformer
    0.08
    0.08
    Act Density 0.277%
    No Known Activations