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

    Neuronpedia

    INDEX
    Explanations

    chunk

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     sınav
    -0.08
    erer
    -0.08
    ati
    -0.08
     TEST
    -0.07
    idente
    -0.07
    _subscribe
    -0.07
     sosyal
    -0.07
     Sofia
    -0.07
    泽
    -0.07
     optical
    -0.07
    POSITIVE LOGITS
     chunk
    0.12
     chuck
    0.10
     chunks
    0.10
    chunk
    0.09
    Chuck
    0.09
     Chuck
    0.09
    Chunks
    0.09
     Chunk
    0.08
    chunks
    0.08
    _chunks
    0.08
    Act Density 0.005%

    No Known Activations