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

    Neuronpedia

    INDEX
    Explanations

    string extraction

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    ે
    -0.08
    (range
    -0.08
    aran
    -0.07
     touring
    -0.07
    หล
    -0.07
     Supplements
    -0.07
     applaud
    -0.07
     stunt
    -0.07
     sanctioned
    -0.07
    imenta
    -0.07
    POSITIVE LOGITS
     delim
    0.15
     delimiter
    0.14
    找到
    0.13
    Delimiter
    0.12
     Най
    0.12
     найти
    0.12
     separators
    0.12
     separator
    0.12
     occurrences
    0.12
     delimit
    0.12
    Act Density 0.011%

    No Known Activations