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

    Neuronpedia

    INDEX
    Explanations

    strings "No" or "Ha ha" that often indicate disagreement or negativity

    oai_token-act-pair · gemini-2.0-flash

    No

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     سكانية
    -0.69
     <=",
    -0.67
     still
    -0.57
    liothèque
    -0.54
     AssemblyCulture
    -0.53
    fromJson
    -0.53
    principalColumn
    -0.52
    GEBURTSDATUM
    -0.52
    mistak
    -0.51
    ########.
    -0.51
    POSITIVE LOGITS
    Problem
    0.73
     problem
    0.73
     Problem
    0.67
     probleem
    0.66
     PROBLEM
    0.65
     Probl
    0.64
    problem
    0.63
     probleme
    0.59
     problems
    0.58
     problema
    0.58
    Act Density 0.534%

    No Known Activations