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

    Neuronpedia

    INDEX
    Explanations

    words and phrases expressing ethical judgment and professional behavior.

    oai_token-act-pair · gemini-2.0-flash

    Correctness

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    ReusableCell
    -0.96
     المعيارى
    -0.82
    writeFieldEnd
    -0.77
    ImageContext
    -0.77
    TestingModule
    -0.77
    Portale
    -0.75
    MLLoader
    -0.74
    tagHelperRunner
    -0.71
    +#+#
    -0.71
     متعلقه
    -0.69
    POSITIVE LOGITS
     correct
    1.20
    correct
    1.09
     proper
    1.09
     Correct
    1.01
    Correct
    0.98
    Proper
    0.94
     CORRECT
    0.94
    proper
    0.94
     right
    0.93
     Proper
    0.91
    Act Density 1.215%

    No Known Activations