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

    Neuronpedia

    INDEX
    Explanations

    placeholder and replace instructions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     overkill
    0.46
     methane
    0.43
     थर्मल
    0.40
     arterial
    0.38
     turbines
    0.37
     ortho
    0.37
     Interference
    0.37
     turbine
    0.36
     milder
    0.36
     thermal
    0.36
    POSITIVE LOGITS
     placeholder
    1.02
     placeholders
    0.95
    placeholder
    0.95
     Placeholder
    0.94
    Placeholder
    0.93
    Replace
    0.70
     template
    0.68
    template
    0.68
    テンプレート
    0.67
    模板
    0.66
    Act Density 0.159%

    No Known Activations