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

    Neuronpedia

    INDEX
    Explanations

    assumptions and potential observations

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    et
    0.52
     sticky
    0.50
     lom
    0.49
     reread
    0.47
     dur
    0.47
     risky
    0.45
     Sticky
    0.45
     PHP
    0.45
     غير
    0.45
    y
    0.45
    POSITIVE LOGITS
    ین
    0.52
     असल्यास
    0.52
    Discover
    0.49
     olympiques
    0.49
    Fare
    0.49
    ด
    0.49
    𝗥
    0.49
    STATES
    0.48
     राजांनी
    0.48
    LookAnd
    0.48
    Act Density 0.000%

    No Known Activations