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

    Neuronpedia

    INDEX
    Explanations

    easy, recovery, light activity

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ICIO
    0.41
     prisoners
    0.39
     परिषद
    0.38
     przedstaw
    0.37
     தாவ
    0.37
     prisoner
    0.37
     Ranking
    0.37
     contestants
    0.36
     imprison
    0.36
    yaratan
    0.36
    POSITIVE LOGITS
     easy
    0.75
     EASY
    0.73
     recovery
    0.72
     Easy
    0.72
     mileage
    0.69
    easy
    0.69
     taper
    0.66
    Easy
    0.66
    輕鬆
    0.66
     conversational
    0.64
    Act Density 0.030%

    No Known Activations