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

    Neuronpedia

    INDEX
    Explanations

    lmsys.org and Vicuna

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Sund
    0.38
    bru
    0.37
    file
    0.36
    zo
    0.36
    Tap
    0.35
     Sund
    0.34
    转化
    0.34
    Bru
    0.34
    lines
    0.34
    Pend
    0.33
    POSITIVE LOGITS
     Sts
    0.41
     Interference
    0.39
    ogenen
    0.39
     Lant
    0.39
     roofing
    0.38
     Lan
    0.38
     ಅರ್ಜಿ
    0.38
     sœurs
    0.38
     reserv
    0.37
     ইনস্টি
    0.37
    Act Density 0.004%

    No Known Activations