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

    Neuronpedia

    INDEX
    Explanations

    rules, improving, due to, underlying structure

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Calculations
    0.41
    epine
    0.39
     degassing
    0.37
    ificance
    0.36
     inhomogeneity
    0.36
    ock
    0.36
     উহ
    0.36
    deg
    0.36
     calculations
    0.36
    ieu
    0.36
    POSITIVE LOGITS
     sonra
    0.43
     üzerinde
    0.42
     Margaret
    0.40
     parten
    0.39
     żeby
    0.39
     própria
    0.39
    ലെ
    0.38
     సంవత్సర
    0.38
     অথবা
    0.38
    Margaret
    0.37
    Act Density 0.000%

    No Known Activations