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

    Neuronpedia

    INDEX
    Explanations

    quantifying technical terms

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    раль
    0.42
    ៅ
    0.41
     endonuclease
    0.41
    Aldrich
    0.39
    ᕈ
    0.39
    ᒃ
    0.38
    риса
    0.37
    либ
    0.37
     sailboat
    0.37
    рт
    0.36
    POSITIVE LOGITS
     four
    0.51
     These
    0.50
     Four
    0.49
    それぞれの
    0.47
     Three
    0.47
     all
    0.46
     todas
    0.46
     Clearly
    0.45
     Each
    0.45
     tutte
    0.45
    Act Density 0.436%

    No Known Activations