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

    Neuronpedia

    INDEX
    Explanations

    variable length sequences and accumulation

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    efic
    0.41
     pos
    0.39
    ຊີ
    0.38
     sor
    0.36
     syn
    0.36
    ifice
    0.36
     populace
    0.35
    ificado
    0.35
     schle
    0.34
     ниже
    0.34
    POSITIVE LOGITS
     JSONArray
    0.43
    ట్
    0.43
     Interstate
    0.43
     Andover
    0.42
     iteratively
    0.41
    dwell
    0.40
    innam
    0.39
     hesit
    0.39
    hamton
    0.39
     اتمنى
    0.38
    Act Density 0.001%

    No Known Activations