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

    Neuronpedia

    INDEX
    Explanations

    non-specific or issue

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     r
    0.45
    esign
    0.44
    í
    0.43
     pix
    0.43
    ayscale
    0.42
    t
    0.42
    inserted
    0.41
     t
    0.41
     unison
    0.41
    resized
    0.40
    POSITIVE LOGITS
    по
    0.49
     забо
    0.49
    Теперь
    0.48
    अब
    0.48
    Existe
    0.46
    Otra
    0.46
     גם
    0.45
     इतर
    0.45
     હવે
    0.45
     másik
    0.45
    Act Density 0.003%

    No Known Activations