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

    Neuronpedia

    INDEX
    Explanations

    breakdown followed by categorization

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     implementar
    0.64
    מה
    0.63
     відповідно
    0.63
     pourtant
    0.62
     tasked
    0.61
     implement
    0.60
     плану
    0.60
     solides
    0.59
    shrink
    0.59
     complex
    0.59
    POSITIVE LOGITS
    categor
    1.26
     category
    1.23
     categorized
    1.23
     categor
    1.16
     Categor
    1.14
     classified
    1.07
     categorie
    1.07
     categorization
    1.06
     CATEG
    1.05
    Categor
    1.04
    Act Density 0.303%

    No Known Activations