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

    Neuronpedia

    INDEX
    Explanations

    state of being or processing

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     efectuar
    0.20
     проводить
    0.19
     informar
    0.19
     onları
    0.19
     tört
    0.19
     jaunes
    0.19
     kullanarak
    0.19
     fick
    0.18
     favorables
    0.18
     fikk
    0.18
    POSITIVE LOGITS
     treated
    0.39
     kept
    0.38
     handled
    0.37
     taken
    0.34
     brought
    0.33
     tackled
    0.32
     pushed
    0.32
     examined
    0.31
     given
    0.31
     put
    0.31
    Act Density 0.133%

    No Known Activations