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

    Neuronpedia

    INDEX
    Explanations

    you'll find or get

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     zorgt
    1.02
     tries
    0.88
     helps
    0.83
     помогают
    0.83
     Helps
    0.82
     seeks
    0.81
     помогает
    0.81
     Tells
    0.81
     lets
    0.79
     ayud
    0.78
    POSITIVE LOGITS
     encounter
    1.52
     find
    1.40
    encounter
    1.29
    find
    1.24
    遇到
    1.17
     encounters
    1.13
     encountering
    1.11
     finds
    1.09
     discover
    1.08
     finden
    1.04
    Act Density 0.280%

    No Known Activations