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

    Neuronpedia

    INDEX
    Explanations

    parsing arguments and code structure

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    חי
    0.35
    াকা
    0.34
    ূ
    0.34
    庞
    0.34
    फ
    0.33
     বাবা
    0.33
    פ
    0.33
    劇場
    0.32
    ":[],
    0.32
    ໂ
    0.32
    POSITIVE LOGITS
     spice
    0.34
     determinar
    0.33
     Spice
    0.33
     Arias
    0.33
     zmian
    0.33
     Alpes
    0.32
     besonderen
    0.32
     Vorg
    0.32
     adjective
    0.31
     trat
    0.31
    Act Density 0.305%

    No Known Activations