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

    Neuronpedia

    INDEX
    Explanations

    list items or sentences

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ակ
    0.39
     optional
    0.37
     suboptimal
    0.35
    ઐ
    0.35
     möglicherweise
    0.34
     ხოლო
    0.33
     sections
    0.33
     possibly
    0.33
     caveats
    0.32
     seterusnya
    0.32
    POSITIVE LOGITS
     рыб
    0.40
    オープ
    0.37
     гром
    0.36
     Instructor
    0.36
     ră
    0.36
     صدای
    0.36
     শ
    0.36
     যে
    0.35
     рав
    0.35
     आठवीं
    0.35
    Act Density 0.022%

    No Known Activations