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

    Neuronpedia

    INDEX
    Explanations

    explicitly diverge dual

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    json
    0.52
     Hayward
    0.51
     فرمایا
    0.49
    खबर
    0.48
    最为
    0.48
    明
    0.48
    နာ
    0.47
     четы
    0.46
    灣
    0.46
    星期
    0.46
    POSITIVE LOGITS
     antidepressants
    0.49
     recycle
    0.48
    `/
    0.47
     towel
    0.46
     mildew
    0.45
     recycler
    0.44
     cell
    0.44
    °,
    0.43
    \*
    0.43
    \
    0.43
    Act Density 0.000%

    No Known Activations