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

    Neuronpedia

    INDEX
    Explanations

    names like Charlie and Susie

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    an
    3.85
    ان
    2.60
    Џ
    2.59
    т
    2.59
    на
    2.58
    م
    2.55
    ين
    2.48
    ല
    2.47
    ен
    2.46
    า
    2.44
    POSITIVE LOGITS
    ต้อง
    2.64
    런
    2.60
    ń
    2.54
    ați
    2.52
    ी
    2.50
    ٔ
    2.44
    glycer
    2.39
    ा
    2.39
     caranya
    2.39
    们
    2.33
    Act Density 0.098%

    No Known Activations