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

    Neuronpedia

    INDEX
    Explanations

    git, group, Greek

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    <unused2147>
    0.65
     வான
    0.62
    த்தனை
    0.61
    আত্ম
    0.60
    úrate
    0.60
     পাঁ
    0.58
    PathComponent
    0.58
     মণ
    0.56
     અભ
    0.56
    釵
    0.55
    POSITIVE LOGITS
    G
    1.96
    GC
    1.82
    Gs
    1.81
     G
    1.78
    GR
    1.78
    Г
    1.77
    GS
    1.76
     Gr
    1.74
    GI
    1.73
    GP
    1.73
    Act Density 4.477%

    No Known Activations