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

    Neuronpedia

    INDEX
    Explanations

    ligma joke

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    er
    0.65
    い
    0.55
    र
    0.48
    ل
    0.47
     leuc
    0.46
    선
    0.45
    いる
    0.44
    erase
    0.43
    स्ट
    0.43
    いろ
    0.43
    POSITIVE LOGITS
     tersedia
    0.60
     terzo
    0.58
    fera
    0.55
     irritated
    0.52
    时候
    0.52
     prezzo
    0.52
     该
    0.51
    טים
    0.51
     salió
    0.51
    ਨ
    0.50
    Act Density 0.001%

    No Known Activations