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

    Neuronpedia

    INDEX
    Explanations

    train

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     erm
    -0.07
     zcela
    -0.07
     erupted
    -0.07
    umber
    -0.07
    Swap
    -0.06
     Jeb
    -0.06
    odule
    -0.06
     rapport
    -0.06
     foss
    -0.06
    综合
    -0.06
    POSITIVE LOGITS
     train
    0.10
     Train
    0.08
    音
    0.07
     statistically
    0.07
    ิถ
    0.07
    イン
    0.07
     fish
    0.07
            ↵        ↵
    0.07
    िण
    0.07
    鋼
    0.06
    Act Density 0.008%

    No Known Activations