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

    Neuronpedia

    INDEX
    Explanations

    Shrek, washing, ascending track

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     excuse
    1.26
    }$
    1.20
     sobra
    1.19
     dismiss
    1.13
    Ե
    1.13
    குதி
    1.13
    ﺏ
    1.12
    렉
    1.12
     concerns
    1.11
     איז
    1.07
    POSITIVE LOGITS
    𝗶
    1.68
    er
    1.67
    iid
    1.63
    vedere
    1.62
    ه
    1.53
    arikat
    1.50
    eine
    1.49
    能夠
    1.49
    iin
    1.47
    вропей
    1.43
    Act Density 0.005%

    No Known Activations