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

    Neuronpedia

    INDEX
    Explanations

    technical keywords and code context

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ባቸው
    0.43
    펌
    0.40
    ovsk
    0.39
     easements
    0.39
    Easing
    0.38
    čar
    0.38
    𝓮
    0.38
     landings
    0.37
     caption
    0.36
    ulen
    0.36
    POSITIVE LOGITS
     reg
    0.40
     Fill
    0.38
     Reg
    0.36
     cycle
    0.36
    Fill
    0.35
     никак
    0.35
    ා
    0.35
     Level
    0.34
    führer
    0.34
     Mongolian
    0.34
    Act Density 0.000%

    No Known Activations