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

    Neuronpedia

    INDEX
    Explanations

    I

    np_max-act-logits · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ರ್ಭ
    0.41
     බල
    0.40
    elio
    0.39
     Hv
    0.39
     bois
    0.38
     Hoe
    0.38
     şə
    0.38
     лично
    0.38
    মালা
    0.38
    为例
    0.38
    POSITIVE LOGITS
     sorry
    0.57
     Sorry
    0.53
     knocked
    0.51
     signalé
    0.46
     Tidak
    0.46
     سوری
    0.46
    sorry
    0.44
    Sorry
    0.42
     am
    0.42
     نیست
    0.42
    Act Density 0.060%

    No Known Activations