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

    Neuronpedia

    INDEX
    Explanations

    introduces or describes concepts

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    洿
    0.48
     optimizes
    0.44
     Auflage
    0.44
     tailles
    0.43
    悆
    0.43
    ূল
    0.43
    Raster
    0.43
     bőr
    0.42
     Optimize
    0.42
     Elektr
    0.42
    POSITIVE LOGITS
    ست
    0.49
    하자
    0.46
     जेव्हा
    0.45
    fæ
    0.45
    when
    0.45
    र्म
    0.45
     त्याला
    0.44
    他和
    0.44
     layak
    0.44
    ፧
    0.44
    Act Density 0.000%

    No Known Activations