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

    Neuronpedia

    INDEX
    Explanations

    glitch, GLUE, Glendale, glibc, glans, glides

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    grep
    0.42
    实际
    0.39
    澡
    0.39
    ব্যাপ
    0.39
    ৪
    0.37
    evidence
    0.36
     معه
    0.36
    appropriate
    0.35
    aspetto
    0.35
    ちゃんと
    0.35
    POSITIVE LOGITS
     gl
    1.01
     Gl
    0.93
    Gl
    0.93
     GL
    0.84
     гла
    0.80
     ग्ल
    0.79
    GL
    0.78
     ग्ले
    0.75
     glu
    0.74
     ग्ला
    0.74
    Act Density 0.031%

    No Known Activations