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

    Neuronpedia

    INDEX
    Explanations

    as you can see

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Seeing
    0.89
    seeing
    0.88
     seeing
    0.88
    Seeing
    0.88
    Saw
    0.85
     saw
    0.82
     Saw
    0.80
    saw
    0.72
     دید
    0.66
     SAW
    0.66
    POSITIVE LOGITS
     plainly
    0.59
     plain
    0.51
     clearly
    0.49
     ఇప్పుడు
    0.45
    plain
    0.45
     hopefully
    0.43
    clearly
    0.42
     clairement
    0.42
     claramente
    0.42
     ansatz
    0.41
    Act Density 0.007%

    No Known Activations