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

    Neuronpedia

    INDEX
    Explanations

    coding and development tasks

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    3
    0.50
    9
    0.45
    2
    0.44
    1
    0.44
    4
    0.42
     در
    0.40
     говорят
    0.39
    ב
    0.39
    6
    0.39
    7
    0.39
    POSITIVE LOGITS
     expedit
    0.45
     outset
    0.42
     without
    0.40
     workflows
    0.39
     또는
    0.39
     archi
    0.39
     but
    0.38
    🏗
    0.38
     intending
    0.38
     iteratively
    0.38
    Act Density 0.007%

    No Known Activations