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

    Neuronpedia

    INDEX
    Explanations

    past states of being

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     जातात
    2.01
     करतात
    1.97
     دارند
    1.80
     देतात
    1.74
     सांगतात
    1.73
     येतात
    1.72
     असतात
    1.72
     помогают
    1.72
     позволяют
    1.68
     होतात
    1.68
    POSITIVE LOGITS
     was
    3.26
    was
    2.30
     wasn
    2.29
     came
    2.10
     был
    2.04
     had
    2.02
     became
    2.02
     était
    1.98
     began
    1.98
     была
    1.94
    Act Density 0.132%

    No Known Activations