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

    Neuronpedia

    INDEX
    Explanations

    optimizer steps

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     venerable
    0.47
     finger
    0.46
     mActivity
    0.45
     steeple
    0.45
     provenance
    0.44
     waterfowl
    0.44
    )
    0.44
     apres
    0.43
     access
    0.43
     marble
    0.43
    POSITIVE LOGITS
    铧
    0.56
    ці
    0.52
     Archived
    0.43
    galaxys
    0.42
    cheid
    0.42
    zwischen
    0.42
    خدم
    0.41
    ცია
    0.41
    Wins
    0.41
     देणे
    0.40
    Act Density 0.000%

    No Known Activations