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

    Neuronpedia

    INDEX
    Explanations

    bracketed placeholders

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     lessons
    0.81
     bundles
    0.78
     latency
    0.76
     shutdown
    0.74
     analges
    0.74
     sluggish
    0.73
     tests
    0.73
     zombies
    0.72
     integer
    0.72
     latent
    0.72
    POSITIVE LOGITS
    ometown
    1.25
     Name
    1.19
     Dates
    1.18
     Date
    1.15
     Bezeichnung
    1.13
     Organizational
    1.12
     Address
    1.11
     Contact
    1.07
     должность
    1.07
     полномо
    1.07
    Act Density 0.285%

    No Known Activations