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

    Neuronpedia

    INDEX
    Explanations

    Code

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     rush
    -0.07
     Protocol
    -0.07
     Evaluation
    -0.07
     interpol
    -0.07
    allocator
    -0.06
    outline
    -0.06
     synthesis
    -0.06
    Fallback
    -0.06
     scope
    -0.06
     Specialty
    -0.06
    POSITIVE LOGITS
       ↵
    0.07
     amaç
    0.07
    chied
    0.07
    ображ
    0.06
    ，让
    0.06
     ）↵
    0.06
     под
    0.06
     carrots
    0.06
    雙
    0.06
     النو
    0.06
    Act Density 0.091%

    No Known Activations