© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-2B
    5. 4-CLT-HP
    7. 477
    Prev
    Next
    INDEX
    Explanations

    about

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     which
    -0.68
    which
    -0.57
     Which
    -0.54
    Which
    -0.52
     meant
    -0.52
     venido
    -0.47
     cuja
    -0.47
     cuyo
    -0.45
     WHICH
    -0.45
     cuyos
    -0.44
    POSITIVE LOGITS
     AssemblyTitle
    0.83
     GreatSchools
    0.72
     فريبيس
    0.70
    transQ
    0.68
    RegressionTest
    0.66
    Története
    0.65
     '\\;'
    0.65
    iocene
    0.65
     JpaRepository
    0.64
    schemes
    0.63
    Activations Density 0.002%

    No Known Activations