© 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-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 112371
    Prev
    Next
    INDEX
    Explanations

    debug and trace logging

    np_acts-logits-general · gemini-2.5-flash-lite

    debugging-related code elements, particularly debug flags, verbose settings, and trace statements in programming contexts.

    oai_token-act-pair · claude-4-5-sonnetTriggered by @bfgcn7frf5
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    :
    -1.58
     famosa
    -1.48
     longtime
    -1.48
     that
    -1.38
     todėl
    -1.37
     various
    -1.35
     steadfast
    -1.33
     K
    -1.30
     übernimmt
    -1.30
     esteemed
    -1.30
    POSITIVE LOGITS
    umerable
    1.58
     RICHARD
    1.48
     and
    1.45
     Sociale
    1.45
    Algemeen
    1.44
     RALPH
    1.43
     każdego
    1.42
     toep
    1.42
     ALAN
    1.42
     zoude
    1.41
    Activations Density 0.015%

    No Known Activations