© 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. 22-GEMMASCOPE-RES-131K
    7. 26654
    Prev
    Next
    INDEX
    Explanations

    starting position or penalty

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
     augusti
    -1.15
     ロン
    -1.12
    sekten
    -1.08
    nachweise
    -1.01
    message
    -1.00
    olanda
    -0.99
     Endgame
    -0.99
    cisi
    -0.98
     sagde
    -0.98
    нент
    -0.97
    POSITIVE LOGITS
     will
    1.48
     position
    1.30
     on
    1.17
     préfère
    1.14
     memudahkan
    1.07
     avantaj
    1.06
    \,,
    1.05
     виправи
    1.05
     ahead
    1.04
     for
    1.02
    Activations Density 0.014%

    No Known Activations