© 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. 99070
    Prev
    Next
    INDEX
    Explanations

    drop out or withdrawal

    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_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
     construire
    -1.69
     insidious
    -1.52
    pensione
    -1.48
    仍然
    -1.47
     baffling
    -1.46
    可以
    -1.45
     penser
    -1.45
     opět
    -1.44
     imaginar
    -1.43
    非常
    -1.41
    POSITIVE LOGITS
     on
    1.89
     in
    1.59
     using
    1.48
     pursuing
    1.45
     there
    1.44
     providing
    1.43
     for
    1.42
    pecabe
    1.41
    s
    1.39
     into
    1.38
    Activations Density 0.022%

    No Known Activations