© 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. 128721
    Prev
    Next
    INDEX
    Explanations

    code keywords and structures

    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
     váll
    -1.08
     verifique
    -0.95
    ↵↵↵↵↵
    -0.89
     heisst
    -0.88
    露出
    -0.88
     präsentiert
    -0.87
     valmis
    -0.86
     ”
    -0.85
    ↵↵↵↵↵↵↵↵↵↵↵↵
    -0.84
     bezieht
    -0.83
    POSITIVE LOGITS
    ディダス
    1.08
     اختلال
    1.00
     ofs
    0.98
     tantal
    0.97
     Bursa
    0.93
     TRX
    0.93
     オーダー
    0.92
    と思った
    0.92
     deo
    0.91
     hearth
    0.91
    Activations Density 0.022%

    No Known Activations