© 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. 34-GEMMASCOPE-RES-131K
    7. 93332
    Prev
    Next
    INDEX
    Explanations

    periods and breaks

    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_34/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
     at
    -1.27
    duma
    -0.82
     ditutup
    -0.80
    cec
    -0.78
     concepts
    -0.78
     daytime
    -0.76
     フード
    -0.76
     первого
    -0.73
     detuvo
    -0.72
     flooring
    -0.72
    POSITIVE LOGITS
     halftime
    1.33
     intermission
    1.29
    ORANGE
    0.79
    voerd
    0.78
     recess
    0.76
     oranges
    0.76
    捎
    0.75
     fı
    0.75
    inter
    0.75
    runOnUiThread
    0.75
    Activations Density 0.019%

    No Known Activations