© 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. 128928
    Prev
    Next
    INDEX
    Explanations

    perhaps, followed by emphasis

    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
    覘
    -1.84
     PARTE
    -1.84
    ”？
    -1.83
     sympa
    -1.81
     alin
    -1.76
    autique
    -1.76
     idyl
    -1.76
    歩道
    -1.73
    ânsito
    -1.73
     experimenta
    -1.73
    POSITIVE LOGITS
    3
    2.00
     sogar
    1.99
    4
    1.96
    8
    1.95
     will
    1.86
    呿
    1.78
     двох
    1.73
    erialize
    1.73
    嵛
    1.72
     only
    1.70
    Activations Density 0.011%

    No Known Activations