© 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. 25341
    Prev
    Next
    INDEX
    Explanations

    unit tests, test suites

    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
    搅拌
    -0.96
    的笑
    -0.94
    ős
    -0.89
    を持つ
    -0.84
     attempt
    -0.82
    明细
    -0.82
    </strong>
    -0.80
    ameters
    -0.79
     говорю
    -0.79
    观的
    -0.79
    POSITIVE LOGITS
    )-\
    1.04
    бочка
    0.98
     maailman
    0.96
     conoce
    0.92
    dubbo
    0.91
    不仅
    0.89
     reconnaît
    0.88
     mariée
    0.87
    França
    0.86
    Variante
    0.85
    Activations Density 0.015%

    No Known Activations