© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Gemma-2-27B
    3. 22-GEMMASCOPE-RES-131K
    4. 35645
    Prev
    Next
    INDEX
    Explanations

    JSDoc parameter types

    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
    //};
    -1.36
     under
    -1.21
     about
    -1.21
    を書く
    -1.21
     Verständ
    -1.19
    regalo
    -1.18
     städ
    -1.16
    ∷
    -1.10
    aceite
    -1.10
     around
    -1.10
    POSITIVE LOGITS
    </strong>
    1.64
    </i>
    1.34
    or
    1.27
     henne
    1.23
     problemet
    1.22
     lyder
    1.21
     finner
    1.18
     herrera
    1.15
     honom
    1.14
     uan
    1.14
    Activations Density 0.002%

    No Known Activations