© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 27669
    Prev
    Next
    INDEX
    Explanations

    fashion and design

    np_acts-logits-general · gemini-2.5-flash-lite

    The neuron fires on URL fragments and surrounding navigation/instruction text (e.g. “http://…”, domain names, and words like “Download,” “manual,” “locating”) that make up hyperlinks or link‐related directions.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    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
     моего
    -1.18
    gigabyte
    -1.09
     what
    -1.01
    conocimiento
    -1.01
     предприятий
    -1.00
     път
    -1.00
     черный
    -1.00
    abbix
    -0.99
    gamis
    -0.98
     Pří
    -0.98
    POSITIVE LOGITS
     grotte
    1.20
    pertino
    1.14
    ściu
    1.13
     promesse
    1.11
     {$\
    1.07
    ствия
    1.04
    ências
    1.02
     cliquer
    1.02
     legislación
    1.02
     ſu
    1.02
    Activations Density 0.005%

    No Known Activations