© 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
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 12968
    Prev
    Next
    INDEX
    Explanations

    code and configuration keywords

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

    The neuron spotlights programming identifiers—names of variables, types, and other code symbols (often CamelCased or underscored and sometimes including digits).

    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
     treino
    -1.01
     ehemal
    -1.00
    sapato
    -0.99
     ettiği
    -0.93
     outback
    -0.93
    fdata
    -0.93
     Regionen
    -0.92
    blems
    -0.92
     概要
    -0.92
    esticide
    -0.91
    POSITIVE LOGITS
    ☎
    1.09
     questo
    0.93
     álbum
    0.88
     Dung
    0.88
     „
    0.87
     guste
    0.85
     ,,
    0.85
     muchas
    0.84
     niveau
    0.83
     musique
    0.82
    Activations Density 0.032%

    No Known Activations