© 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. 73132
    Prev
    Next
    INDEX
    Explanations

    code or data link import

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

    The neuron fires on unusually long runs of whitespace (e.g. extended indentations or blank‐line spans).

    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
     two
    -1.11
     three
    -1.02
    </em>
    -1.01
     several
    -0.95
     where
    -0.95
    gunakan
    -0.95
     دندان
    -0.91
    理石
    -0.91
    TextMessage
    -0.89
     varios
    -0.89
    POSITIVE LOGITS
     Эти
    1.00
     приготовить
    0.99
     tramway
    0.96
     behandeling
    0.94
    ↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵
    0.92
     outlander
    0.88
    ↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵
    0.88
     artistico
    0.87
     Cæsar
    0.87
     schreef
    0.87
    Activations Density 0.004%

    No Known Activations