© 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. 34-GEMMASCOPE-RES-131K
    7. 99791
    Prev
    Next
    INDEX
    Explanations

    search query parameter

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

    The neuron is picking out programming‐style tokens—especially code symbols (parentheses, commas, function names, etc.) and other snippets of source code.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
    манов
    -0.81
    kne
    -0.80
    дарь
    -0.74
     さま
    -0.72
     Jungs
    -0.72
    numsp
    -0.72
    Kne
    -0.71
    tyka
    -0.71
    Decline
    -0.70
    Dann
    -0.69
    POSITIVE LOGITS
     term
    2.39
     query
    2.31
     keywords
    2.14
     keyword
    2.00
    Term
    1.99
    query
    1.95
    term
    1.92
     terms
    1.91
    Query
    1.83
     Term
    1.68
    Activations Density 0.061%

    No Known Activations