© 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. 10-GEMMASCOPE-RES-131K
    7. 100307
    Prev
    Next
    INDEX
    Explanations

    code symbols and citations

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

    technical or formal text containing special characters, particularly code, mathematical notation, or structured data formats.

    oai_token-act-pair · claude-4-5-sonnetTriggered by @bfgcn7frf5

    This neuron lights up on tokens that look like technical identifiers or acronyms—e.g. code names, file or function names, stock tickers, and other uppercase or punctuation-rich labels.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
     But
    -2.73
    鋮
    -2.69
    ↵↵
    -2.69
    大多
    -2.23
     Zwecke
    -2.16
    月下
    -2.14
     レスト
    -2.14
     lakini
    -2.13
    鳢
    -2.13
     tetapi
    -2.11
    POSITIVE LOGITS
    x
    2.50
    齮
    2.44
     they
    2.42
    你們
    2.33
     étan
    2.31
    as
    2.31
    i
    2.30
    かっ
    2.27
    付けて
    2.19
    してもら
    2.19
    Activations Density 0.000%

    No Known Activations