© 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
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 104909
    Prev
    Next
    INDEX
    Explanations

    programming code and errors

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

    underscores in programming code identifiers, particularly in C++ and similar languages.

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

    The neuron fires on code‐style identifiers containing underscores (e.g. macro or function names, header filenames).

    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
    '
    -2.16
     ‘
    -1.74
     incluso
    -1.64
    ’
    -1.60
    end
    -1.58
     fichiers
    -1.55
    ‘
    -1.52
     
    -1.52
    の作成
    -1.52
     themselves
    -1.52
    POSITIVE LOGITS
    Skocz
    1.70
    Список
    1.67
    ヤ
    1.65
    bilden
    1.63
     動
    1.59
    ampaikan
    1.59
    VIATIONS
    1.57
    poň
    1.56
    ujem
    1.52
     tradicionales
    1.52
    Activations Density 0.001%

    No Known Activations