© 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. 12727
    Prev
    Next
    INDEX
    Explanations

    code keywords and symbols

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

    The neuron spikes on code‐style identifiers and keywords—especially long C/C++ tokens like typedef, struct, C++, and similar symbol‐heavy names.

    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
     完了
    -2.09
    …….
    -1.86
     ….
    -1.84
    ……….
    -1.76
    ……..
    -1.63
    ………..
    -1.53
     ……
    -1.52
    and
    -1.52
    ….
    -1.50
    味的
    -1.50
    POSITIVE LOGITS
    )—
    1.66
    Ž
    1.48
     $^{\
    1.43
    🔺
    1.41
     booming
    1.41
    ⬇
    1.39
     =
    1.38
    ☆☆☆☆☆
    1.33
    –
    1.33
    ☑
    1.32
    Activations Density 39.119%

    No Known Activations