© 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. 93056
    Prev
    Next
    INDEX
    Explanations

    code delimiters and dictionaries

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

    This neuron fires on programming‐language tokens—especially keywords, type names, identifiers, and sequences of punctuation (e.g. “if”, “var”, “string”, “Element”, “)))”, “;”) that occur in code snippets.

    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
     kült
    -0.85
    Royal
    -0.85
    出産
    -0.81
    tences
    -0.81
    谁能
    -0.79
     люди
    -0.79
     sustaining
    -0.77
     and
    -0.76
    宜蘭
    -0.75
    枇
    -0.74
    POSITIVE LOGITS
     bluse
    1.14
     have
    0.93
     CARTOON
    0.92
    斯的
    0.91
     madu
    0.91
     också
    0.91
     reconnaître
    0.90
     noexcept
    0.89
    omst
    0.88
     freddo
    0.88
    Activations Density 0.004%

    No Known Activations