© 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. 13271
    Prev
    Next
    INDEX
    Explanations

    code and special characters

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

    The neuron flags programming‐language tokens—especially code constructs like preprocessor directives, generic angle brackets, and common identifiers—indicating that it detects 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
    From
    -2.14
    when
    -2.09
    Хороший
    -1.95
    When
    -1.83
    Having
    -1.80
    I
    -1.78
    焢
    -1.76
    Nearly
    -1.75
     this
    -1.71
    Most
    -1.70
    POSITIVE LOGITS
     museos
    1.66
     escritores
    1.62
     integrantes
    1.56
     alrededores
    1.55
     complementos
    1.52
    🪆
    1.52
    籥
    1.50
    kyat
    1.48
     yanı
    1.45
     antiguos
    1.44
    Activations Density 0.090%

    No Known Activations