© 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. 34-GEMMASCOPE-RES-131K
    7. 59976
    Prev
    Next
    INDEX
    Explanations

    classifier.fit

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

    The neuron is strongly activated by programming-style tokens (e.g., code keywords, identifiers, and syntax) indicating it detects code or code-like text.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
     openers
    -0.81
     Lire
    -0.73
    ネ
    -0.73
    FontAwesome
    -0.72
    вок
    -0.70
     ネ
    -0.69
    ídio
    -0.69
    寺
    -0.68
    Про
    -0.68
     Schicht
    -0.68
    POSITIVE LOGITS
     fit
    3.77
     fitted
    3.64
    fit
    3.45
     fitting
    3.16
    fitted
    3.14
     Fit
    2.97
    Fit
    2.89
    fitting
    2.75
    Fitting
    2.73
     fits
    2.64
    Activations Density 0.020%

    No Known Activations