© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 48900
    Prev
    Next
    INDEX
    Explanations

    nginx, http, aliases, or Windows

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

    The neuron strongly activates on code‐like tokens—especially uppercase identifiers and macro names (e.g. HALIDE, IDE, LAMBDA_H) and other programming symbols.

    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
    ♑
    -0.70
    ೀ
    -0.69
    ruppen
    -0.69
     Sue
    -0.67
    吓
    -0.67
     intimidate
    -0.67
    xmin
    -0.66
     Ridley
    -0.66
    介
    -0.66
     nay
    -0.66
    POSITIVE LOGITS
     jät
    0.71
     matter
    0.71
     unequivocally
    0.70
    ينه
    0.68
    这篇
    0.64
     merec
    0.64
     Matter
    0.64
     boule
    0.63
     urządzenia
    0.63
    ingresso
    0.62
    Activations Density 0.066%

    No Known Activations