© 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. 119378
    Prev
    Next
    INDEX
    Explanations

    Q followed by specific words

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

    The neuron strongly activates on the “.” separators used in code (especially Java package/class qualifiers) and other statement delimiters like semicolons.

    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
    daad
    -0.78
     воспринима
    -0.77
    wnd
    -0.76
     отделения
    -0.75
     blunt
    -0.73
     Amherst
    -0.72
    acją
    -0.71
    鲛
    -0.71
    guero
    -0.70
     headaches
    -0.69
    POSITIVE LOGITS
    member
    1.01
    Member
    0.99
    Board
    0.98
     VO
    0.96
    VO
    0.93
     member
    0.90
     Member
    0.88
    Insets
    0.86
     пыле
    0.81
     Board
    0.80
    Activations Density 0.006%

    No Known Activations