© 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. 70340
    Prev
    Next
    INDEX
    Explanations

    Thomas Flynn begin

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

    This neuron picks up long runs of whitespace characters (indentation) at the starts of lines.

    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
     still
    -0.87
     from
    -0.84
     zaterdag
    -0.83
    =>$
    -0.80
     giornata
    -0.80
     modelAndView
    -0.79
     сахар
    -0.79
     Lakukan
    -0.79
    logisch
    -0.78
    كتور
    -0.77
    POSITIVE LOGITS
     شروع
    1.45
     begun
    1.33
     commenced
    1.31
    開始
    1.26
    开始
    1.23
     began
    1.21
     commencing
    1.14
     commences
    1.12
     started
    1.11
    began
    1.11
    Activations Density 0.000%

    No Known Activations