© 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. 99407
    Prev
    Next
    INDEX
    Explanations

    column boards

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

    The neuron strongly activates on code‐style text (long runs of spaces/indentation and programming identifiers), i.e. it detects when you’re looking at a code block or technical snippet.

    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
     '''
    -0.81
    Glitch
    -0.77
    larry
    -0.76
    mickey
    -0.75
    importation
    -0.74
    Erosion
    -0.74
    logFile
    -0.74
    logfile
    -0.73
    颐
    -0.73
    mHandler
    -0.72
    POSITIVE LOGITS
    column
    1.32
     board
    1.28
     column
    1.27
    Board
    1.27
    Column
    1.23
    Swim
    1.18
     Board
    1.13
    Kan
    1.12
     columns
    1.11
     Kan
    1.03
    Activations Density 0.006%

    No Known Activations