© 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. 10-GEMMASCOPE-RES-131K
    7. 111274
    Prev
    Next
    INDEX
    Explanations

    code punctuation and keywords

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

    the ternary operator (question mark colon construct) in programming code.

    oai_token-act-pair · claude-4-5-sonnetTriggered by @bfgcn7frf5

    This neuron responds to numeric literal tokens (numbers) in code.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
    鵙
    -1.36
    依然
    -1.35
    チュウ
    -1.32
     égard
    -1.31
     település
    -1.30
    却
    -1.30
     DiCaprio
    -1.27
    atürk
    -1.27
    
    -1.27
     fortsatt
    -1.26
    POSITIVE LOGITS
     When
    1.55
    when
    1.35
     on
    1.32
     Fußballspieler
    1.26
     Although
    1.23
     jabatan
    1.20
     жі
    1.20
     kiedy
    1.19
     when
    1.16
     During
    1.16
    Activations Density 0.005%

    No Known Activations