© 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. 22-GEMMASCOPE-RES-131K
    7. 54138
    Prev
    Next
    INDEX
    Explanations

    Dunrobin

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

    The neuron is primarily triggered by leading‐whitespace tokens (indentation) 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_22/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
     kabát
    -1.24
     peluche
    -1.18
     klinik
    -1.12
     ACTUALLY
    -1.11
     triko
    -1.09
     muñeco
    -1.08
     conforto
    -1.03
     bluza
    -1.01
     strukt
    -0.99
     instruk
    -0.99
    POSITIVE LOGITS
     it
    1.21
     by
    1.10
     some
    1.00
     while
    0.94
     above
    0.88
     if
    0.87
    
    0.85
    谁能
    0.85
    Some
    0.83
     stately
    0.82
    Activations Density 0.014%

    No Known Activations