© 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
    2. Gemma-2-27B
    3. 22-GEMMASCOPE-RES-131K
    4. 130890
    Prev
    Next
    INDEX
    Explanations

    The neuron responds strongly to programming‐style tokens, especially uppercase identifiers (like macros) and file‐path separators (slashes).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    technical/scientific terms

    np_acts-logits-general · gemini-2.5-flash-lite
    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
    (※
    -1.63
     with
    -1.41
     comes
    -1.41
     its
    -1.35
     becomes
    -1.34
    !!!!!
    -1.34
    !!!!
    -1.32
     when
    -1.24
     as
    -1.21
     seemingly
    -1.20
    POSITIVE LOGITS
     allerdings
    1.45
     tambin
    1.40
     longterm
    1.34
    
    1.33
     COULD
    1.31
     uph
    1.30
     toutefois
    1.29
    헴
    1.27
     EIGHT
    1.27
    哪个
    1.27
    Activations Density 0.000%

    No Known Activations