© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 22501
    Prev
    Next
    INDEX
    Explanations

    code structure punctuation

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

    The neuron detects JavaScript “arrow‐function” syntax, i.e. the occurrence of “(…) =>” in the 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
    Should
    -1.73
    Essentially
    -1.72
    Without
    -1.70
    焿
    -1.67
     PRACT
    -1.66
    Whether
    -1.64
     EMERG
    -1.62
    Once
    -1.57
    Contrary
    -1.57
    羮
    -1.56
    POSITIVE LOGITS
     on
    2.61
     мужская
    2.08
    értelmű
    1.90
     Schauspielerin
    1.75
     женская
    1.71
     детская
    1.66
     especially
    1.64
     as
    1.63
     Eröffnung
    1.61
     Freundschaft
    1.59
    Activations Density 0.007%

    No Known Activations