© 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. 22-GEMMASCOPE-RES-131K
    7. 127987
    Prev
    Next
    INDEX
    Explanations

    The neuron is keyed to programming‐language tokens—especially JavaScript/Angular identifiers, keywords, and symbols—marking code snippets rather than ordinary text.

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

    angular controller or component definition

    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
     just
    -1.09
     more
    -1.05
     during
    -1.05
     securing
    -1.02
     if
    -1.02
     for
    -1.01
     exuberant
    -0.97
     numerous
    -0.96
     both
    -0.96
     epitom
    -0.95
    POSITIVE LOGITS
     heute
    1.08
     décisions
    1.07
     coûts
    1.05
    State
    1.05
     تصميم
    1.04
    howe
    1.02
     avancer
    1.02
    ี้ยง
    1.01
    
    1.00
     partager
    0.99
    Activations Density 0.007%

    No Known Activations