© 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. 25208
    Prev
    Next
    INDEX
    Explanations

    code constructs and definitions

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

    The neuron activates on SQL DDL keywords related to creating or dropping tables (e.g. CREATE TABLE, DROP TABLE, IF EXISTS).

    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
     their
    -2.13
     all
    -2.08
     only
    -2.06
     usually
    -2.02
     two
    -1.99
     mainly
    -1.99
     because
    -1.98
     mostly
    -1.95
     first
    -1.91
     almost
    -1.91
    POSITIVE LOGITS
     maniere
    1.68
     Dinas
    1.61
    菒
    1.57
     applau
    1.55
     bricola
    1.54
     interrom
    1.53
     delightfully
    1.53
    ſſen
    1.53
    訁
    1.53
    躄
    1.52
    Activations Density 0.023%

    No Known Activations