© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 26389
    Prev
    Next
    INDEX
    Explanations

    SQL `ON` clauses

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

    The neuron primarily fires on tokens that are numeric literals (i.e. sequences containing digits).

    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
    Soorten
    -1.16
    lossen
    -1.06
    *****/
    -1.04
    Maur
    -1.02
     comentar
    -1.02
    pport
    -1.02
     gelang
    -1.00
    ïc
    -1.00
    ≾
    -1.00
    柟
    -1.00
    POSITIVE LOGITS
     it
    0.99
    de
    0.98
    am
    0.96
    ss
    0.94
    to
    0.93
     a
    0.91
    en
    0.89
    o
    0.88
     people
    0.86
     bestimmte
    0.85
    Activations Density 0.018%

    No Known Activations