© 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. 33622
    Prev
    Next
    INDEX
    Explanations

    journal citations and drug names

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

    numerical references in academic or scientific texts, particularly publication years or section numbers.

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi

    This neuron detects numeric tokens (years, page numbers, percentages, and other standalone numbers).

    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
     just
    -1.08
     every
    -1.06
     usługi
    -1.05
     at
    -0.99
     because
    -0.97
    🌻
    -0.95
     mellem
    -0.95
     through
    -0.95
    kait
    -0.93
     in
    -0.92
    POSITIVE LOGITS
     drug
    1.27
     potenti
    0.99
     drugs
    0.98
    drug
    0.94
     culin
    0.93
    otropy
    0.93
     Pharmacol
    0.92
    ,;
    0.92
    NotNull
    0.91
    ódromo
    0.91
    Activations Density 0.001%

    No Known Activations