© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 85922
    Prev
    Next
    INDEX
    Explanations

    ElIZABETH HEALTH CENTER

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

    citations and references, including footnotes, scientific notation, and academic references in text.

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

    The neuron fires on academic‐style citation and technical‐jargon tokens—author names, footnote markers, reference markers, gene or chemical names, and formula fragments.

    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
     a
    -1.99
     their
    -1.68
     that
    -1.54
     with
    -1.52
     as
    -1.52
    ggars
    -1.52
    当时
    -1.39
     another
    -1.37
     what
    -1.35
     today
    -1.35
    POSITIVE LOGITS
     biedt
    1.63
    回目の
    1.59
     даже
    1.54
    HYDR
    1.46
    Geschiedenis
    1.42
    Comté
    1.42
    FORME
    1.36
     même
    1.34
     azule
    1.34
    特的
    1.33
    Activations Density 0.582%

    No Known Activations