© 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. 20848
    Prev
    Next
    INDEX
    Explanations

    MG abbreviation or unit

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

    The neuron identifies "MG" followed by numbers, typically representing model numbers, product codes, or chemical compounds.

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

    The neuron detects occurrences of the uppercase bigram “MG” (as in abbreviations or names like “MG Road”).

    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
    that
    -2.30
    回想
    -2.27
    isations
    -2.16
    嫫
    -2.16
    我们
    -2.16
     Possibly
    -2.14
     Saying
    -2.14
    戆
    -2.11
    But
    -2.06
    箆
    -2.06
    POSITIVE LOGITS
    ’
    3.72
    ?"
    2.52
    ⧠
    2.20
    казания
    2.17
    纾
    2.11
    www
    2.08
     has
    2.08
     Для
    2.05
    жника
    2.03
    寕
    2.03
    Activations Density 0.006%

    No Known Activations