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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 41920
    Prev
    Next
    INDEX
    Explanations

    metoclopramide, perioperative, nicotine

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

    words with diacritical marks or special characters.

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

    The neuron activates on tokens containing accented characters (especially the “Mé” prefix), i.e., words with diacritics.

    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
     ''
    -2.67
     Beſ
    -2.38
     BOTH
    -2.38
    I
    -2.30
    柃
    -2.25
    溷
    -2.25
     мужчина
    -2.23
     """
    -2.23
    鲮
    -2.23
    ''
    -2.22
    POSITIVE LOGITS
     has
    2.98
    ⤿
    2.58
    的一个
    2.45
    殽
    2.34
    ”،
    2.33
    
    2.31
    萑
    2.30
     beginnings
    2.28
     leichten
    2.23
    .
    2.20
    Activations Density 0.010%

    No Known Activations