© 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. 22-GEMMASCOPE-RES-131K
    7. 130370
    Prev
    Next
    INDEX
    Explanations

    The main thing this neuron does is detect commas (punctuation marks separating clauses).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    comma then word

    np_acts-logits-general · gemini-2.5-flash-lite
    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
     äldre
    -1.12
    楽器
    -1.11
     gånger
    -1.08
     الإن
    -1.06
    そして
    -1.04
     समीक्षक
    -1.04
     särsk
    -1.02
    لیت
    -1.02
    記述
    -1.02
     egna
    -1.01
    POSITIVE LOGITS
     orizz
    1.16
     يون
    1.11
     acreditar
    1.10
     melalui
    1.09
     there
    1.09
     Cię
    1.09
     faceva
    1.05
     you
    1.05
    cuerpo
    1.05
     piccole
    1.03
    Activations Density 0.004%

    No Known Activations