© 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. 60083
    Prev
    Next
    INDEX
    Explanations

    names and years after punctuation

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

    The neuron selectively activates on the numeric and punctuation patterns of legal or bibliographic citations (case names, reporter volumes, page numbers, dates in parentheses).

    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
     aprecia
    -1.44
     mogelijkheid
    -1.42
    ximadamente
    -1.34
     tärke
    -1.31
     Redacción
    -1.30
    После
    -1.27
     retira
    -1.27
     decía
    -1.26
    蚨
    -1.23
    ޏ
    -1.23
    POSITIVE LOGITS
     on
    1.43
    相应的
    1.35
    cterms
    1.23
    щита
    1.20
     adesso
    1.19
     sidor
    1.19
    婭
    1.15
     конечно
    1.13
     from
    1.13
     fria
    1.09
    Activations Density 0.021%

    No Known Activations