© 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. 60287
    Prev
    Next
    INDEX
    Explanations

    citations and references

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

    The neuron locks onto punctuation and connector words that occur around citations and section‐break lists in academic text—e.g. closing brackets, commas, and contrastive “but.”

    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
     help
    -1.28
     Rücksicht
    -1.23
     zatem
    -1.20
    潆
    -1.19
     neues
    -1.19
     our
    -1.18
     problemów
    -1.18
     dépasser
    -1.17
     kinds
    -1.16
    . 
    -1.16
    POSITIVE LOGITS
     from
    1.97
     and
    1.59
    (!__
    1.26
     iNdEx
    1.24
     Podcasts
    1.19
    尽管
    1.13
    </strong>
    1.12
     begon
    1.09
     طويلة
    1.09
    —
    1.09
    Activations Density 0.056%

    No Known Activations