© 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. 71233
    Prev
    Next
    INDEX
    Explanations

    function, purpose, or characteristic

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

    The neuron fires on formal, specification-style words and phrases that typically introduce or qualify quantitative, measured, or time-stamped information (e.g. “shall,” “mainly,” “dominated,” “prerequisite,” “AM/PM”).

    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
    鵟
    -1.65
     anhänger
    -1.55
     uczniów
    -1.55
    糁
    -1.55
    acterísticas
    -1.55
    ໊
    -1.48
    熥
    -1.45
    鱨
    -1.44
    ついに
    -1.43
     verdienen
    -1.43
    POSITIVE LOGITS
     our
    1.64
     get
    1.54
     they
    1.53
     Do
    1.45
     do
    1.41
     We
    1.38
     those
    1.36
     unsere
    1.34
     events
    1.34
     use
    1.34
    Activations Density 0.031%

    No Known Activations