© 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
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 110609
    Prev
    Next
    INDEX
    Explanations

    one of

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

    This neuron detects the cardinal “one” when it’s used to refer to acquiring or identifying a single item (e.g. in phrases like “get one,” “pick one up,” “one of the first”).

    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
    athed
    -1.41
     combinação
    -1.39
     كورونا
    -1.32
    Gosh
    -1.30
    Muerte
    -1.29
    icône
    -1.29
     जाना
    -1.28
    泐
    -1.23
     rajut
    -1.23
     manquer
    -1.22
    POSITIVE LOGITS
     for
    2.05
     but
    1.80
     can
    1.63
     when
    1.50
     with
    1.48
     of
    1.48
     this
    1.48
     without
    1.45
     that
    1.43
     before
    1.35
    Activations Density 0.077%

    No Known Activations