© 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. 11777
    Prev
    Next
    INDEX
    Explanations

    first book or series

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

    The neuron detects ordinal or numeric mentions that specify an item’s position or count within a series (e.g., “first,” “two,” “five,” “second book,” “third in the series”).

    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
     first
    -1.75
     because
    -1.59
     ....
    -1.58
     only
    -1.56
     both
    -1.55
     .....
    -1.55
     bereits
    -1.49
     by
    -1.48
     expertly
    -1.47
     very
    -1.45
    POSITIVE LOGITS
     Schritte
    1.43
    confused
    1.40
     recentes
    1.34
    𐄁
    1.31
     lokalen
    1.31
     cerâmica
    1.31
    fibonacci
    1.29
     artesanato
    1.27
     Gespräche
    1.27
    как
    1.27
    Activations Density 0.041%

    No Known Activations