© 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. 22-GEMMASCOPE-RES-131K
    7. 130743
    Prev
    Next
    INDEX
    Explanations

    The neuron strongly activates on standalone numeric tokens—especially multi‐digit numbers and percentages—i.e. statistical figures in the text.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    quantities and measurements

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
     where
    -1.70
     without
    -1.69
     with
    -1.68
     as
    -1.65
     almost
    -1.59
     these
    -1.58
     probably
    -1.58
     one
    -1.56
     that
    -1.55
     which
    -1.55
    POSITIVE LOGITS
     refroid
    1.88
     mélang
    1.73
     réjou
    1.66
     rassemb
    1.57
     relâche
    1.55
     rafraî
    1.51
     récol
    1.48
     nécess
    1.45
     récomp
    1.43
     trám
    1.41
    Activations Density 0.051%

    No Known Activations