© 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. 34-GEMMASCOPE-RES-131K
    7. 55716
    Prev
    Next
    INDEX
    Explanations

    further reading or note

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

    The neuron selectively activates on numeric digit sequences (i.e. tokens composed of numbers).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
    電話番号
    -1.12
     sebagainya
    -1.06
     fabrica
    -1.04
    ﹖
    -1.04
    getMenu
    -1.02
     expérience
    -0.99
    Peru
    -0.99
    ԅ
    -0.99
     Basta
    -0.99
     preço
    -0.97
    POSITIVE LOGITS
     afield
    1.72
    emore
    1.63
    most
    1.55
    ed
    1.46
     than
    1.45
    more
    1.43
    nore
    1.43
    ance
    1.32
    MORE
    1.23
     details
    1.22
    Activations Density 0.024%

    No Known Activations