© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 1454
    Prev
    Next
    INDEX
    Explanations

    The neuron fires on any numeric token or sequence containing digits (e.g. phone numbers, prices, dates, IDs).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-12b-pt/resid_post/layer_31_width_16k_l0_medium
    Prompts (Dashboard)
    392,802 prompts, 256 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    冪
    0.58
     ONLINE
    0.57
     Jut
    0.55
     PRICE
    0.52
    zarchiwizowane
    0.52
    ="$
    0.52
    賏
    0.52
    頂き
    0.51
    invokeLater
    0.50
    MakeRange
    0.50
    POSITIVE LOGITS
     phone
    1.27
     telephone
    1.25
    phone
    1.24
     tel
    1.23
    Tel
    1.20
     Tel
    1.16
     phones
    1.16
    telephone
    1.16
     telephones
    1.13
    Telephone
    1.13
    Activations Density 0.149%

    No Known Activations