© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 4022
    Prev
    Next
    INDEX
    Explanations

    The neuron specializes in spotting numerical tokens and numeric sequences (e.g. dates, phone numbers, prices, percentages) in the text.

    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_24_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
     жүктөө
    0.46
    angleright
    0.45
    kün
    0.45
    伀
    0.43
     Казино
    0.43
    би
    0.42
    >=</
    0.42
    नो
    0.42
    ὗ
    0.41
    ન્
    0.41
    POSITIVE LOGITS
    that
    0.61
     that
    0.52
     Because
    0.49
    That
    0.49
     they
    0.48
     They
    0.48
    Because
    0.48
    They
    0.48
    Which
    0.47
    which
    0.46
    Activations Density 0.262%

    No Known Activations