© 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. 24-GEMMASCOPE-2-RES-16K
    7. 15028
    Prev
    Next
    INDEX
    Explanations

    actions and processes

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

    The main thing this neuron does is detect numerical tokens—years, dates, quantities, measurements, counts, and similar number‐related information.

    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
    MiddleLeft
    0.56
    CharField
    0.55
    inese
    0.55
    imap
    0.55
    하거나
    0.54
     agrícola
    0.54
    aný
    0.54
    🙍
    0.53
    ubei
    0.52
    ᠮ
    0.52
    POSITIVE LOGITS
     FINAL
    0.90
     ulteriori
    0.87
    สุดท้าย
    0.80
     Final
    0.80
    FINAL
    0.80
     अंतिम
    0.79
    最后的
    0.78
    最後の
    0.77
     கடைசி
    0.75
     finales
    0.75
    Activations Density 1.724%

    No Known Activations