© 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-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 6088
    Prev
    Next
    INDEX
    Explanations

    This neuron detects numeric expressions and arithmetic operators (e.g. numbers and “+” signs) in formula-like contexts.

    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
    רו
    0.96
    בת
    0.90
    은
    0.89
    ק
    0.89
    טה
    0.88
    ਅ
    0.87
     sats
    0.87
     và
    0.86
    +)$
    0.84
    서는
    0.84
    POSITIVE LOGITS
    د
    1.00
    /−
    0.98
     wymien
    0.95
    بدا
    0.94
    intention
    0.93
    အတူ
    0.93
    entreprise
    0.92
     مشتمل
    0.92
    dır
    0.91
    apeutic
    0.91
    Activations Density 0.018%

    No Known Activations