© 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. 12750
    Prev
    Next
    INDEX
    Explanations

    Germany, individual accountability, organizations, Stories

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

    The neuron is highly activated by numeric tokens (especially multi‐digit numbers or decimals), i.e. it finds tokens that represent numerical values.

    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
    ς
    0.91
    ⋮
    0.91
    acide
    0.87
    oser
    0.80
    ات
    0.80
    kw
    0.79
    Clause
    0.79
    icie
    0.79
    rawberry
    0.79
    ﾝ
    0.79
    POSITIVE LOGITS
    並沒有
    0.77
     updatedAt
    0.74
    並且
    0.65
     Congrès
    0.65
     genealogy
    0.64
     يقول
    0.64
     वजह
    0.63
     ব্যবস্থাপ
    0.63
    أي
    0.63
    或許
    0.62
    Activations Density 0.000%

    No Known Activations