© 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
    2. Gemma-3-12B
    3. 24-GEMMASCOPE-2-RES-16K
    4. 11717
    Prev
    Next
    INDEX
    Explanations

    This neuron detects Japanese-style document references and numbering (e.g. “第 1”, “図 4”, full-width punctuation, example/step markers).

    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
    )،
    1.48
     Determining
    1.38
     Applying
    1.34
     Demonstrated
    1.34
    Applying
    1.32
    ؛
    1.31
     Analyzing
    1.30
    Improvements
    1.30
     Considerations
    1.29
     تكعيب
    1.29
    POSITIVE LOGITS
     그
    1.38
    <0xE3>
    1.29
     이
    1.28
    그
    1.20
     ※
    1.19
     
    1.18
    (
    1.15
     고
    1.13
     동
    1.12
    (
    1.11
    Activations Density 0.007%

    No Known Activations