© 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. 12819
    Prev
    Next
    INDEX
    Explanations

    programming languages and social platforms

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

    The neuron strongly activates on standalone section‐heading tokens—like days of the week, unit numbers or titles, sports positions, language names, chakra names, etc.—that label or introduce a new section.

    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
     titers
    0.52
     the
    0.51
    ੱਕ
    0.50
    র্ঘ
    0.47
    ってお
    0.46
    ох
    0.46
    つまり
    0.45
    რ
    0.45
    也就是说
    0.45
    用户的
    0.45
    POSITIVE LOGITS
     ผม
    0.55
     ¿?
    0.55
    /#
    0.54
     ตุลาคม
    0.52
     Terbaik
    0.52
     單
    0.52
     Faites
    0.51
     किमान
    0.51
    ፡
    0.51
    \":
    0.50
    Activations Density 0.020%

    No Known Activations