© 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. 31-GEMMASCOPE-2-RES-16K
    7. 10376
    Prev
    Next
    INDEX
    Explanations

    companies actively update or acknowledge

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

    The neuron detects common auxiliary and linking verbs—especially high‐frequency forms like “is,” “are,” and “will.”

    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_31_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
     died
    0.64
     deal
    0.60
     brink
    0.59
     penthouse
    0.59
     comer
    0.57
     south
    0.56
     lucky
    0.56
     ridge
    0.56
     nested
    0.55
     जीतकर
    0.55
    POSITIVE LOGITS
    禁止
    0.75
    公布
    0.72
    鼓励
    0.70
     memperkenalkan
    0.69
     홈페이지
    0.68
    发布
    0.64
     발행
    0.64
    umumkan
    0.64
     발표
    0.64
     запре
    0.63
    Activations Density 0.195%

    No Known Activations