© 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. 5895
    Prev
    Next
    INDEX
    Explanations

    or controlling or improving

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

    This neuron detects the logical-connector and modal words (e.g. “or,” “are,” “may,” etc.) typical of enumerated legal or policy clauses.

    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
     karate
    0.65
     watts
    0.63
    watts
    0.62
    酊
    0.62
     बैंग
    0.61
     Pedal
    0.59
     тебя
    0.59
    ตัวเอง
    0.59
    𐰴
    0.59
     melancholy
    0.58
    POSITIVE LOGITS
     proxies
    0.71
     zumindest
    0.61
    RUPTION
    0.60
     Prox
    0.57
    ادرة
    0.55
     closely
    0.54
     असलेले
    0.54
     являются
    0.54
     proxy
    0.53
    STARTED
    0.53
    Activations Density 0.124%

    No Known Activations