© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 9830
    Prev
    Next
    INDEX
    Explanations

    ethical and appropriateness concerns

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

    The neuron fires on high-content words—especially prominent action verbs or transaction terms (e.g. “purchase,” “buy,” “act”) that carry key informational load.

    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
     ด่า
    1.03
     itd
    0.99
    K
    0.95
     unsurprisingly
    0.91
     onboarding
    0.90
     يمكنك
    0.88
     しっかり
    0.88
     すご
    0.88
     żeby
    0.86
     توی
    0.86
    POSITIVE LOGITS
     perplexing
    0.94
     quale
    0.90
     neoplastic
    0.90
     progeny
    0.87
     pictorial
    0.87
     nahezu
    0.82
     nomenclature
    0.81
    萃
    0.80
     fanciful
    0.80
    करिता
    0.78
    Activations Density 0.586%

    No Known Activations