© 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. 4922
    Prev
    Next
    INDEX
    Explanations

    data-driven, secure, intuitive

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

    This neuron detects hyphenated or compound attribute adjectives—like “user-centric,” “data-driven,” “high-performance,” “non-invasive,” etc.

    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
     bzw
    0.56
     cosa
    0.52
    uję
    0.52
    (
    0.51
     तुम्हारे
    0.51
    :
    0.50
     Continued
    0.49
     Ibid
    0.48
     auc
    0.48
    W
    0.47
    POSITIVE LOGITS
     প্রয়োগ
    0.55
    ρας
    0.55
    styling
    0.55
     للغاية
    0.54
    ंग्रेस
    0.54
     optimizations
    0.53
     filmmaking
    0.52
    解决方案
    0.51
    फारिश
    0.50
    <unused184>
    0.50
    Activations Density 0.001%

    No Known Activations