© 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. 5447
    Prev
    Next
    INDEX
    Explanations

    efficiency and dexterity improvement

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

    The neuron detects the main product or specification keywords—i.e. prominent nouns and measurement terms—in item descriptions.

    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
     impedir
    0.77
     يجب
    0.75
     문제를
    0.73
     얘기를
    0.72
    🤬
    0.71
     に対
    0.70
     언급
    0.70
    (!(
    0.70
     precludes
    0.70
     산업
    0.68
    POSITIVE LOGITS
     sleek
    1.02
     offrant
    0.92
     elegant
    0.89
     sesuai
    0.89
     wellness
    0.89
     elegante
    0.89
     delightfully
    0.88
     tonos
    0.87
     bangle
    0.87
     delightful
    0.86
    Activations Density 0.697%

    No Known Activations