© 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. 5958
    Prev
    Next
    INDEX
    Explanations

    drip of, stuffy, time during, musculoskeletal issues

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

    The neuron activates on content nouns (both concrete things like “trees” and abstract concepts like “peace,” “justice,” or “trading”).

    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
    ?
    0.64
    You
    0.62
     youll
    0.61
    您的
    0.60
     you
    0.59
    您
    0.58
     your
    0.55
     expressly
    0.55
    說明
    0.55
    Use
    0.55
    POSITIVE LOGITS
     situación
    0.87
     少し
    0.86
     mediocre
    0.85
     commotion
    0.82
     conseguiu
    0.81
     helplessly
    0.81
    😔
    0.81
     sporad
    0.80
     väldigt
    0.80
    😥
    0.80
    Activations Density 0.065%

    No Known Activations