© 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. 7027
    Prev
    Next
    INDEX
    Explanations

    include or included

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

    The main thing this neuron does is pick out the unusually informative or high-weight tokens in a passage—words (including numbers, common function words, and key content terms) that carry the most semantic or structural importance.

    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
     privately
    0.66
     renfer
    0.60
     starfish
    0.60
     उससे
    0.59
     loved
    0.58
     ஏற்பட்டு
    0.55
     einzel
    0.55
     D
    0.55
     effic
    0.55
    Cour
    0.55
    POSITIVE LOGITS
     Included
    1.12
    included
    1.09
     शामिल
    0.99
    Included
    0.99
     součástí
    0.99
     inclusions
    0.94
    inclusion
    0.93
     included
    0.93
    CLUDED
    0.91
     inclusion
    0.90
    Activations Density 1.234%

    No Known Activations