© 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. 31-GEMMASCOPE-2-RES-16K
    7. 7644
    Prev
    Next
    INDEX
    Explanations

    everything

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

    The neuron strongly activates on plural common nouns (i.e., tokens ending in “s” that name items, services, classes, amenities, 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_31_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
     बजाय
    0.84
     which
    0.83
     (
    0.80
     especially
    0.80
     particularly
    0.80
     также
    0.76
     instead
    0.75
     (/
    0.74
    尤其是
    0.73
     نیز
    0.73
    POSITIVE LOGITS
     EVERYTHING
    1.61
     everything
    1.55
     wszystko
    1.54
    这一切
    1.54
     tudo
    1.52
     semuanya
    1.51
    everything
    1.48
     всё
    1.40
    Everything
    1.37
     Everything
    1.30
    Activations Density 0.307%

    No Known Activations