© 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
    2. Gemma-3-12B
    3. 31-GEMMASCOPE-2-RES-16K
    4. 397
    Prev
    Next
    INDEX
    Explanations

    "i" followed by verbs

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

    The neuron fires on noun tokens that name categories or groups of things—especially plural (or collective) nouns.

    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
     $\%$
    0.41
     chainsaw
    0.41
     immunization
    0.39
     œ
    0.39
     Fiverr
    0.39
     densely
    0.38
     cybersecurity
    0.38
    轧
    0.38
     GIL
    0.38
     ${
    0.38
    POSITIVE LOGITS
     Кла
    0.41
     అభి
    0.38
    <unused544>
    0.38
    érie
    0.38
     Шо
    0.37
    <unused479>
    0.37
     板
    0.37
     Suggest
    0.37
     Жи
    0.37
     lendemain
    0.36
    Activations Density 0.004%

    No Known Activations