© 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. 14034
    Prev
    Next
    INDEX
    Explanations

    verbs after punctuation

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

    This neuron activates on rare, content‐specific proper nouns and named entities (people, organizations, places, dates).

    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
     theres
    0.88
     the
    0.82
    ,【
    0.82
     there
    0.79
    ,
    0.77
     –,
    0.75
    ،
    0.73
     everyone
    0.72
     teh
    0.71
     indiv
    0.71
    POSITIVE LOGITS
     has
    0.96
     heeft
    0.86
    has
    0.82
    은
    0.78
    は
    0.76
     имеет
    0.76
     specializes
    0.76
     knows
    0.75
     deserves
    0.74
     ਨੇ
    0.74
    Activations Density 0.237%

    No Known Activations