© 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. 16021
    Prev
    Next
    INDEX
    Explanations

    parenthesis, brackets, and dollar signs

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

    This neuron fires on content words that kick off new clauses or sentences (i.e. the first “real” word after a sentence or clause boundary).

    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
    僚
    0.69
     Однако
    0.68
     однако
    0.66
    事情
    0.65
    きた
    0.64
    きて
    0.64
    spending
    0.64
    ANIES
    0.64
     intento
    0.64
     ک
    0.63
    POSITIVE LOGITS
     Dora
    0.85
     folhas
    0.85
    orel
    0.84
    Y
    0.84
    ör
    0.81
    ડી
    0.81
     realidade
    0.79
    iz
    0.79
     mistura
    0.79
     léz
    0.78
    Activations Density 0.000%

    No Known Activations