© 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. 15853
    Prev
    Next
    INDEX
    Explanations

    when you

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

    The neuron detects question-opening function words and bigrams (e.g. “How do…,” “Does…have…”) that mark the start of a user question.

    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
     등의
    0.72
     (
    0.68
     हेतु
    0.66
    』（
    0.65
    들의
    0.64
    0.62
    合せ
    0.61
     （
    0.61
     evoking
    0.60
    （
    0.60
    POSITIVE LOGITS
     bassa
    0.68
     tubo
    0.66
     Quindi
    0.66
     ആണ്
    0.65
    十几
    0.63
    🫢
    0.62
     انا
    0.62
     rekao
    0.62
     morreu
    0.60
    চু
    0.60
    Activations Density 0.007%

    No Known Activations