© 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. 31-GEMMASCOPE-2-RES-16K
    7. 7951
    Prev
    Next
    INDEX
    Explanations

    give you the best

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

    This neuron detects direct address to the reader—especially second-person “you” (and related future-oriented phrasing like “you will”) in instructional or advisory text.

    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
    ために
    0.65
    it
    0.60
    ﾅ
    0.59
    ですし
    0.59
    なくな
    0.57
    Wear
    0.57
    ことで
    0.57
    دل
    0.56
    你说
    0.55
    Vs
    0.54
    POSITIVE LOGITS
     guys
    1.04
    uuuu
    0.88
     bè
    0.80
    guys
    0.73
    uu
    0.73
     morgen
    0.72
     closer
    0.70
     pouvez
    0.70
     beiden
    0.70
    шают
    0.70
    Activations Density 0.314%

    No Known Activations