© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 1963
    Prev
    Next
    INDEX
    Explanations

    affirmative or interrogative endings

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

    The neuron strongly activates on common short conversational “filler” and address words (greetings, interjections, apologies, quick responses) in dialogue.

    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
    当該
    0.70
     ("
    0.64
    此外
    0.63
    <eos>
    0.62
     —
    0.61
    ↵↵↵
    0.59
    ↵↵↵↵
    0.59
     ==>
    0.59
    该公司
    0.59
     ⇒
    0.57
    POSITIVE LOGITS
     babe
    0.96
     hermano
    0.82
    !!”
    0.80
     okay
    0.77
     letzte
    0.77
     mamá
    0.77
     प्लीज
    0.77
     bebé
    0.74
    ?!"
    0.73
     انا
    0.73
    Activations Density 0.060%

    No Known Activations