© 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. 15673
    Prev
    Next
    INDEX
    Explanations

    Da Vinci, Da Funk, Daikon

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

    The neuron fires strongly whenever it sees the token “Da” (as in names or brand prefixes).

    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
     lectores
    0.63
    verständlich
    0.59
     physi
    0.57
     rétr
    0.56
     atteints
    0.56
    ॰ऍ
    0.56
    
    0.55
     clazz
    0.54
    TimeDict
    0.54
    颔
    0.54
    POSITIVE LOGITS
    ğın
    0.72
    ğ
    0.69
    UGHT
    0.56
    пот
    0.55
     Vinci
    0.55
    섯
    0.55
    ków
    0.54
    ำ
    0.53
     capo
    0.52
    ース
    0.52
    Activations Density 0.160%

    No Known Activations