© 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. 12239
    Prev
    Next
    INDEX
    Explanations

    This is

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

    The neuron fires on comment‐ or markup‐opening tokens (e.g. “/*” or “<em>”), i.e. the starts of code comments or inline markup.

    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
    с
    1.20
    ्
    1.18
    ре
    1.12
    لية
    1.10
    گ
    1.09
    ү
    1.05
     inviting
    1.04
    з
    1.04
    ے
    1.04
    ер
    1.03
    POSITIVE LOGITS
    
    2.09
    ból
    1.43
    funcs
    1.32
     Drž
    1.31
    bangan
    1.29
    йки
    1.26
    
    1.20
    тности
    1.19
    ‌اند
    1.16
    ‌تر
    1.13
    Activations Density 0.000%

    No Known Activations