© 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
    2. Gemma-3-12B
    3. 24-GEMMASCOPE-2-RES-16K
    4. 13758
    Prev
    Next
    INDEX
    Explanations

    while also

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

    The neuron is highly activated by English discourse-connective words—especially temporal subordinators like “once,” “on,” and “while”—that introduce clauses.

    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
    ни
    2.06
    м
    1.45
    ע
    1.35
    ן
    1.23
    ро
    1.20
    ernetes
    1.20
    ла
    1.08
    ol
    1.06
    сию
    1.06
    rika
    1.05
    POSITIVE LOGITS
    ف
    1.72
    🠀
    1.18
    ্ড
    1.16
     simultaneously
    1.16
     simult
    1.14
    تون
    1.14
    deki
    1.13
     दुसरी
    1.13
     concurrently
    1.12
    llä
    1.12
    Activations Density 0.151%

    No Known Activations