© 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. 12166
    Prev
    Next
    INDEX
    Explanations

    on (1) or off (0)

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

    This neuron fires strongly on the very first word or symbol at the start of a new text segment or paragraph, effectively marking segment boundaries.

    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
    Jeśli
    0.90
    Lesson
    0.78
     harshly
    0.78
    as
    0.77
     woe
    0.75
    রামর্শ
    0.74
    施
    0.74
    Df
    0.71
    Ironically
    0.71
    فة
    0.71
    POSITIVE LOGITS
     هدف
    0.77
     деятельность
    0.77
     يست
    0.76
    니다
    0.73
    tedir
    0.73
     reptile
    0.73
     pelanggan
    0.72
     материал
    0.72
     pekerjaan
    0.71
    猊
    0.71
    Activations Density 0.000%

    No Known Activations