© 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. 323
    Prev
    Next
    INDEX
    Explanations

    contextual conjunctions

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

    This neuron responds to Japanese (and similar ideographic) technical terms—especially Kanji words relating to semiconductor underfill resin and curing processes.

    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
     ameryka
    0.49
    وضح
    0.48
    ؛
    0.44
     ඔබේ
    0.44
    ޏ
    0.44
     függvény
    0.43
     التربيع
    0.43
    )،
    0.42
     Buchstaben
    0.42
     تكعيب
    0.41
    POSITIVE LOGITS
    ，
    0.54
    만
    0.51
    、
    0.50
     不
    0.47
    的
    0.47
    不
    0.47
    大
    0.47
     자
    0.46
     몇
    0.46
    就
    0.46
    Activations Density 0.045%

    No Known Activations