© 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. 8382
    Prev
    Next
    INDEX
    Explanations

    dialogue markers and quotes

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

    This neuron fires strongly on tokens inside quoted dialogue, effectively spotting direct speech segments.

    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
     underserved
    0.60
     koriste
    0.59
    L
    0.59
     multifaceted
    0.57
     utilize
    0.57
     strives
    0.56
     n
    0.55
     elusive
    0.55
    7
    0.55
     J
    0.55
    POSITIVE LOGITS
    옇
    0.75
    ákat
    0.74
     চাহিয়া
    0.71
     మీకు
    0.71
     Rowing
    0.70
     ပြော
    0.70
    thorace
    0.69
    íte
    0.69
    Ꮁ
    0.69
     বুঝিতে
    0.66
    Activations Density 0.002%

    No Known Activations