© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 5429
    Prev
    Next
    INDEX
    Explanations

    * AS _ , at least, code

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

    The neuron fires on tokens used in formal factual or legal reporting—e.g. ownership declarations (“owns”) and legislative citations (“Leg”).

    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
     カジュアル
    0.87
     ג
    0.86
     ค
    0.84
     โร
    0.84
     Ү
    0.81
     В
    0.80
     จ
    0.79
     ฟ
    0.79
     సన్ని
    0.79
     ก
    0.79
    POSITIVE LOGITS
    MRI
    0.85
    SCO
    0.77
    l
    0.76
    TG
    0.75
    CAB
    0.75
    pou
    0.75
    CAM
    0.74
    exempl
    0.74
    LI
    0.70
    Oku
    0.70
    Activations Density 0.000%

    No Known Activations