© 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. 2885
    Prev
    Next
    INDEX
    Explanations

    predicting or analyzing

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

    This neuron fires on very high‐frequency function words (common articles, prepositions, auxiliaries, etc.).

    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
     เดือน
    0.88
     เรื่อง
    0.81
     mencari
    0.78
    }>{
    0.76
    рованные
    0.76
     propias
    0.76
     quảng
    0.74
     chữa
    0.74
    ówki
    0.73
     Nghị
    0.73
    POSITIVE LOGITS
    INST
    0.74
     уче
    0.74
    вате
    0.73
    colors
    0.72
    Основ
    0.72
    幹
    0.70
    сите
    0.69
    tensor
    0.67
     тео
    0.67
    銑
    0.67
    Activations Density 0.000%

    No Known Activations