© 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-3-27B
    5. 31-GEMMASCOPE-2-RES-262K
    7. 21008
    Prev
    Next
    INDEX
    Explanations

    Korean '지' followed by '향'

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

    Korean greetings and names

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-27b-pt/resid_post/layer_31_width_262k_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
     besten
    0.44
    伯
    0.44
    fills
    0.42
    ்சி
    0.42
    лі
    0.41
     ഈ
    0.41
    ttle
    0.41
     betre
    0.41
    Accepted
    0.41
     بها
    0.40
    POSITIVE LOGITS
     wine
    0.45
    ™,
    0.44
    ¹.
    0.44
    اشی
    0.43
     डिन
    0.42
     vect
    0.42
     interactions
    0.41
     virulent
    0.41
    ominator
    0.41
     narc
    0.41
    Activations Density 0.008%

    No Known Activations