© 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. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 22948
    Prev
    Next
    INDEX
    Explanations

    native country or region

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    adamkarvonen/qwen3-32b-saes/saes_Qwen_Qwen3-32B_batch_top_k/resid_post_layer_32
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    _DS
    -0.10
    çļĤ
    -0.09
    åIJĽåŃIJ
    -0.09
    æ£Ģ
    -0.09
    ä¸ŃèĢĥ
    -0.08
    Convention
    -0.08
    kah
    -0.08
    adt
    -0.08
    ä¸įçIJĨ
    -0.08
    ç³ł
    -0.08
    POSITIVE LOGITS
    æķħ乡
    0.17
    æ¯įæł¡
    0.17
    家乡
    0.17
    ç¥ĸåĽ½
    0.15
    ä¹ĭ乡
    0.15
    èĢģ家
    0.13
     alma
    0.12
    ä¹ĭåľ°
    0.12
     homeland
    0.12
    æīĢåľ¨
    0.12
    Activations Density 0.221%

    No Known Activations