© 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. 30692
    Prev
    Next
    INDEX
    Explanations

    place names or entities

    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
    åįģéĩĮ
    -0.09
    太太
    -0.09
    ocre
    -0.08
    âĸ·
    -0.08
    åģ¶
    -0.08
    æĭħä¿Ŀ
    -0.08
    ç¼ĵæħ¢
    -0.08
    ried
    -0.08
    çĮ¥
    -0.08
    -corner
    -0.08
    POSITIVE LOGITS
    寻
    0.09
    å¹´å¤ľ
    0.09
    å·¥ä½ľä½ľé£İ
    0.09
    è¿Ļä¹Ī大
    0.09
    éĽ³
    0.08
    鸳
    0.08
    ä¸Ģå¹´ä¸Ģ度
    0.08
    mun
    0.08
    åij³
    0.08
    为代表
    0.08
    Activations Density 0.020%

    No Known Activations