© 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. 18804
    Prev
    Next
    INDEX
    Explanations

    turning, looking behind, around

    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
    亲æīĭ
    -0.10
    ç¬ı
    -0.09
    åľ°ä¸Ń
    -0.09
    æģ¹
    -0.08
    çľ¼ä¸ĭ
    -0.08
    æ´»çĿĢ
    -0.08
    æľĢ好ä¸įè¦ģ
    -0.08
    æ§Ľ
    -0.08
    Characters
    -0.08
    æīĭæĮģ
    -0.08
    POSITIVE LOGITS
    åĽŀ头
    0.11
    ego
    0.10
    Ñĥй
    0.09
    azzi
    0.09
    ç¬ij
    0.09
    åĽŀ头çľĭ
    0.09
    éĿ
    0.09
     ÑģÑĤÑĢа
    0.09
     Conway
    0.09
    quis
    0.09
    Activations Density 0.017%

    No Known Activations