© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 36374
    Prev
    Next
    INDEX
    Explanations

    actions and completions

    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
    çī¹æ®ĬæĥħåĨµ
    -0.11
    å¾Ĺå¾Ī
    -0.10
    çī¹ãģ«
    -0.10
    æ§Ľ
    -0.09
    caffold
    -0.09
    快讯
    -0.09
    éļĶçĥŃ
    -0.09
    诫
    -0.09
     realmente
    -0.09
     дейÑģÑĤвиÑĤелÑĮно
    -0.09
    POSITIVE LOGITS
    åIJĦ项
    0.12
    åIJĦ个æĸ¹éĿ¢
    0.11
    åIJĦç±»
    0.11
    çļĦåIJĦ项
    0.10
    åIJĦé¡¹å·¥ä½ľ
    0.09
    åIJĦæĸ¹éĿ¢
    0.09
    äºĨä¸Ģéģį
    0.09
    ast
    0.09
    æ¯ıä¸Ģ天
    0.09
    åIJĦä½į
    0.09
    Activations Density 0.263%

    No Known Activations