© 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. 7050
    Prev
    Next
    INDEX
    Explanations

    the comma ","

    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
    åIJij社ä¼ļ
    -0.08
    åIJij
    -0.08
     Sind
    -0.08
    ickle
    -0.08
    HUD
    -0.08
    åľ¨ä»ĸçļĦ
    -0.08
    è·Ł
    -0.08
    åľ¨æĪijçļĦ
    -0.07
    åľŃ
    -0.07
    ä¸įè¿ĩ
    -0.07
    POSITIVE LOGITS
    æĪij们就
    0.14
    æĪijå°±
    0.12
    æĪijæĥ³
    0.12
    éĥ½ä¸įçŁ¥éģĵ
    0.12
    ä½łå°±
    0.11
    ä¸įå¾Ĺä¸į说
    0.11
     maka
    0.11
    æĪijåıĪ
    0.11
    就说
    0.11
     ÙģØ¥ÙĨ
    0.11
    Activations Density 0.259%

    No Known Activations