© 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. 39605
    Prev
    Next
    INDEX
    Explanations

    add to

    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
    è§ģè¿ĩ
    -0.10
     Consort
    -0.09
     Osborne
    -0.09
    è¸ı
    -0.09
    éĺ¶æ¢¯
    -0.09
    空åīį
    -0.09
    å¤ļä½ĻçļĦ
    -0.08
    ìĹĺ
    -0.08
    ãĤĬ
    -0.08
    =format
    -0.08
    POSITIVE LOGITS
    å®ĮåĸĦ
    0.12
    ä¸įåĥı
    0.09
    æī¹
    0.08
    积累
    0.08
    ogg
    0.08
    涵
    0.08
    ä¸įæĸŃå®ĮåĸĦ
    0.08
     established
    0.08
    ç¼ł
    0.08
    稿
    0.08
    Activations Density 0.035%

    No Known Activations