© 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. 56257
    Prev
    Next
    INDEX
    Explanations

    asking you questions

    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
    åħ¥åºĵ
    -0.09
    åIJĽ
    -0.09
     Spo
    -0.09
    eyer
    -0.08
    åij¼åIJ¸éģĵ
    -0.08
    eer
    -0.08
    ä¸ĭäºĨ
    -0.08
    OP
    -0.07
    è½»è½»åľ°
    -0.07
    åij¼åĶ¤
    -0.07
    POSITIVE LOGITS
    帮æĪij
    0.13
    ç»ĻäºĨæĪij
    0.10
    -cookie
    0.10
    åīįè¿°
    0.10
    幫
    0.10
    ãģĦãĤī
    0.10
    èģĮæĿĥ
    0.10
    åķ¬
    0.09
    _ALLOW
    0.09
    batim
    0.09
    Activations Density 0.187%

    No Known Activations