© 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. 2262
    Prev
    Next
    INDEX
    Explanations

    quantitative comparison or change

    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
    åIJĪ计
    -0.09
    ouser
    -0.09
    æŃ»
    -0.09
    zee
    -0.09
    atches
    -0.09
    æĪIJç«ĭäºĨ
    -0.08
    ailable
    -0.08
    éķ¿éķ¿çļĦ
    -0.08
    eway
    -0.08
    æѤåĪ»
    -0.08
    POSITIVE LOGITS
    åħ¶ä¸Ńæľī
    0.11
    æľīæīĢ
    0.09
    åħ¶ä¸ŃåĮħæĭ¬
    0.09
    ç»´æĮģ
    0.08
     neutral
    0.08
    éĤĽ
    0.08
     penetration
    0.08
    ä¸ģ
    0.08
    pen
    0.08
    ouched
    0.08
    Activations Density 0.051%

    No Known Activations