© 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. 56076
    Prev
    Next
    INDEX
    Explanations

    scientific specialization

    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
    æĥħ人
    -0.09
    åħ¨ä¸ĸçķĮ
    -0.09
    ıy
    -0.08
    razy
    -0.08
    urer
    -0.08
    _OCC
    -0.08
    ä¸Ģçķª
    -0.08
    æİ·
    -0.08
    _OCCURRED
    -0.08
    lass
    -0.08
    POSITIVE LOGITS
    æīĭä¸ŃçļĦ
    0.09
     воÑģ
    0.08
    åĶ¯ä¸ĢçļĦ
    0.08
    ix
    0.08
    å½¢æĢģ
    0.08
    eld
    0.08
    å¾¼
    0.08
    æľīæľº
    0.08
     Santiago
    0.08
    zers
    0.08
    Activations Density 0.657%

    No Known Activations