© 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. 4373
    Prev
    Next
    INDEX
    Explanations

    civil tax, objection, peace activist, Hindu Karma Yoga

    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«®
    -0.09
    æıIJæĹ©
    -0.09
     hor
    -0.09
    ear
    -0.09
    岫
    -0.08
    æIJŀ好
    -0.08
    èĢĹ
    -0.08
    AREA
    -0.08
    arty
    -0.08
    å§Ĵ
    -0.08
    POSITIVE LOGITS
    å¾®éĩı
    0.10
     unarmed
    0.10
    zell
    0.09
    æĴĴ
    0.09
     Surre
    0.09
    å½¹
    0.09
    æļ´åĬĽ
    0.09
    коÑģ
    0.09
    å´ĩå°ļ
    0.09
     militar
    0.09
    Activations Density 0.071%

    No Known Activations