© 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. 33027
    Prev
    Next
    INDEX
    Explanations

    life without

    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
     Onion
    -0.09
    宽æķŀ
    -0.09
    误ä¼ļ
    -0.09
    ething
    -0.09
     táºŃn
    -0.09
    ifton
    -0.08
    èļĤ
    -0.08
    rott
    -0.08
    å¿ĥå¾Ĺä½ĵä¼ļ
    -0.08
    essional
    -0.08
    POSITIVE LOGITS
    没äºĨ
    0.19
     without
    0.18
     sans
    0.18
     بدÙĪÙĨ
    0.18
    without
    0.18
     без
    0.17
    _without
    0.17
     WITHOUT
    0.17
    Without
    0.16
     Without
    0.16
    Activations Density 0.239%

    No Known Activations