© 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. 49713
    Prev
    Next
    INDEX
    Explanations

    models and codes

    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
    åīĶ
    -0.09
    доб
    -0.09
     Saud
    -0.09
    áŀĹ
    -0.08
    _FMT
    -0.08
     SWT
    -0.08
    enze
    -0.08
    çģ¿
    -0.08
    亿ä¸ĩ
    -0.08
    jets
    -0.08
    POSITIVE LOGITS
    åĽŀåIJĪ
    0.11
    ä¸Ģ款
    0.11
    ç¢İ
    0.10
    åģľä¸ĭæĿ¥
    0.10
    模åŀĭ
    0.10
     ëıĮ
    0.10
    æ¼Ķ
    0.09
    íķ´ì£¼
    0.09
     Graz
    0.09
     ká»ĭch
    0.09
    Activations Density 0.142%

    No Known Activations