© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 10601
    Prev
    Next
    INDEX
    Explanations

    digits followed by letters or symbols

    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
    uttle
    -0.11
    æĵ¬
    -0.10
    æĭŁ
    -0.10
    æ¦Ľ
    -0.09
    pies
    -0.09
    аÑģÑģ
    -0.09
    åī¯
    -0.09
    访
    -0.09
    á»ĩu
    -0.09
    igli
    -0.09
    POSITIVE LOGITS
    é¡·
    0.08
     compliant
    0.08
    è¹Ĭ
    0.07
    æĻ®
    0.07
    çī¹åĪ«æĺ¯åľ¨
    0.07
    оÑĢÑĤ
    0.07
    æݥ触åĪ°
    0.07
    çī¹åĪ«æĺ¯
    0.07
     pro
    0.07
    éĩįè¦ģæĦıä¹ī
    0.07
    Activations Density 0.254%

    No Known Activations