© 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. 28044
    Prev
    Next
    INDEX
    Explanations

    manage tasks and resources

    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
    ä¸Ģ个好çļĦ
    -0.09
    илÑĮ
    -0.09
    ÙģÙĦ
    -0.09
    Mgr
    -0.08
    管çIJĨå±Ĥ
    -0.08
    é«Ģ
    -0.08
    管çIJĨåĪ¶åº¦
    -0.08
    派人
    -0.08
    å¾ij
    -0.08
    CEL
    -0.08
    POSITIVE LOGITS
    èµ·æĿ¥äºĨ
    0.11
    able
    0.11
    çĿ¬
    0.10
    æѤäºĭ
    0.10
    uten
    0.09
    æĹºçĽĽ
    0.09
    äºĨä¸Ģåľº
    0.09
    èµ·æĿ¥
    0.09
    itarian
    0.09
     ÑģиÑĤÑĥаÑĨии
    0.09
    Activations Density 0.050%

    No Known Activations