© 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. 56337
    Prev
    Next
    INDEX
    Explanations

    U.S

    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
    æĪijåĽ½
    -0.09
    èĢģ人家
    -0.09
    åIJĦèĩªçļĦ
    -0.09
    äºĨæĪijçļĦ
    -0.09
    æĪijåĸľæ¬¢
    -0.09
    çļĦåIJĮåѦ
    -0.08
    åѤåįķ
    -0.08
    楼æĪ¿
    -0.08
    æ½®
    -0.08
    inden
    -0.08
    POSITIVE LOGITS
    anno
    0.09
    æĢ»åħ±
    0.09
    inant
    0.09
    ç»ĩ
    0.08
    åħ±æľī
    0.08
     nons
    0.08
    fone
    0.08
    -wide
    0.08
    顶级
    0.08
    Ñĥни
    0.08
    Activations Density 0.068%

    No Known Activations