© 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. 7180
    Prev
    Next
    INDEX
    Explanations

    trained in

    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
    éļıçĿĢæŶéĹ´
    -0.10
     cán
    -0.09
    俶
    -0.09
    æĥĬåı¹
    -0.08
    羸
    -0.08
    nam
    -0.08
    çĸ½
    -0.08
    æ¯Ķè¾ĥ大
    -0.08
    inals
    -0.08
    sole
    -0.08
    POSITIVE LOGITS
    ä¸ĢéŨ
    0.12
    æĸ¹éĿ¢çļĦ
    0.09
    åĬŁè¯¾
    0.09
    è¿Ļç§įæĸ¹å¼ı
    0.09
     Dillon
    0.09
     unusual
    0.09
    LTR
    0.08
    åĿ¨
    0.08
    缸åħ³æĶ¿çŃĸ
    0.08
    bred
    0.08
    Activations Density 0.029%

    No Known Activations