© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 8549
    Prev
    Next
    INDEX
    Explanations

    X, Y variables

    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
    绿èī²çݯä¿Ŀ
    -0.11
    RestController
    -0.10
    å·§
    -0.09
    åŀĨ
    -0.08
    paralle
    -0.08
    å¢ŀåĢ¼
    -0.08
    åħ¬æĬ¥
    -0.08
    .drawString
    -0.08
     Garner
    -0.08
    踪
    -0.08
    POSITIVE LOGITS
    ç¾ŀ
    0.09
    åľ¨ä»ĸ
    0.08
    å°±ä¸įèĥ½
    0.08
    åľ¨ä»ĸçļĦ
    0.08
    æĬĬä»ĸ
    0.08
    gay
    0.08
     idiot
    0.07
    毫æĹłçĸij
    0.07
    QR
    0.07
     Fat
    0.07
    Activations Density 0.257%

    No Known Activations