© 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. 21638
    Prev
    Next
    INDEX
    Explanations

    others

    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
    æ¾Į
    -0.11
    ä»·åĢ¼éĵ¾
    -0.09
    Streaming
    -0.09
    AGER
    -0.09
    tÃŃ
    -0.09
    请éĢīæĭ©
    -0.09
    orges
    -0.09
    Corp
    -0.09
    è¾½
    -0.09
    aky
    -0.08
    POSITIVE LOGITS
    ä»ĺ
    0.11
    æĭ´
    0.09
    POP
    0.08
     Bek
    0.08
    rect
    0.08
    ÙĬرة
    0.08
    ä¸ĭæ»ij
    0.08
     доÑģÑĤ
    0.07
    åŀ«
    0.07
     autoload
    0.07
    Activations Density 0.003%

    No Known Activations