© 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. 57015
    Prev
    Next
    INDEX
    Explanations

    rolled out

    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
    .ssl
    -0.09
    åĸĿ
    -0.09
    èĿī
    -0.08
    ä½łä¸į
    -0.08
    æĵĺ
    -0.08
     اÙĦÙħسÙĦÙħ
    -0.08
    è´²
    -0.08
    iri
    -0.08
    åĸĿäºĨ
    -0.08
    лиз
    -0.08
    POSITIVE LOGITS
    æݨ广
    0.21
    æݨ广åºĶçĶ¨
    0.17
    æݨè¡Į
    0.15
    è¯ķçĤ¹
    0.15
    è¯ķè¡Į
    0.14
    æĻ®åıĬ
    0.13
    è¯ķçĤ¹å·¥ä½ľ
    0.13
     ÑĢаÑģпÑĢоÑģÑĤÑĢан
    0.13
    大éĿ¢ç§¯
    0.12
    æݨåĩº
    0.12
    Activations Density 0.093%

    No Known Activations