© 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. 36572
    Prev
    Next
    INDEX
    Explanations

    review quality scores

    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
    è·¨è¶Ĭå¼ı
    -0.09
    åį´æĺ¯
    -0.09
    è²Įä¼¼
    -0.08
    éĽģ
    -0.08
    éĻ¤æѤ
    -0.08
    .opts
    -0.08
    çIJĩ
    -0.08
    é¢Ĩè¡Ķ
    -0.08
    ä¸į解
    -0.08
    mt
    -0.08
    POSITIVE LOGITS
    éļ¾å¿ĺ
    0.09
    erne
    0.09
    æĮĩæķ°
    0.08
     лиÑĩно
    0.08
    èĨ
    0.08
    Hack
    0.08
     thesis
    0.08
    çļĦèī¯å¥½
    0.08
     typical
    0.08
    åĨ¤
    0.08
    Activations Density 0.052%

    No Known Activations