© 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. 22026
    Prev
    Next
    INDEX
    Explanations

    lost and

    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
    auc
    -0.10
    Ī
    -0.09
    orth
    -0.09
     влаÑģÑĤи
    -0.08
    etre
    -0.08
    hort
    -0.08
    storm
    -0.08
    aire
    -0.08
    ç«ĭãģ¦
    -0.08
    ets
    -0.08
    POSITIVE LOGITS
    è¯Ŀ说
    0.12
    ness
    0.11
     Angeles
    0.11
    anza
    0.11
    è¿Ļ对
    0.10
    ãĤĪãģĨãģª
    0.10
    å±ĢéĿ¢
    0.10
    è¿Ļæł·çļĦè¯Ŀ
    0.10
    åIJĦè¡ĮåIJĦä¸ļ
    0.10
    abar
    0.09
    Activations Density 0.015%

    No Known Activations