© 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. 27008
    Prev
    Next
    INDEX
    Explanations

    deficiency or defect

    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
    çľĭå¾ħ
    -0.12
    æ»ĭåij³
    -0.10
    å§Ĭ
    -0.10
    äºĨä¸Ģ份
    -0.10
    éĢļå¾Ģ
    -0.10
    åĴĮå®ĮåĸĦ
    -0.10
    æł·çļĦ
    -0.09
    éĢļè¡Į
    -0.09
    åĬ¿å¤´
    -0.09
    çļĦæ¨¡æł·
    -0.09
    POSITIVE LOGITS
    éĹ®é¢ĺ
    0.10
    ç»ĵæŀĦæĢ§
    0.09
    ä¸ĬçļĦ
    0.09
    rror
    0.08
     error
    0.08
     problems
    0.08
    é£ĺ
    0.08
    æĸ¹éĿ¢çļĦ
    0.08
    被
    0.08
    ä¸Ńå¤ĸ
    0.08
    Activations Density 0.599%

    No Known Activations