© 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. 39429
    Prev
    Next
    INDEX
    Explanations

    overall assessment

    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
    ä»ħä»ħæĺ¯
    -0.11
    ä¸ĩä¸Ģ
    -0.10
    ç¯
    -0.10
    äºİæĺ¯
    -0.10
    ä»ħä»ħ
    -0.09
    çĽ²
    -0.09
     ÑģобÑģÑĤвенно
    -0.08
    ierz
    -0.08
     mere
    -0.08
     certain
    -0.08
    POSITIVE LOGITS
    æķ´ä½ĵ
    0.13
    æĢ»ä½ĵ
    0.12
    çļĦæķ´ä½ĵ
    0.11
    Overall
    0.09
     overall
    0.09
    çļĦè¿ĽæŃ¥
    0.09
    æķ´ä¸ª
    0.09
     пÑĢедÑģÑĤавлÑı
    0.09
    ä¸Ńå°ıåŀĭ
    0.09
    æķ´åĢĭ
    0.09
    Activations Density 0.047%

    No Known Activations