© 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. 53719
    Prev
    Next
    INDEX
    Explanations

    well, over, above, below, worth

    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
    éĿŀè¦ģ
    -0.09
    æ²»
    -0.08
    æĽ´å¤§çļĦ
    -0.08
    æĽ´å¤§
    -0.08
    race
    -0.08
    æ¶ī
    -0.08
    ntl
    -0.08
    à¹īà¸Ńย
    -0.08
     ster
    -0.08
    ciÄħ
    -0.08
    POSITIVE LOGITS
    讫
    0.09
    åĩłæŃ¥
    0.09
    箪
    0.09
    zes
    0.09
    è¶ħè¿ĩäºĨ
    0.09
    ä¸ĬæĿ¥
    0.09
    ãģĹãģ¦ãģıãĤĮ
    0.09
    ç¼ļ
    0.08
    è¿ĩäºĨ
    0.08
    ä¸Ńåįİæ°ijæĹı
    0.08
    Activations Density 0.200%

    No Known Activations