© 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. 54631
    Prev
    Next
    INDEX
    Explanations

    Harry Potter characters and related phrases

    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
    æłĸ
    -0.09
    çĩĶ
    -0.09
    诰
    -0.09
    åŀ£
    -0.09
    ylum
    -0.09
     Urb
    -0.08
    é£ŀ
    -0.08
    _BOUNDS
    -0.08
    æĶ¾è¿ĩ
    -0.08
    å²ģ以ä¸ĭ
    -0.08
    POSITIVE LOGITS
    æ¥Ķ
    0.10
    è¿ĻæŶ
    0.10
    è¯¾æľ¬
    0.09
    ä¸įèĥ½åĨį
    0.09
    åĨįå¤ļ
    0.09
    etre
    0.09
    çľ¼çĿĽ
    0.09
    è¿ĻæŶåĢĻ
    0.09
    à°³
    0.08
    çĸı
    0.08
    Activations Density 0.046%

    No Known Activations