© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 36607
    Prev
    Next
    INDEX
    Explanations

    hard to narrow down

    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
    ç¯ĩç«ł
    -0.09
    èĩªå¸¦
    -0.09
    æľīäºĨ
    -0.08
    å¡Ĭ
    -0.08
    ä¸į平衡
    -0.08
    NullOr
    -0.08
    æ£ĺ
    -0.08
    æĶ¶åĪ°äºĨ
    -0.08
    æĹłäºº
    -0.08
    isors
    -0.08
    POSITIVE LOGITS
    æĺ¯ä¸įåı¯èĥ½
    0.10
    ä¸įåı¯èĥ½
    0.09
    éĢIJä¸Ģ
    0.09
    åĨµä¸Ķ
    0.09
    äºİæĺ¯
    0.09
    ä¸įéĹ´æĸŃ
    0.09
    ä¹Łä¸įåı¯èĥ½
    0.09
    éĻįä»·
    0.08
    åįķçĭ¬
    0.08
    endon
    0.08
    Activations Density 0.056%

    No Known Activations