© 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. 50265
    Prev
    Next
    INDEX
    Explanations

    limits

    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
     intention
    -0.09
    å¤įæ´»
    -0.09
    欢è¿İ
    -0.09
     inst
    -0.09
    éĩį建
    -0.09
    å¾Ĺ以
    -0.08
     Woods
    -0.08
    é¢ĦåĪ¶
    -0.08
    çľļ
    -0.08
    /views
    -0.08
    POSITIVE LOGITS
    åĨįä¹Ł
    0.14
    ä¸įä¼ļåĨį
    0.13
    ä¸įèĥ½åĨį
    0.13
    饱åĴĮ
    0.12
    æµij身
    0.12
    çĵ¶é¢Ī
    0.12
     exhausted
    0.12
    ä¸įåĨį
    0.11
    å¢ŀéĢŁ
    0.11
    åĨįä¹Łä¸į
    0.11
    Activations Density 0.123%

    No Known Activations