© 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. 49774
    Prev
    Next
    INDEX
    Explanations

    basketball hoop and backboard

    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
    çļĦè§Ĵèī²
    -0.11
    å½¹
    -0.09
    éļħ
    -0.09
    ä¸įä¸ĭ
    -0.09
    èŁł
    -0.09
    åIJĪä½ľåħ³ç³»
    -0.09
    loff
    -0.09
    è§Ĵèī²
    -0.08
    æĮ¨
    -0.08
     Premier
    -0.08
    POSITIVE LOGITS
    andin
    0.11
     Ded
    0.09
    åĪ¶å®ļäºĨ
    0.09
    缴æĴŃéĹ´
    0.09
    æī£
    0.09
    æķŀå¼Ģ
    0.09
     rainbow
    0.09
    anging
    0.09
    é«ĺåĩº
    0.09
    强æľīåĬĽçļĦ
    0.08
    Activations Density 0.047%

    No Known Activations