© 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. 55255
    Prev
    Next
    INDEX
    Explanations

    positive descriptive 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
    éĢĻ種
    -0.11
    æį®æѤ
    -0.10
    è¿Ļç§į
    -0.10
    éĢĻ樣
    -0.10
    è¿ĻæŶåĢĻ
    -0.10
    è¿Ļ个è¯į
    -0.10
    è¿Ļæł·çļĦ
    -0.10
     ÑįÑĤомÑĥ
    -0.10
    è¿Ļäºĭ
    -0.10
    è¿Ļæł·æīįèĥ½
    -0.10
    POSITIVE LOGITS
    ä¹Łä¸įä¾ĭå¤ĸ
    0.10
    ìłĢ
    0.10
    ä¸įæĺ¯ä¸Ģ个
    0.09
     happening
    0.09
    åijĬè¯ī大家
    0.09
    æĥ³äºĨ
    0.09
    æĪIJ为ä¸Ģ个
    0.09
    å±ĢéĿ¢
    0.09
    ÑģÑĤоÑı
    0.09
    ãģĭ
    0.09
    Activations Density 0.359%

    No Known Activations