© 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. 17683
    Prev
    Next
    INDEX
    Explanations

    the word after

    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
    æľīäºĨ
    -0.10
    峡
    -0.09
    ä¸Ģç¢Ĺ
    -0.08
    æľĭåıĭ们对
    -0.08
    eron
    -0.08
    äºĴ缸
    -0.08
    ä¸ĢåĪ»
    -0.08
    è¡Ĺä¸Ĭ
    -0.08
    ä¸įæĸŃçļĦ
    -0.08
    纯
    -0.08
    POSITIVE LOGITS
    ively
    0.10
     Ñģобой
    0.10
    æĪĸèĢħ说
    0.09
    lessly
    0.09
    èĩ³äºİ
    0.09
    ãģ°ãģĭãĤĬ
    0.09
    并éĢļè¿ĩ
    0.09
    ä»¿ä½Ľ
    0.09
    ">&#
    0.09
    ologically
    0.09
    Activations Density 0.369%

    No Known Activations