© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 22025
    Prev
    Next
    INDEX
    Explanations

    speaking, writing, describing, referring

    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
    妥åĸĦ
    -0.10
    åĽłä¸ºæĪij们
    -0.10
    à¹ģม
    -0.09
    åį¿
    -0.09
    éļıæŶ
    -0.09
    acho
    -0.09
    ساÙħ
    -0.08
    à¹ģà¸Ī
    -0.08
    äºĨ解ä¸Ģä¸ĭ
    -0.08
     Lan
    -0.08
    POSITIVE LOGITS
    åľ¨çİ°åľº
    0.10
    ruk
    0.09
    è¿Ļ件äºĭ
    0.09
    åľ°çIJĨ
    0.09
     cryst
    0.08
    кÑĢаÑĤ
    0.08
    ä¸Ģä½ĵåĮĸ
    0.08
    CR
    0.08
    主ç¼ĸ
    0.08
    è¶Ĭ
    0.08
    Activations Density 0.050%

    No Known Activations