© 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. 19570
    Prev
    Next
    INDEX
    Explanations

    communities

    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
    åĴĮå¹³
    -0.11
    妪
    -0.09
    ä¼ī
    -0.08
     Belt
    -0.08
    æ»Ķ
    -0.08
    éĥĵ
    -0.08
    åĪĨæķ°çº¿
    -0.08
    .Surface
    -0.08
    desk
    -0.08
    Tri
    -0.08
    POSITIVE LOGITS
    ç¯ĩ
    0.11
    ä¸Ńå°ıåŀĭ
    0.10
    amins
    0.09
    ä¸Ģç«Ļ
    0.09
    uting
    0.09
    ighton
    0.09
    chod
    0.08
     lend
    0.08
    unities
    0.08
    èłķ
    0.08
    Activations Density 0.028%

    No Known Activations