© 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. 50787
    Prev
    Next
    INDEX
    Explanations

    adventures of

    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
    大æĪ·
    -0.09
     toll
    -0.09
    çŀŃ
    -0.08
    æŀ¥
    -0.08
     Canary
    -0.08
    佬
    -0.08
    æĴ©
    -0.08
    å£ķ
    -0.08
    atori
    -0.08
    otas
    -0.08
    POSITIVE LOGITS
    æĬĽå¼ĥ
    0.10
    usted
    0.09
    ç½½
    0.09
    çĺĹ
    0.09
    лиÑĤ
    0.09
    hub
    0.08
    æĺ¯ä¸Ģ款
    0.08
    åĴĮå®ĮåĸĦ
    0.08
    è®°ä½ı
    0.08
    å¼ıçļĦ
    0.08
    Activations Density 0.075%

    No Known Activations