© 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. 7189
    Prev
    Next
    INDEX
    Explanations

    subsidies

    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
    tight
    -0.10
    幡
    -0.09
    çīĪæĿĥå½Ĵ
    -0.09
    å§Ķ
    -0.09
    éĥ½ä¸įæķ¢
    -0.09
    fid
    -0.09
    éķ¿æ±Ł
    -0.08
    è¾ĸ
    -0.08
    æĸIJ
    -0.08
    è¡Į使
    -0.08
    POSITIVE LOGITS
     subsidies
    0.13
    补贴
    0.13
     subsid
    0.12
    æī¶æĮģ
    0.12
     subsidy
    0.12
    æĶ¯æĮģ
    0.10
    äºıæįŁ
    0.10
    æĶ¯æĮģåĬĽåº¦
    0.10
    æĬĢæľ¯æĶ¯æĮģ
    0.10
    æķijåĬ©
    0.09
    Activations Density 0.044%

    No Known Activations