© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 52302
    Prev
    Next
    INDEX
    Explanations

    China, Chinese, countries, debt

    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
     Bus
    -0.10
    åİŁ
    -0.10
    çħ§äº®
    -0.09
    bus
    -0.09
    çĸ¯
    -0.09
     bonus
    -0.09
    Bus
    -0.08
    群éĩĮ
    -0.08
    çĥ¬
    -0.08
     bus
    -0.08
    POSITIVE LOGITS
    æĹ©æĻļ
    0.11
    ãģĽãģļ
    0.09
    åĨįåĪ°
    0.09
    ionale
    0.09
    enaire
    0.08
     sacr
    0.08
     Koreans
    0.08
    china
    0.08
    å¤ļå°ij
    0.08
    ิà¸ĩ
    0.08
    Activations Density 0.111%

    No Known Activations