© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 48027
    Prev
    Next
    INDEX
    Explanations

    reports to

    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
    ç®ĹäºĨ
    -0.10
    çļĦçľ¼åħī
    -0.09
    大æĪ·
    -0.09
    泪
    -0.09
    ãģĤãģ£ãģ¦
    -0.09
    ä¸ĸ家
    -0.09
    éķı
    -0.09
    à¹Ģà¸ī
    -0.09
    éĤ£ä¹Ī容æĺĵ
    -0.09
    åĴĭ
    -0.09
    POSITIVE LOGITS
    è°Īè¯Ŀ
    0.10
    enez
    0.09
    çĨĻ
    0.09
     Ø¥ÙĦÙĬÙĩا
    0.09
     Chip
    0.09
    izi
    0.08
    вен
    0.08
    =index
    0.08
    eci
    0.08
    -ground
    0.08
    Activations Density 0.010%

    No Known Activations