© 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. 56462
    Prev
    Next
    INDEX
    Explanations

    definitive authoritative source

    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
    æ¹¾
    -0.09
     вÑĭбÑĢаÑĤÑĮ
    -0.09
    éĥ´
    -0.09
    好åĩłä¸ª
    -0.09
    æĮīè§Ħå®ļ
    -0.08
    æ»Ĥ
    -0.08
    ве
    -0.08
    å±IJ
    -0.08
    usses
    -0.08
     dum
    -0.08
    POSITIVE LOGITS
    æĿĥå¨ģ
    0.10
    ilter
    0.09
    Authority
    0.09
    theses
    0.08
     benchmarks
    0.08
    iltr
    0.08
     Authority
    0.08
    è¿ĩ硬
    0.08
    ä¹ĭä¸Ģ
    0.08
    åľ°éģĵ
    0.08
    Activations Density 0.064%

    No Known Activations