© 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. 37921
    Prev
    Next
    INDEX
    Explanations

    separators and placeholders

    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
    dra
    -0.10
     pale
    -0.09
    åİĨæŶ
    -0.09
    ä¸Ĭå¸Ĥ
    -0.09
    åģļä¸įåĪ°
    -0.09
    ãĤ¬ãĤ¤
    -0.09
    eka
    -0.09
    æ¡´
    -0.09
    pel
    -0.08
    çĤ¸
    -0.08
    POSITIVE LOGITS
     isempty
    0.10
    æĺ±
    0.10
    éĦ
    0.09
    è¿ĻåĿĹ
    0.09
    ä¹Łåıªæĺ¯
    0.08
    ruk
    0.08
     wholly
    0.08
    åģ¶
    0.08
    æĪijå·²ç»ı
    0.08
     fur
    0.08
    Activations Density 0.044%

    No Known Activations