© 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. 17431
    Prev
    Next
    INDEX
    Explanations

    concepts and vocabulary

    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
    æł·çļĦ
    -0.09
    管家
    -0.08
    æĸ¹åIJij
    -0.08
     ÑĢÑĭ
    -0.08
    æĭį
    -0.08
    æł·
    -0.08
    çĻ»
    -0.08
    容
    -0.08
    rowse
    -0.08
    æ·ĺ
    -0.08
    POSITIVE LOGITS
    æŃ£å¸¸ä½¿çĶ¨
    0.13
    æĮĩçĿĢ
    0.10
    è¦Ĩ
    0.09
    ä¸įä¸ĭ
    0.09
    æ¸ħæĸ°
    0.09
    lá
    0.09
    çĬģ
    0.09
     words
    0.09
    æĻ®éĢļè¯Ŀ
    0.09
    çŃĮ
    0.08
    Activations Density 0.156%

    No Known Activations