© 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. 53685
    Prev
    Next
    INDEX
    Explanations

    each node / block / row / game

    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
    çļĦæīĢæľī
    -0.09
    æīĢæľī
    -0.09
    redo
    -0.08
    åĽ
    -0.08
    é¦ĸæī¹
    -0.08
    zer
    -0.08
    两款
    -0.08
     Tüm
    -0.08
    åı£å¾Ħ
    -0.08
    ç¢İ
    -0.08
    POSITIVE LOGITS
    éĥ½æľīèĩªå·±
    0.13
    æ¯ıä¸Ģ
    0.13
    Each
    0.10
     ÙħÙĨÙĩا
    0.10
    æ¯ı个
    0.10
    ä¸Ģè§Ĵ
    0.10
     каждой
    0.10
     each
    0.09
    æ¯ıåĢĭ
    0.09
    ä¸Ģæľµ
    0.09
    Activations Density 0.093%

    No Known Activations