© 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. 30994
    Prev
    Next
    INDEX
    Explanations

    side of

    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
    举éĥ¨
    -0.10
    çĻ¾å®¶
    -0.10
    ä¸ĩæĪ·
    -0.10
    olest
    -0.09
    icode
    -0.09
    åıªåī©ä¸ĭ
    -0.09
    utton
    -0.09
    ihar
    -0.09
    ä¼°
    -0.09
    æ¬ł
    -0.09
    POSITIVE LOGITS
    ä¸Ģéģĵ
    0.09
     incl
    0.09
    izer
    0.08
    æĶ¹éĿ©åĪĽæĸ°
    0.08
     flip
    0.08
    ati
    0.08
    ä¸ĢèĤ¡
    0.08
    åĦ¿
    0.08
    (Node
    0.07
    åĪļæīį
    0.07
    Activations Density 0.034%

    No Known Activations