© 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. 30735
    Prev
    Next
    INDEX
    Explanations

    asking best/worst/first

    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
    å±
    -0.09
    æĻ¯å¾·
    -0.09
    èĢĦ
    -0.09
    迸
    -0.08
    å¹´æľ«
    -0.08
    çļĦä¸ĢéĿ¢
    -0.08
    讪
    -0.08
    æ¯ı次éĥ½
    -0.08
    äs
    -0.08
    çĪ±äºº
    -0.08
    POSITIVE LOGITS
    idis
    0.10
    htt
    0.09
    æ¸į
    0.09
    çĶ±æѤ
    0.09
    enis
    0.09
    å¹¼ç¨ļ
    0.09
    IVES
    0.09
    å°±æĺ¯åľ¨
    0.08
    å°ļ
    0.08
    interop
    0.08
    Activations Density 0.044%

    No Known Activations