© 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. 18287
    Prev
    Next
    INDEX
    Explanations

    if there is

    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
    ä¸ĢèĪ¬
    -0.09
    riad
    -0.08
    两类
    -0.08
    ãĥ¬ãĤ¹
    -0.08
    eg
    -0.08
    èĥĮéĥ¨
    -0.08
    éĵį
    -0.08
    çĶŁæ´»æĸ¹å¼ı
    -0.08
    è¯ķè¯ķ
    -0.08
    assen
    -0.08
    POSITIVE LOGITS
    æľīä»Ģä¹Ī
    0.10
    æİ
    0.09
    è¿ĺæľīä»Ģä¹Ī
    0.09
    æŁ¢
    0.09
     doch
    0.09
    åIJĮåѦ
    0.09
    å¤ļä¹Ī
    0.09
     ÑħоÑĤÑĮ
    0.09
    èĪĤ
    0.09
    /if
    0.09
    Activations Density 0.025%

    No Known Activations