© 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. 32958
    Prev
    Next
    INDEX
    Explanations

    code headers and comments

    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
     кÑĢÑĥ
    -0.09
    è¯
    -0.08
    ä½Ĩä¸įéĻIJäºİ
    -0.08
     brag
    -0.08
    OWN
    -0.07
     Marvin
    -0.07
    æݨéĶĢ
    -0.07
    üh
    -0.07
    æĦıä¹īä¸ĬçļĦ
    -0.07
    åį¯
    -0.07
    POSITIVE LOGITS
    æīĭèĦļ
    0.10
    ynes
    0.09
    æ¼ł
    0.09
    èĨĽ
    0.09
    .Xr
    0.08
    èѬå¦Ĥ
    0.08
    æľ¦èĥ§
    0.08
    itim
    0.08
    çģ«çĥ§
    0.08
    detector
    0.08
    Activations Density 0.595%

    No Known Activations