© 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. 33963
    Prev
    Next
    INDEX
    Explanations

    granular and granules

    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
     ÑĦоÑĢ
    -0.09
    гаÑĢ
    -0.09
    毳
    -0.09
    è¡Ģèħ¥
    -0.09
    è£ģ
    -0.09
    å¸
    -0.08
    缸è§ģ
    -0.08
     Instructor
    -0.08
    æĥĬ
    -0.08
    /gtest
    -0.08
    POSITIVE LOGITS
    æķ£
    0.12
    è¦ĨçĽĸ
    0.10
    è¦ĨçĽĸçİĩ
    0.10
    ç¢İ
    0.10
    ढ
    0.09
    iek
    0.09
    åįłæį®äºĨ
    0.09
    磨
    0.09
     screws
    0.09
    trib
    0.08
    Activations Density 0.076%

    No Known Activations