© 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. 49590
    Prev
    Next
    INDEX
    Explanations

    with

    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
    æ³¥
    -0.09
    è¢Ń
    -0.08
    ç¼ķ
    -0.08
    guards
    -0.08
    æ§Ł
    -0.08
     eg
    -0.08
    å»ī
    -0.08
    .dm
    -0.08
    误
    -0.08
    æĬ±
    -0.08
    POSITIVE LOGITS
    ä¼ļåIJĮ
    0.18
    ä¼Ļ
    0.14
    ä¸İåħ¶ä»ĸ
    0.13
    大家ä¸Ģèµ·
    0.12
     together
    0.12
    along
    0.12
    æĪijåĴĮ
    0.11
    åħ±åIJĮ
    0.11
    并ä¸İ
    0.11
     junto
    0.11
    Activations Density 0.093%

    No Known Activations