© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 50996
    Prev
    Next
    INDEX
    Explanations

    end user

    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¸ĭåįĬ
    -0.08
    una
    -0.08
    æĸ°å¸¸æĢģ
    -0.08
    3
    -0.08
    å°ıçĭĹ
    -0.08
    åĮĸå·¥
    -0.08
    æ¼ķ
    -0.08
    ube
    -0.08
    пÑĥÑģк
    -0.08
    åĢ¼å¾Ĺ
    -0.07
    POSITIVE LOGITS
    çĶ¨æĪ·æıIJä¾Ľ
    0.12
    wards
    0.10
    åĵ©
    0.10
    çĶ¨æĪ·çļĦ
    0.10
    æĬµ
    0.09
     Nev
    0.09
    WARDS
    0.09
     koÅĦ
    0.09
    èĤ¢ä½ĵ
    0.09
    ë°ij
    0.09
    Activations Density 0.032%

    No Known Activations