© 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. 1622
    Prev
    Next
    INDEX
    Explanations

    with descriptions of features

    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
    åĸī
    -0.10
    çłļ
    -0.09
    ripe
    -0.09
    ç¼£
    -0.09
    è¿Ļæī¹
    -0.09
    ä¸įå¦Ĥ
    -0.09
    磾
    -0.09
    สà¸Ķ
    -0.09
    ä¾ĿçĦ¶æĺ¯
    -0.09
    夺
    -0.09
    POSITIVE LOGITS
    æ¼ıæ´ŀ
    0.11
    个åĪ«
    0.10
    æ·
    0.10
     عÙĦÙĬÙĩا
    0.09
     holes
    0.09
    holes
    0.09
    èªī
    0.09
    embedded
    0.09
    å¾½
    0.08
    oci
    0.08
    Activations Density 0.103%

    No Known Activations