© 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. 53418
    Prev
    Next
    INDEX
    Explanations

    roles

    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
    èĭŀ
    -0.09
    éĢ¢
    -0.09
    portlet
    -0.09
    æīĢåľ¨åľ°
    -0.08
     ven
    -0.08
    atab
    -0.08
    ç½ijåıĭ
    -0.08
    对èĩªå·±çļĦ
    -0.08
    楮
    -0.08
    æ¯ıä½į
    -0.08
    POSITIVE LOGITS
     roles
    0.19
     Roles
    0.17
    è§Ĵèī²
    0.16
    çļĦè§Ĵèī²
    0.16
    Roles
    0.15
    æĭħä»»
    0.14
    roles
    0.14
    _roles
    0.14
    .Roles
    0.13
    å·¥ä½ľå²Ĺä½į
    0.13
    Activations Density 0.146%

    No Known Activations