© 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. 20074
    Prev
    Next
    INDEX
    Explanations

    first of the year

    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
    ses
    -0.11
    绦
    -0.10
    ä¸īå¹´
    -0.10
     backstage
    -0.10
    å¤ļå¹´
    -0.09
    åĻ±
    -0.09
    æĪijå¸ĮæľĽ
    -0.09
    åįģå¹´
    -0.09
    å¤ļå¹´çļĦ
    -0.09
    ä¸ĥå¹´
    -0.09
    POSITIVE LOGITS
    åĶ±
    0.09
    æĶ¯
    0.09
    éĢıæĺİ
    0.08
     Wins
    0.08
     Rice
    0.08
    onta
    0.08
    xin
    0.08
    vice
    0.08
     Sing
    0.08
    Âł
    0.08
    Activations Density 0.042%

    No Known Activations