© 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. 27692
    Prev
    Next
    INDEX
    Explanations

    .component

    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
    éģ´
    -0.09
    åĬŁå¤«
    -0.09
    ãģ¨ãģĦãģĨãĤĤãģ®
    -0.09
    åĻľ
    -0.08
     giá»ijng
    -0.08
    çĤ¹èµŀ
    -0.08
    ÑĢод
    -0.08
    łĢ
    -0.08
    ders
    -0.08
    ventory
    -0.08
    POSITIVE LOGITS
    OUCH
    0.10
    åĪ¨
    0.09
    venta
    0.09
    æĥ®
    0.08
    éĻĶ
    0.08
    å½»
    0.08
     erk
    0.08
    yme
    0.08
    breadcrumb
    0.08
    å®Įæķ´æĢ§
    0.08
    Activations Density 0.205%

    No Known Activations