© 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. 30710
    Prev
    Next
    INDEX
    Explanations

    typesetting typography

    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
    妲
    -0.11
    娥
    -0.10
    åıĻ
    -0.10
    zion
    -0.09
    ç»ĺ
    -0.09
    åŤ
    -0.09
    wire
    -0.09
    亲
    -0.09
    ç¬ij声
    -0.09
    ä½İä½į
    -0.08
    POSITIVE LOGITS
    åį°åĪ·
    0.11
     Fut
    0.09
    ems
    0.08
    bir
    0.08
    è´ŁéĿ¢
    0.08
    éĽ¢ãĤĮ
    0.08
    emb
    0.08
    åį°
    0.08
    临è¿ij
    0.08
    isp
    0.08
    Activations Density 0.136%

    No Known Activations