© 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. 32057
    Prev
    Next
    INDEX
    Explanations

    flavor and aroma

    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
     Passive
    -0.10
    èĬ³é¦Ļ
    -0.09
     passive
    -0.09
    樵
    -0.08
    åıijæĮ¥äºĨ
    -0.08
    бÑĭ
    -0.08
     GRAPH
    -0.08
    mé
    -0.08
    ä¼ļè°Ī
    -0.08
    管çIJĨåijĺ
    -0.08
    POSITIVE LOGITS
    çĶ·åŃ©
    0.09
    ç¾İè²Į
    0.09
    ä¸Ģ次æĢ§
    0.08
    èĦ¾æ°Ķ
    0.08
     PTR
    0.08
    åĺı
    0.08
    ä¸Ģ身
    0.08
    umann
    0.08
    ä¼ł
    0.08
    模仿
    0.08
    Activations Density 0.031%

    No Known Activations