© 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. 3243
    Prev
    Next
    INDEX
    Explanations

    words appearing at the end of clauses

    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
    Ùij
    -0.09
    “
    -0.09
     Nor
    -0.08
    èĢĮæĺ¯
    -0.08
    ross
    -0.08
    ä¹ĭä¸Ģ
    -0.08
    ende
    -0.07
    åºĶ该å¦Ĥä½ķ
    -0.07
    alu
    -0.07
    ali
    -0.07
    POSITIVE LOGITS
    åıĺäºĨ
    0.10
    oyer
    0.10
    åĵĪå°Ķ
    0.09
    æĪĸèĢħ说
    0.09
    åĺĽ
    0.09
    åķľ
    0.09
    ä¸įæŃ¢
    0.09
     zwar
    0.09
    ãģĦãĤĦ
    0.08
    деÑĢжан
    0.08
    Activations Density 0.158%

    No Known Activations