© 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. 62945
    Prev
    Next
    INDEX
    Explanations

    Whittaker Chambers, Scopes Trial, Breitbart, Ford

    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
    Cour
    -0.09
    æ°¸ä¹ħ
    -0.09
     Cour
    -0.09
    utton
    -0.08
     subs
    -0.08
    idden
    -0.08
    ĥģ
    -0.08
     пÑĢием
    -0.08
    åľ°æĿ¿
    -0.08
     Foster
    -0.08
    POSITIVE LOGITS
    :NS
    0.09
    è¿Ļ个人
    0.09
    ä¸į足以
    0.09
    çľĭçĶµå½±
    0.09
     Ø¢ÙĦ
    0.09
     Shirt
    0.09
    çľĭçĿĢ
    0.08
    模
    0.08
    iore
    0.08
    循
    0.08
    Activations Density 0.046%

    No Known Activations