© 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. 23468
    Prev
    Next
    INDEX
    Explanations

    names like Rooney, Ryder, Nani

    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
    æį
    -0.09
    æĵ¦
    -0.09
    åŃķ
    -0.09
    о
    -0.09
    æ´ŀ
    -0.09
    ä¸İæĹ¶ä¿±è¿Ľ
    -0.09
     Pey
    -0.08
    åģı
    -0.08
    eda
    -0.08
    ledo
    -0.08
    POSITIVE LOGITS
    æµij
    0.09
    ç©°
    0.09
    åıĪ好
    0.09
    çŃIJ
    0.08
    çijĽ
    0.08
    é«ĺä½į
    0.08
    åģĥ
    0.08
    bil
    0.08
    ä¸Ģ个å¤ļ
    0.08
    acha
    0.08
    Activations Density 0.022%

    No Known Activations