© 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. 35223
    Prev
    Next
    INDEX
    Explanations

    Marquette University, Villanova University

    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
    å¹ķ
    -0.09
    å¤
    -0.09
    éľ
    -0.09
    çĤĶ
    -0.08
    ä¸ĸçķĮçļĦ
    -0.08
    ãĤĤãģ¡ãĤįãĤĵ
    -0.08
    rtle
    -0.08
    è¯į
    -0.08
    cole
    -0.08
    ivate
    -0.08
    POSITIVE LOGITS
    leep
    0.10
    åij½è¿IJ
    0.09
    нÑı
    0.09
    ender
    0.09
    اج
    0.09
    ampo
    0.09
    law
    0.09
    å¥ĭè¿Ľ
    0.08
    acen
    0.08
    ä¸Ģ款
    0.08
    Activations Density 0.004%

    No Known Activations