© 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. 5281
    Prev
    Next
    INDEX
    Explanations

    normal, usual, traditional, default

    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
    èĢĮéĿŀ
    -0.10
    å½Ĵ
    -0.09
    帰
    -0.09
    ÑĤин
    -0.09
    .esp
    -0.08
    uck
    -0.08
    顶å±Ĥ
    -0.08
    赶ä¸Ĭ
    -0.08
    azo
    -0.08
    åıĺè¿ģ
    -0.08
    POSITIVE LOGITS
    æ¸
    0.09
    å¸Ĥåİ¿
    0.09
     ones
    0.09
    iki
    0.08
     offerings
    0.08
    è¦ģèĢĥèĻij
    0.08
     Fet
    0.08
    è¦ģåѦä¼ļ
    0.08
    åIJĮç±»
    0.08
    ider
    0.08
    Activations Density 0.118%

    No Known Activations