© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 47623
    Prev
    Next
    INDEX
    Explanations

    cone

    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
    Negative Logits
    件
    -0.09
     bolt
    -0.09
    ente
    -0.09
    æĬī
    -0.08
    cation
    -0.08
    aren
    -0.08
    æĻ¯å¾·
    -0.08
    èıģ
    -0.08
    errer
    -0.08
    wer
    -0.08
    POSITIVE LOGITS
    å°ĸ
    0.11
    egin
    0.10
    ä¸Ģæł¹
    0.10
    ights
    0.09
    chio
    0.09
    butt
    0.09
    åĿIJçĿĢ
    0.09
    空åīį
    0.09
    è¦ģåѦä¼ļ
    0.09
    ä¸ļæĢģ
    0.08
    Activations Density 0.011%

    No Known Activations