© 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. 601
    Prev
    Next
    INDEX
    Explanations

    non-english characters or symbols

    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
    getService
    -0.09
    getPost
    -0.09
    èĭį
    -0.08
    ako
    -0.08
    为代表çļĦ
    -0.08
    anga
    -0.08
    ker
    -0.08
    æŀľçĦ¶
    -0.08
    itter
    -0.07
    elay
    -0.07
    POSITIVE LOGITS
    驹
    0.09
    ç«
    0.09
    ä½Ĩä¸įéĻIJäºİ
    0.09
    ë§Īëĭ¤
    0.09
    åĿļå®ŀçļĦ
    0.09
    è¹Ĭ
    0.09
    _PERIOD
    0.09
    imation
    0.09
    佶
    0.09
    社ä¼ļæķĪçĽĬ
    0.08
    Activations Density 0.511%

    No Known Activations