© 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. 22774
    Prev
    Next
    INDEX
    Explanations

    pilot studies

    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
    ä¸ĸ家
    -0.09
    rc
    -0.09
    å±ıéļľ
    -0.08
    éĩįè§Ĩ
    -0.08
    umm
    -0.08
    лик
    -0.08
    olian
    -0.08
    ource
    -0.08
    wed
    -0.08
    rs
    -0.08
    POSITIVE LOGITS
    ä¸Ģ个éĹ®é¢ĺ
    0.10
    åķ¼
    0.10
    å¼ģ
    0.10
    æ»´æ»´
    0.09
    ä¼ļåıijçĶŁ
    0.09
     reins
    0.09
    èĬ
    0.08
    ãģ¦ãģĦ
    0.08
    lane
    0.08
    ä¸įæİĴéĻ¤
    0.08
    Activations Density 0.018%

    No Known Activations