© 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. 4048
    Prev
    Next
    INDEX
    Explanations

    not only, not all, not every

    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
    å¤ĸæĿ¥
    -0.08
    *D
    -0.08
    oser
    -0.08
    å¾Ĥ
    -0.08
    armed
    -0.08
    cul
    -0.08
    зал
    -0.08
    çIJĨæĥ³
    -0.08
    мол
    -0.08
    ãģĬãģĭ
    -0.08
    POSITIVE LOGITS
    çļĦä¸Ģéĥ¨åĪĨ
    0.12
    æīĢæľīçļĦ
    0.11
    ä¸Ģéĥ¨åĪĨ
    0.11
    æīĢæľī人éĥ½
    0.11
    éĥ½ä¼ļæľī
    0.11
    çļĦçģµéŃĤ
    0.11
    æīĢæľī人
    0.10
     모ëĵł
    0.10
    åĵĦ
    0.10
     every
    0.09
    Activations Density 0.021%

    No Known Activations