© 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. 59176
    Prev
    Next
    INDEX
    Explanations

    talking again, communication, relationships

    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
    羡æħķ
    -0.09
    åīįè¾Ī
    -0.09
    æķıéĶIJ
    -0.09
    空
    -0.08
     Alert
    -0.08
    ä¸įçĶĺ
    -0.08
    çīºçī²
    -0.08
    éĢļè¿ĩ对
    -0.08
    èѦ
    -0.08
    erties
    -0.08
    POSITIVE LOGITS
    è§ģéĿ¢
    0.11
    缸è§ģ
    0.11
    éļĶ
    0.10
    aiser
    0.10
    æݥ触
    0.10
    è¿ŀ线
    0.10
    亲æĥħ
    0.09
    äºĴåĬ¨
    0.09
    SetActive
    0.09
    忤
    0.09
    Activations Density 0.097%

    No Known Activations