© 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. 16838
    Prev
    Next
    INDEX
    Explanations

    friends, sisters, partners, spouse

    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
    è¿ĩ度
    -0.10
    anko
    -0.09
    两æīĭ
    -0.09
    æĪijçļĦ
    -0.09
    sex
    -0.09
    åĪ©çĶ¨æĤ¨çļĦ
    -0.09
    dana
    -0.08
    VICE
    -0.08
    åħ¥æīĭ
    -0.08
    verte
    -0.08
    POSITIVE LOGITS
    åľ¨åIJĮä¸Ģ
    0.11
    é»ĺå¥ij
    0.11
    亲å¯Ĩ
    0.11
    å§IJ妹
    0.10
    äºĴ缸
    0.10
    äºĴè¡¥
    0.10
    åľ¨ä¸Ģèµ·
    0.09
     inse
    0.09
    å½¼æѤ
    0.09
    宿
    0.09
    Activations Density 0.080%

    No Known Activations