© 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. 40184
    Prev
    Next
    INDEX
    Explanations

    press followed by action

    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
    èµŃ
    -0.10
    åħ¨èº«
    -0.09
    iverse
    -0.09
    åľ°å¸¦
    -0.09
    å¹²
    -0.09
     Lowell
    -0.08
     Aires
    -0.08
    æĮ¤
    -0.08
    çļĦ身影
    -0.08
    çĶ¨æĪ·æıIJä¾Ľ
    -0.08
    POSITIVE LOGITS
    anan
    0.09
    odem
    0.08
    reta
    0.08
    èĩªçĦ¶èĢĮ
    0.08
    å°±è¿ij
    0.08
    ä¸ĭåİ»
    0.08
    å¾Ĺå¤ļ
    0.08
    æĶ¾åģĩ
    0.08
    ãģĦãģŁãĤī
    0.08
    ãĤ¬ãĤ¤
    0.08
    Activations Density 0.053%

    No Known Activations