© 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. 26262
    Prev
    Next
    INDEX
    Explanations

    metadata or structured attributes

    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
    å¦ĩ
    -0.09
    è£
    -0.09
    åĽ½å®¶æĪĺçķ¥
    -0.09
    婦
    -0.09
     VE
    -0.09
    åĤ¢
    -0.08
    èĮµ
    -0.08
     Transitional
    -0.08
    ä¹Łå¥½
    -0.08
    newInstance
    -0.08
    POSITIVE LOGITS
    ares
    0.08
    çħ®
    0.08
    azo
    0.08
    æĹłæĦı
    0.08
     taxing
    0.08
    .Typed
    0.08
    atta
    0.08
    aud
    0.07
     asses
    0.07
    ever
    0.07
    Activations Density 0.248%

    No Known Activations