© 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. 47598
    Prev
    Next
    INDEX
    Explanations

    names after initials or titles

    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
    -0.08
    w
    -0.08
     scratch
    -0.08
     underlying
    -0.08
     comp
    -0.07
    åĪ¹
    -0.07
     Rubio
    -0.07
    ä½ı
    -0.07
    象
    -0.07
    æ´»
    -0.07
    POSITIVE LOGITS
    ени
    0.10
    _Parms
    0.10
    åıĤèĤ¡
    0.10
    æŁ¥çľĭæĽ´å¤ļ
    0.10
     VIR
    0.10
    eri
    0.09
    æĹĮ
    0.09
    åIJĪèµĦ
    0.09
    åŁķ
    0.09
    .mask
    0.09
    Activations Density 0.264%

    No Known Activations