© 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. 575
    Prev
    Next
    INDEX
    Explanations

    butterflies and moths

    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
     amen
    -0.11
    éĺ¡
    -0.10
    éĦī
    -0.09
    çĸı
    -0.09
    éĶĦ
    -0.09
    çļĦæĪIJéķ¿
    -0.08
    èĢĥæŁ¥
    -0.08
    AMI
    -0.08
    çĵĴ
    -0.08
    Verb
    -0.08
    POSITIVE LOGITS
    çĦļ
    0.10
    åĪĴ
    0.09
    éķľåŃIJ
    0.09
    ãģĹãĤĩãģĨ
    0.08
    æĸ°éĹ»ä¸Ńå¿ĥ
    0.08
    导åIJij
    0.08
    umu
    0.08
     centrally
    0.08
    jid
    0.08
    shot
    0.08
    Activations Density 0.163%

    No Known Activations