© 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. 16755
    Prev
    Next
    INDEX
    Explanations

    sophisticated

    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
    ings
    -0.10
    绶
    -0.09
    æľIJ
    -0.09
    çķĮçļĦ
    -0.08
    ç¼¥
    -0.08
    /or
    -0.08
    å®ŀéĻħæݧåĪ¶
    -0.08
    æķ´é½IJ
    -0.08
    quist
    -0.08
    аÑģ
    -0.08
    POSITIVE LOGITS
    ly
    0.13
    ness
    0.11
    -minded
    0.11
     '",
    0.10
    ally
    0.10
    ately
    0.09
    iate
    0.09
    ierz
    0.09
    LY
    0.09
    eroon
    0.09
    Activations Density 0.004%

    No Known Activations