© 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. 830
    Prev
    Next
    INDEX
    Explanations

    learned from

    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
    åİħ
    -0.09
    廳
    -0.09
    æī¾åĽŀ
    -0.08
    æĬ¡
    -0.08
    å·¥åľ°
    -0.08
    æľºç¥¨
    -0.08
    .getTable
    -0.08
    æķ£
    -0.08
    çĶŁäº§åĬĽ
    -0.08
    ainers
    -0.08
    POSITIVE LOGITS
    ricks
    0.10
    åIJĦèĩªçļĦ
    0.09
    fully
    0.09
    èĩªèº«
    0.08
    è¿ĩåİ»çļĦ
    0.08
    åĬ¨æĦŁ
    0.08
    vro
    0.08
    ror
    0.08
    èĩªèº«çļĦ
    0.08
     mistakes
    0.08
    Activations Density 0.009%

    No Known Activations