© 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. 47644
    Prev
    Next
    INDEX
    Explanations

    quoted link, box, button, or option

    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
    ç¼Ń
    -0.11
    äºĨä¸įå°ij
    -0.10
    аÑģÑĤ
    -0.09
    åħ¶ä»ĸçļĦ
    -0.09
    éĢµ
    -0.09
    è¯
    -0.09
    rud
    -0.09
    IRD
    -0.09
    ç®Ĺæĺ¯
    -0.08
    ç¼ij
    -0.08
    POSITIVE LOGITS
     approach
    0.11
     portion
    0.10
    idy
    0.09
    portion
    0.09
    appro
    0.09
    åĽŀäºĭ
    0.09
    els
    0.08
    eters
    0.08
    åŃĹæł·
    0.08
    举åĬ¨
    0.08
    Activations Density 0.191%

    No Known Activations