© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 5290
    Prev
    Next
    INDEX
    Explanations

    scientific journals and research papers

    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
    Not in Any Lists

    No Comments

    Negative Logits
    åłª
    -0.09
    æĭħ
    -0.09
    zung
    -0.08
    åħ¢
    -0.08
    ê²ģ
    -0.08
    åĶ®
    -0.07
    渲
    -0.07
     caus
    -0.07
    åıįé¦Ī
    -0.07
    çļĦæĪ¿åŃIJ
    -0.07
    POSITIVE LOGITS
    èµĽåĮº
    0.09
    ฯ
    0.09
    itore
    0.09
    æ¢ĵ
    0.09
    lete
    0.08
    æĸĩåĮĸ建设
    0.08
    èµ°è¿ij
    0.08
    èĥ¤
    0.08
    å©·
    0.08
    å«ī
    0.08
    Activations Density 0.151%

    No Known Activations