© 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. 54033
    Prev
    Next
    INDEX
    Explanations

    inferring conclusions and causality

    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
    è¾Ł
    -0.10
    oden
    -0.10
    æľīæĿ¡ä»¶
    -0.09
    å¸Ľ
    -0.09
    æĬ±
    -0.08
    èµĺ
    -0.08
    æİ¥åħ¥
    -0.08
     Teen
    -0.08
    elf
    -0.08
    ç¿ķ
    -0.08
    POSITIVE LOGITS
    æݨçIJĨ
    0.14
    ä»İä¸Ń
    0.14
     infer
    0.13
    å¾Ĺåĩº
    0.13
     extrapol
    0.12
     ÙħÙĨÙĩ
    0.12
     dara
    0.12
     deduct
    0.11
    ç»ĵ论
    0.11
     conclusions
    0.10
    Activations Density 0.089%

    No Known Activations