© 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. 18468
    Prev
    Next
    INDEX
    Explanations

    .github or .micro

    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
    以人æ°ij
    -0.10
    çļĦåħ³éĶ®
    -0.09
    æ²¥
    -0.09
    ç±³ç²ī
    -0.08
    éĺħ读åħ¨æĸĩ
    -0.08
    å»īä»·
    -0.08
    æĿ¥ä¸´
    -0.08
    anja
    -0.08
    éļ¾åħ³
    -0.08
    åľ¨å¹¿å·ŀ
    -0.08
    POSITIVE LOGITS
    ETO
    0.09
    èĩ¬
    0.09
     SC
    0.09
    apa
    0.09
    RITE
    0.08
     Dut
    0.08
    åIJIJ
    0.08
     hundred
    0.08
    uliar
    0.08
    vida
    0.08
    Activations Density 0.001%

    No Known Activations