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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 37477
    Prev
    Next
    INDEX
    Explanations

    this post

    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
    çīĪæĿĥ声æĺİ
    -0.09
    .Lang
    -0.09
    ä½°
    -0.09
    æľ¬ç½ijç«Ļ
    -0.08
    olla
    -0.08
    ç«¥
    -0.08
    åįī
    -0.08
    quiv
    -0.08
     trÆ°ng
    -0.08
    æºľ
    -0.08
    POSITIVE LOGITS
    Ign
    0.09
    åĨħ涵
    0.08
    æIJĢ
    0.08
     Ign
    0.08
    ÑģÑĤеÑĢ
    0.08
    hoe
    0.08
     clo
    0.08
     lean
    0.08
    çļĦç²¾ç¥ŀ
    0.08
    æ²ĸ
    0.08
    Activations Density 0.020%

    No Known Activations