© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 52364
    Prev
    Next
    INDEX
    Explanations

    of the

    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
    åIJĵ
    -0.09
    icks
    -0.09
    äººåľ¨
    -0.08
    ÑģÑı
    -0.08
    .Deep
    -0.08
    åħľ
    -0.08
    ï¸ı
    -0.08
    å®ĺ宣
    -0.08
     Bul
    -0.08
    dish
    -0.08
    POSITIVE LOGITS
    æĪijå¾Ī
    0.09
    ç¨Ķ
    0.09
    éĽĨèģļ
    0.08
    ç½Ķ
    0.08
    ä¸Ģå¹´ä¸Ģ度
    0.08
    沿çĿĢ
    0.08
    æľīäºĽäºº
    0.08
    讨åİĮ
    0.08
    .epoch
    0.08
    лаг
    0.08
    Activations Density 0.098%

    No Known Activations