© 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. 50646
    Prev
    Next
    INDEX
    Explanations

    idea of

    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
    alien
    -0.09
    ä¿Ń
    -0.09
    chy
    -0.09
    /root
    -0.09
    æĪij说
    -0.08
     GRAT
    -0.08
     momentum
    -0.08
    äºĴè¡¥
    -0.08
    eron
    -0.08
    åĬłåĪĨ
    -0.08
    POSITIVE LOGITS
    å¿ĥ缮
    0.11
    ä»Ģä¹Īåı«
    0.10
    æĸ¹å¼ı
    0.09
    สว
    0.08
    èĩ§
    0.08
    æ°Ļ
    0.08
    çĪ°
    0.08
    æīĢè°ĵ
    0.08
    abouts
    0.08
    à¹ģà¸Ĺ
    0.08
    Activations Density 0.169%

    No Known Activations