© 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. 52350
    Prev
    Next
    INDEX
    Explanations

    bug fixes

    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
    æĮ¯
    -0.09
    misc
    -0.09
    èĨij
    -0.09
    ãģ£ãģįãĤĬ
    -0.08
    åģļçļĦ
    -0.08
    ë¡Ŀ
    -0.08
    صاب
    -0.08
    ä¸ŃåİŁ
    -0.08
    为ä»Ģä¹Īä¸į
    -0.08
    ä¸Ģäºĭ
    -0.08
    POSITIVE LOGITS
    bug
    0.10
    ç¾ļ
    0.09
    /errors
    0.09
    Gratis
    0.09
    è®´
    0.09
     ÐijеÑģ
    0.09
    çĬĩ
    0.09
    ç½ħ
    0.08
    大åıĶ
    0.08
    åĬ²
    0.08
    Activations Density 0.045%

    No Known Activations