© 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. 56918
    Prev
    Next
    INDEX
    Explanations

    receive a

    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
    ights
    -0.10
    说ä»Ģä¹Ī
    -0.10
    ä»Ģä¹ĪæŶåĢĻ
    -0.10
    راس
    -0.09
    ë§ī
    -0.09
    å§Ķ书记
    -0.09
    оÑģÑĮ
    -0.09
    IGHT
    -0.09
    å®ļäºĨ
    -0.09
    ör
    -0.09
    POSITIVE LOGITS
    ance
    0.11
    ivity
    0.11
    oder
    0.10
    ables
    0.10
    à¸ļาà¸Ķ
    0.10
    ably
    0.09
    /send
    0.09
     ÙħÙĨÙĩا
    0.09
    asm
    0.09
    ê°ģ
    0.09
    Activations Density 0.082%

    No Known Activations