© 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. 27282
    Prev
    Next
    INDEX
    Explanations

    at best / least / your option

    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
    æĢİä¹Īåģļ
    -0.09
    idden
    -0.09
    é²į
    -0.09
    åı¦ä¸ĢåįĬ
    -0.09
    lds
    -0.09
    é¾Ł
    -0.09
    ä»Ģä¹ĪæĦıæĢĿ
    -0.08
    å®īåħ¨æĢ§
    -0.08
     Fletcher
    -0.08
    몰
    -0.08
    POSITIVE LOGITS
    èµ·çłģ
    0.10
    éģij
    0.10
    满äºĨ
    0.10
    æĽ´ä½ķåĨµ
    0.09
    çļĦæł¹æľ¬
    0.09
    æľĢå¤ļ
    0.09
    ç®ħ
    0.09
     worst
    0.09
    æľĢå°ij
    0.09
    à¹Ģà¸Ĺ
    0.08
    Activations Density 0.042%

    No Known Activations