© 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. 56886
    Prev
    Next
    INDEX
    Explanations

    deserved, worthy, capable

    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
    å±ij
    -0.10
    è°¢è°¢ä½ł
    -0.10
    èħ¾é£ŀ
    -0.09
    æĿ¥æī¾
    -0.09
    æįĨ
    -0.09
    itta
    -0.09
    åĬłçĽĸ
    -0.09
    tings
    -0.09
    é¾Ļæ¹ĸ
    -0.08
    ViewById
    -0.08
    POSITIVE LOGITS
    åĢ¼å¾Ĺ
    0.10
    èĢħ
    0.10
    å®Įåħ¨åı¯ä»¥
    0.09
     alt
    0.09
    ERE
    0.08
    å¾Ĺèµ·
    0.08
    ombat
    0.08
    ridge
    0.08
    igits
    0.08
    bam
    0.08
    Activations Density 0.004%

    No Known Activations