© 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. 5033
    Prev
    Next
    INDEX
    Explanations

    report

    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
    ç¢ĵ
    -0.11
    æī¹å¤į
    -0.10
    åĩºåħ·
    -0.09
    éļĹ
    -0.09
    æĸĩ书
    -0.09
    åĪĨå·¥
    -0.09
    ä¹ĭäºĭ
    -0.08
    ISTA
    -0.08
    ä¸Ģç¯ĩæĸĩç«ł
    -0.08
    åİī
    -0.08
    POSITIVE LOGITS
     зÑĢ
    0.10
    æľīèī²
    0.09
    adol
    0.09
    andes
    0.09
    ди
    0.09
    å±ħ
    0.09
    åĿª
    0.08
    _align
    0.08
    åĹŁ
    0.08
    ivals
    0.08
    Activations Density 0.025%

    No Known Activations