© 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. 26930
    Prev
    Next
    INDEX
    Explanations

    carbon fibre or box structure

    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
    èĸ®
    -0.10
    ลาà¸Ļ
    -0.09
    å«Į
    -0.09
     Ber
    -0.09
    磨æįŁ
    -0.09
    å®ķ
    -0.09
    å¸Ńåį·
    -0.09
    èĥª
    -0.09
    ä¸Ģæŀļ
    -0.09
    潤
    -0.08
    POSITIVE LOGITS
    ç«¿
    0.10
    egra
    0.09
    æ¡¢
    0.08
    æ¡ħ
    0.08
    大éĻĨ
    0.08
     Truyá»ģn
    0.08
    ipples
    0.08
     forg
    0.08
    èĢĮå·²
    0.08
     haircut
    0.08
    Activations Density 0.109%

    No Known Activations