© 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. 54074
    Prev
    Next
    INDEX
    Explanations

    awards for best category

    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
    为ä¸Ńå¿ĥ
    -0.10
    ä¸įåIJĮç¨ĭ度
    -0.09
    åħ¬åħ±èµĦæºIJ
    -0.09
    eo
    -0.08
    éĢĻ麼
    -0.08
    èĻij
    -0.08
    essel
    -0.07
    åıijåĶ®
    -0.07
    -fi
    -0.07
    åģļä¸ĢäºĽ
    -0.07
    POSITIVE LOGITS
    æľĢä½³
    0.14
    å¥ĸ项
    0.13
    çļĦæľĢä½³
    0.12
    åįķ项
    0.12
    æľĢåħ·
    0.11
    æıIJåIJį
    0.11
    å¥ĸ
    0.11
    -best
    0.10
    .best
    0.10
    两项
    0.10
    Activations Density 0.077%

    No Known Activations