© 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. 3756
    Prev
    Next
    INDEX
    Explanations

    permitted, prohibited, modify, must

    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
     hed
    -0.10
    å¾ģéĽĨ
    -0.09
    INES
    -0.09
    åij¨ä¸Ģ
    -0.09
    åĽĽæĸ¹
    -0.08
    åĨ²
    -0.08
    æīĵæĭĽåij¼
    -0.08
    æľĢåıĹ欢è¿İ
    -0.08
    ä¸įåħĭ
    -0.08
    审
    -0.08
    POSITIVE LOGITS
     Narc
    0.09
    èĥĮåIJİçļĦ
    0.09
    .bel
    0.08
    å¡ŀ
    0.08
    _ble
    0.08
     bush
    0.08
     backing
    0.08
     бÑİ
    0.08
     pog
    0.08
    çļĦä¸Ģéĥ¨åĪĨ
    0.08
    Activations Density 0.010%

    No Known Activations