© 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. 1386
    Prev
    Next
    INDEX
    Explanations

    door and frame components

    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
    åij¼å£°
    -0.09
    å¤ĩçĶ¨
    -0.09
    æ·¹
    -0.08
     coolest
    -0.08
    aced
    -0.08
    æĥ³ä¸įåĪ°
    -0.08
    å¹´æĬ¥
    -0.08
    ãĢĿ
    -0.08
    å¤ļå¹´çļĦ
    -0.08
    Began
    -0.08
    POSITIVE LOGITS
    éŨåı£
    0.16
    éŨ
    0.15
    éĸĢ
    0.14
    åħ¥éŨ
    0.14
    door
    0.13
    åħ¥åı£
    0.13
     doorway
    0.13
    éŨæĪ·
    0.12
     door
    0.12
    æ´ŀ
    0.12
    Activations Density 0.055%

    No Known Activations