© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 19362
    Prev
    Next
    INDEX
    Explanations

    body parts: limbs, head, torso

    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
    çľ¨çľ¼
    -0.10
    åݨæĪ¿
    -0.09
    <pair
    -0.09
    èĢħçļĦ
    -0.09
    èĢħãģ®
    -0.09
    è¿ĩç¨ĭå½ĵä¸Ń
    -0.08
     eh
    -0.08
    çľģéĴ±
    -0.08
    å®ļä½į
    -0.08
    fall
    -0.08
    POSITIVE LOGITS
    èĭ¥æľī
    0.11
    ä¹ĥèĩ³
    0.10
    è¿Ļä¸ĢçĤ¹
    0.10
    ëģ¼
    0.09
    اÙĪر
    0.09
    两级
    0.09
    蹶
    0.09
    ç£ķ
    0.08
    娣
    0.08
    èĥ¤
    0.08
    Activations Density 0.050%

    No Known Activations