© 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. 35870
    Prev
    Next
    INDEX
    Explanations

    tracking physical layer information

    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
    erator
    -0.10
    æĮ½åĽŀ
    -0.09
    ools
    -0.09
    æĬķ产
    -0.09
    åĿĿ
    -0.09
    被çĽĹ
    -0.08
     dotyc
    -0.08
    och
    -0.08
    esta
    -0.08
    è¿İæİ¥
    -0.08
    POSITIVE LOGITS
    ãĤĪãģĨãģª
    0.09
    ÑĭÑĪ
    0.09
    anten
    0.09
    人士
    0.09
    _msgs
    0.09
    assets
    0.09
    æĢ§å¼º
    0.09
    å¥ģ
    0.08
    çļĦ缮åħī
    0.08
    ÑĢай
    0.08
    Activations Density 0.008%

    No Known Activations