© 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. 61323
    Prev
    Next
    INDEX
    Explanations

    spatial relation

    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
    纷纷
    -0.10
    åIJĦåľ°
    -0.10
    ä¹Łå¥½
    -0.10
    ä¹ĭæĦı
    -0.09
    asis
    -0.09
    åģ·
    -0.09
    åŃ
    -0.08
    åĨį次
    -0.08
    à¸Ļà¸Ħร
    -0.08
    æľºèĥ½
    -0.08
    POSITIVE LOGITS
    same
    0.11
    éĿłè¿ij
    0.11
    è·Ŀ离
    0.10
    æİ¥è¿ij
    0.10
     proximity
    0.10
     nearby
    0.09
    缴æݥ影åĵį
    0.09
    缸éĤ»
    0.09
    è·ĿéĽ¢
    0.09
    缸è¿ij
    0.09
    Activations Density 0.068%

    No Known Activations