© 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. 50907
    Prev
    Next
    INDEX
    Explanations

    breathing room and space

    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
    æİ¥åĬĽ
    -0.11
    æİ¥åħ¥
    -0.10
    à°¡
    -0.10
    éģĹ
    -0.10
    渡
    -0.10
    æ¼Ķ
    -0.09
    éģĹåĿĢ
    -0.09
    èģĮ
    -0.09
    åŁ¹è®Ń
    -0.09
    åŀĴ
    -0.09
    POSITIVE LOGITS
    空éĹ´
    0.17
     space
    0.15
    çļĦ空éĹ´
    0.15
    空éĸĵ
    0.15
    åıijå±ķ空éĹ´
    0.13
     breathing
    0.13
    space
    0.11
    ãĤ¹ãĥļãĥ¼ãĤ¹
    0.11
    åij¼åIJ¸
    0.11
     spaces
    0.11
    Activations Density 0.025%

    No Known Activations