© 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. 27583
    Prev
    Next
    INDEX
    Explanations

    roofs of buildings

    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
    ÑĤÑĢен
    -0.10
    æ°¤
    -0.10
    Injected
    -0.10
    åĵī
    -0.10
    æĻ¯
    -0.09
    ä¸įçĶ±
    -0.09
    inne
    -0.09
    æ±²
    -0.09
    çļĦçľ¼ç¥ŀ
    -0.09
    åłĩ
    -0.09
    POSITIVE LOGITS
    æ¯ĶåĪ©
    0.09
     bowling
    0.09
    æĿ¡ä¾ĭ
    0.09
    å±ĭ
    0.09
    cht
    0.08
    å±İ
    0.08
    æĿ¿æĿIJ
    0.08
    秣
    0.08
    èĩªæĿ¥
    0.08
    éķ¿éĢĶ
    0.08
    Activations Density 0.129%

    No Known Activations