© 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. 27062
    Prev
    Next
    INDEX
    Explanations

    speculation or possibility

    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
    好象
    -0.13
    ä¼¼ä¹İ
    -0.12
     reportedly
    -0.12
     apparently
    -0.11
    好åĥı
    -0.11
     supposedly
    -0.10
    éĢļ常
    -0.10
    éĢļ常æĺ¯
    -0.10
    çĸijä¼¼
    -0.10
    ä»¿ä½Ľ
    -0.09
    POSITIVE LOGITS
     somewhere
    0.11
    没ä»Ģä¹Ī
    0.10
    æŁIJä¸Ģ
    0.10
    ä½łä¼ļ
    0.09
    ä¹Łæ²¡ä»Ģä¹Ī
    0.09
     algún
    0.09
     пÑĢидеÑĤÑģÑı
    0.09
    æľīä¸ĢäºĽ
    0.09
    缸å½ĵ
    0.09
     займ
    0.09
    Activations Density 0.083%

    No Known Activations