© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 21010
    Prev
    Next
    INDEX
    Explanations

    continued growth and reduction

    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
    çµ®
    -0.10
     róż
    -0.09
    庶
    -0.09
    é»ı
    -0.08
    -block
    -0.08
    åĮ¿åIJį
    -0.08
    åĵĻ
    -0.08
    ropol
    -0.08
    读书
    -0.08
    åĬ£åĬ¿
    -0.08
    POSITIVE LOGITS
    ç»ıæµİå¢ŀéķ¿
    0.10
    çļĦå¢ŀéķ¿
    0.10
    å¢ŀéķ¿
    0.09
    æģĥ
    0.09
    éŃĶæ³ķå¸Ī
    0.09
    æĭ¯æķij
    0.09
    çĬĦ
    0.09
    å¢ŀéĩı
    0.09
    èµĦ
    0.09
    åıĺäºĨ
    0.09
    Activations Density 0.082%

    No Known Activations