© 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. 63232
    Prev
    Next
    INDEX
    Explanations

    afraid of, woos, perfect, business

    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
    ä¸ĢæŶ
    -0.10
     Bean
    -0.09
    å½ĵæŶ
    -0.08
     spe
    -0.08
    éģ¿
    -0.08
    Bean
    -0.08
     dispar
    -0.08
     particular
    -0.08
    ä¹į
    -0.08
    æݢ索
    -0.08
    POSITIVE LOGITS
    èĩŃ
    0.12
    亥
    0.11
    åѤç«ĭ
    0.10
    æ°ijæĦı
    0.10
    ä¿¡èªī
    0.10
    è§īæĤŁ
    0.10
    æİ·
    0.09
    æ°ijå¿ĥ
    0.09
    çļĪ
    0.09
    群ä¼Ĺ
    0.09
    Activations Density 0.095%

    No Known Activations