© 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. 9438
    Prev
    Next
    INDEX
    Explanations

    et al

    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
    rame
    -0.09
    åįĥä¼ı
    -0.09
    oodle
    -0.09
    绪
    -0.08
     and
    -0.08
    碴
    -0.08
    åħ¬æ°ij
    -0.08
    éĩĮæĸ¯
    -0.08
    åŁºè°ĥ
    -0.08
    æĦ§
    -0.08
    POSITIVE LOGITS
    们
    0.09
    们çļĦ
    0.09
    å¿ĥä¸ŃçļĦ
    0.09
    ETER
    0.08
    æīĭä¸ŃçļĦ
    0.08
     chÃŃ
    0.08
    -animate
    0.08
    ä¸Ģåħ±
    0.08
    ä»ĵ
    0.08
    åľ¨åĨħçļĦ
    0.08
    Activations Density 0.086%

    No Known Activations