© 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. 37249
    Prev
    Next
    INDEX
    Explanations

    traditional

    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
    à¹ģà¸ĭ
    -0.09
    è·³è·ĥ
    -0.08
    éľ¹
    -0.08
    åľ¨åĽ½å¤ĸ
    -0.08
    ä¸įçĶ¨
    -0.08
    MBED
    -0.08
    åľºåIJĪ
    -0.08
    第ä¸Ģå±Ĭ
    -0.08
    åľ¨é¦Ļ港
    -0.08
    ä»ij
    -0.08
    POSITIVE LOGITS
    çļĦä¼łç»Ł
    0.18
    ä¼łç»ŁçļĦ
    0.16
    ä¼łç»Ł
    0.15
     traditional
    0.13
    åĽºå®ļçļĦ
    0.12
    traditional
    0.12
     traditionally
    0.12
    åĤ³çµ±
    0.12
    fixed
    0.11
    Traditional
    0.11
    Activations Density 0.311%

    No Known Activations