© 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. 9452
    Prev
    Next
    INDEX
    Explanations

    fine, mind, care, with

    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
    çļĦéĩįè¦ģæĢ§
    -0.09
     Horde
    -0.09
    жи
    -0.09
    æĬĽ
    -0.09
    涵
    -0.09
    åĬłæ²¹
    -0.09
    èĻ½çĦ¶
    -0.08
    ä¾ĽéľĢ
    -0.08
    ä»Ģä¹ĪæĦıæĢĿ
    -0.08
    好象
    -0.08
    POSITIVE LOGITS
    ä½Ĩæĺ¯å¦Ĥæŀľ
    0.10
    ä»»ä½ķ
    0.10
    ä»»ä½ķå½¢å¼ı
    0.09
    ä½Ĩå¦Ĥæŀľ
    0.09
    çļĤ
    0.09
    èĩªæ²»
    0.09
    çļĦå°±æĺ¯
    0.08
    å®ī康
    0.08
    à¸ĸา
    0.08
    heed
    0.08
    Activations Density 0.041%

    No Known Activations