© 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. 8374
    Prev
    Next
    INDEX
    Explanations

    nearly, almost

    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
    该æĢİä¹Ī
    -0.10
    æıĨ
    -0.10
    家ä½ı
    -0.09
     Vox
    -0.09
    æļĤæŶ
    -0.09
    æľĢå¿«
    -0.08
    erah
    -0.08
    finder
    -0.08
    佬
    -0.08
    æļĤ
    -0.08
    POSITIVE LOGITS
    è®®
    0.10
    æ³®
    0.10
    åĬŁå¤«
    0.10
    峤
    0.10
    haul
    0.09
    ä¸Ńå°ıåŀĭ
    0.09
    ernal
    0.09
    ç¬ijè¯Ŀ
    0.09
    é£İéĩĩ
    0.09
    ä¿Ŀ温
    0.09
    Activations Density 0.020%

    No Known Activations