© 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. 8832
    Prev
    Next
    INDEX
    Explanations

    moving away from or shift to

    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
    çļĦåľ°ä½į
    -0.10
    èĵī
    -0.09
    è°ĥåīĤ
    -0.09
     Ref
    -0.08
    _Ref
    -0.08
    irit
    -0.08
    æ²IJæµ´
    -0.08
    çłº
    -0.08
    ior
    -0.08
    _ref
    -0.08
    POSITIVE LOGITS
    强è°ĥ
    0.12
    æijĴ
    0.11
    ä¼łç»ŁçļĦ
    0.10
    è¶ĭåĬ¿
    0.10
    addock
    0.10
    ä¹ĭ举
    0.10
    åĢ¾åIJijäºİ
    0.10
    ä¸Ńå¼ı
    0.10
     emphasis
    0.10
    离å¼Ģ
    0.10
    Activations Density 0.068%

    No Known Activations