© 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. 8156
    Prev
    Next
    INDEX
    Explanations

    unsupported browser download

    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
    çļĦçĭ¬çī¹
    -0.10
    寤
    -0.09
    è¯Ħå®ļ
    -0.09
    fo
    -0.09
    Mixin
    -0.09
    åIJ¸å¼ķæĽ´å¤ļ
    -0.08
    鼶è·Ŀ离
    -0.08
    çĭ¬ä¸ĢæĹłäºĮ
    -0.08
    å·²ç»ıè¾¾åĪ°
    -0.08
    æĿ
    -0.08
    POSITIVE LOGITS
    æµıè§ĪåĻ¨
    0.09
     widely
    0.08
    UDGE
    0.08
    è´»
    0.08
    agers
    0.08
    广æĴŃçĶµè§Ĩ
    0.08
    elps
    0.08
    ç§ijæķĻ
    0.08
    browser
    0.08
    å¸Ńåį·
    0.08
    Activations Density 0.041%

    No Known Activations