© 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. 20973
    Prev
    Next
    INDEX
    Explanations

    until followed by descriptive words

    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
    CallCheck
    -0.10
    æıĨ
    -0.10
    æĢ»æĺ¯
    -0.09
    natal
    -0.09
    orts
    -0.09
    總æĺ¯
    -0.09
    次è¦ģ
    -0.09
     вÑģегда
    -0.09
    åºĶåıĬæŶ
    -0.09
     Oval
    -0.09
    POSITIVE LOGITS
     either
    0.10
    ä¸Ģ人
    0.09
    çĴľ
    0.09
    çIJĨ解åĴĮ
    0.09
    Dub
    0.09
    åĸľ
    0.08
    çİ°åĩº
    0.08
    产çĶŁ
    0.08
    åĪĨè¡Į
    0.08
    usty
    0.08
    Activations Density 0.023%

    No Known Activations