© 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. 37437
    Prev
    Next
    INDEX
    Explanations

    curb

    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
    课æĸĩ
    -0.10
    éĢĥéģ¿
    -0.10
    åĽ¢éķ¿
    -0.09
    åĽ¤
    -0.09
    æīĭç»Ńè´¹
    -0.09
    åĪĽæĸ°åŀĭ
    -0.09
    ress
    -0.09
    é»ijæļĹ
    -0.09
    太平
    -0.08
    çļĻ
    -0.08
    POSITIVE LOGITS
    ToWorld
    0.09
    ä¸ĬçļĦ
    0.09
    ä¸Ĭæľī
    0.09
    ä¸ĬäºĨ
    0.09
    LESS
    0.08
    icz
    0.08
    prises
    0.08
    åı¦ä¸Ģè¾¹
    0.08
    å¢Ģ
    0.08
    loads
    0.08
    Activations Density 0.002%

    No Known Activations