© 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. 8606
    Prev
    Next
    INDEX
    Explanations

    sounds like

    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
    utan
    -0.09
    è¿Ļæł·è¯´
    -0.09
    æĭľ
    -0.09
    string
    -0.08
    ä¸Ģçľ¼
    -0.08
    è¯Ŀ说
    -0.08
    æĥĬèī³
    -0.08
    没æľīæĥ³åĪ°
    -0.08
    åı°
    -0.08
    èĭ¹
    -0.08
    POSITIVE LOGITS
    审议
    0.10
    åıijè´¢
    0.09
     irre
    0.09
    HeaderCode
    0.09
    DRAM
    0.09
    æľĢç»Ī
    0.09
    ÙĦد
    0.09
    IOD
    0.09
    èľľ
    0.08
    _PCM
    0.08
    Activations Density 0.035%

    No Known Activations