© 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. 51902
    Prev
    Next
    INDEX
    Explanations

    lists items with descriptive phrases

    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
    å¥ļ
    -0.09
     le
    -0.08
    UCT
    -0.08
     Kash
    -0.08
     gra
    -0.08
    STA
    -0.08
     Meyer
    -0.07
     ØŃÙĪ
    -0.07
    ãģłãģ£ãģŁãĤī
    -0.07
     UPC
    -0.07
    POSITIVE LOGITS
    èĥĮåIJİ
    0.09
    ä¸Ģç³»åĪĹ
    0.09
    аÑĪ
    0.09
    çı°
    0.09
    ç´Ĭ
    0.09
    irectional
    0.09
    dni
    0.08
    SAM
    0.08
    loff
    0.08
    èĢIJ磨
    0.08
    Activations Density 0.688%

    No Known Activations