© 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. 8802
    Prev
    Next
    INDEX
    Explanations

    Last.fm LINE ASCAP Medline

    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
    ìĪĻ
    -0.09
    èIJ
    -0.09
    ä¸ĢçĻ¾å¤ļ
    -0.09
    éĩįå·¥
    -0.09
     Matte
    -0.09
     mist
    -0.08
    æ³ħ
    -0.08
    rates
    -0.08
    à¸ļา
    -0.08
    ä¸ĭäºĨ
    -0.08
    POSITIVE LOGITS
    çIJ¶
    0.10
    é³ı
    0.08
    acho
    0.08
    /cpp
    0.08
    è¿Ļéĥ¨åĪĨ
    0.08
    åIJ¬
    0.08
    PerPixel
    0.08
    ORIA
    0.08
     PUT
    0.08
     monet
    0.08
    Activations Density 0.002%

    No Known Activations