© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 36127
    Prev
    Next
    INDEX
    Explanations

    tied, tucked, up, bow

    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
    æĬ«
    -0.08
    aver
    -0.08
    éķij
    -0.08
     Sud
    -0.08
    sted
    -0.08
    éĿ¢æĿ¡
    -0.08
    å¹³åİŁ
    -0.08
    vell
    -0.08
    ly
    -0.08
     Sudoku
    -0.08
    POSITIVE LOGITS
    çĿ£
    0.11
    éĨļ
    0.10
    åĪ©äºİ
    0.09
    å§ĭäºİ
    0.09
    ä¼ļåIJĮ
    0.09
    æĥľ
    0.09
     Osw
    0.09
    ä¹ĭäºī
    0.08
    éĿ´
    0.08
    /ts
    0.08
    Activations Density 0.003%

    No Known Activations