© 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. 51384
    Prev
    Next
    INDEX
    Explanations

    names ending in 'itz', 'osh', '[uncommon characters]', 'ali'

    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
    çıī
    -0.09
    æĬ¢
    -0.09
     chief
    -0.09
     Fus
    -0.08
    utters
    -0.08
    天èµĭ
    -0.08
    cape
    -0.08
    amic
    -0.08
    çĦī
    -0.08
    гÑĢам
    -0.08
    POSITIVE LOGITS
    æIJĵ
    0.10
    owitz
    0.10
    椽
    0.10
    uan
    0.10
    Sink
    0.09
    é½
    0.09
    åºŁæŧ
    0.09
    æĢ
    0.09
    éķ¿æľŁä»¥æĿ¥
    0.09
    anton
    0.09
    Activations Density 0.017%

    No Known Activations