© 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. 8785
    Prev
    Next
    INDEX
    Explanations

    experiment or study context

    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
    毫æĹł
    -0.10
    _rng
    -0.10
    ä¹IJè§Ĥ
    -0.09
    ç½Ķ
    -0.09
    æģ¤
    -0.09
     Phil
    -0.08
    éĻ²
    -0.08
    å¡ŀ
    -0.08
    (TIM
    -0.08
    çĥ¨
    -0.08
    POSITIVE LOGITS
    å°ijçĪ·
    0.10
    åľ¨çİ°åľº
    0.10
    çĥ¦æģ¼
    0.09
     extingu
    0.09
    çĿ£å¯¼
    0.09
    é¢ijçİĩ
    0.09
    ullo
    0.09
    èµĮ
    0.09
    èĢĮåİ»
    0.08
    åĽŀåIJĪ
    0.08
    Activations Density 0.302%

    No Known Activations