© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 2102
    Prev
    Next
    INDEX
    Explanations

    different every uniquely

    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
    å¼ķèµ·
    -0.10
     diver
    -0.09
    евиÑĩ
    -0.08
    etta
    -0.08
    дол
    -0.08
    é«ĺ度éĩįè§Ĩ
    -0.08
    å¼ķèµ·äºĨ
    -0.08
    etting
    -0.08
    clc
    -0.08
    ä¸įçĶ±
    -0.08
    POSITIVE LOGITS
    å®ŀæķĪ
    0.10
    ä¸įå°½
    0.10
    éļĶ离
    0.09
    çĭ¬ä¸ĢæĹłäºĮ
    0.09
    ç»ıèIJ¥ç®¡çIJĨ
    0.09
    æŃ£æĸĩ
    0.09
    inte
    0.09
    çĹī
    0.09
    趱
    0.09
    ErrorException
    0.09
    Activations Density 0.044%

    No Known Activations