© 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. 18298
    Prev
    Next
    INDEX
    Explanations

    out of

    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ä¸į
    -0.09
     completionHandler
    -0.08
    å°±æĺ¯è¦ģ
    -0.08
     powerhouse
    -0.08
    ä¸ĢæŃ¥
    -0.08
    åĩºèĩª
    -0.08
    Ñģад
    -0.08
    ÑĢей
    -0.08
     concession
    -0.08
    uko
    -0.08
    POSITIVE LOGITS
    ä½ľä¸ļ
    0.09
     affairs
    0.08
    opers
    0.08
    ä½Ĩä¹Ł
    0.08
    oder
    0.08
    ê±´
    0.08
    èįĢ
    0.08
    hti
    0.08
    ä½Ĩä»İ
    0.08
    å¤
    0.08
    Activations Density 0.013%

    No Known Activations