© 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. 31489
    Prev
    Next
    INDEX
    Explanations

    index, performance, table variable, license

    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
    ä¸ĢéĹª
    -0.10
    ä»¥äººä¸ºæľ¬
    -0.10
    æĹłæĦı
    -0.10
    æĻ¦
    -0.10
    ä¹ŁæŃ£æĺ¯
    -0.09
     Attribution
    -0.09
    å¿ħä¸įåı¯
    -0.09
    éŃĶçİĭ
    -0.09
    è¿ĩ渡
    -0.09
    isor
    -0.09
    POSITIVE LOGITS
    æĹı
    0.09
    _performance
    0.08
    ç¦ģ
    0.08
    à¹Īà¸Ńà¸ĩ
    0.08
     onData
    0.08
    ragment
    0.07
    olvers
    0.07
     движениÑı
    0.07
    ayer
    0.07
    agged
    0.07
    Activations Density 0.134%

    No Known Activations