© 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. 39594
    Prev
    Next
    INDEX
    Explanations

    device warm up/down

    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
    ળ
    -0.09
     tote
    -0.08
    ego
    -0.08
    æľ¬æľŁ
    -0.08
    ëĵł
    -0.08
     Tes
    -0.08
     возможно
    -0.08
    гла
    -0.08
    arth
    -0.08
    _ISO
    -0.08
    POSITIVE LOGITS
    主è¦ģé¢Ĩ导
    0.10
    é¢Ĩ导çıŃåŃIJ
    0.09
    Collections
    0.09
    åIJįçīĮ
    0.09
    æ¸ł
    0.09
    åļ
    0.09
    åĬŁ
    0.09
    ç¨Ķ
    0.08
     Chow
    0.08
    åĪ¶å®ļ
    0.08
    Activations Density 0.035%

    No Known Activations