© 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. 50241
    Prev
    Next
    INDEX
    Explanations

    Oops!

    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
    ä½Ĩä¸įéĻIJäºİ
    -0.09
    èĶij
    -0.09
    usta
    -0.08
    积
    -0.08
    DBG
    -0.08
    ILLS
    -0.08
    iero
    -0.08
    .arc
    -0.08
     ultimately
    -0.08
    足
    -0.08
    POSITIVE LOGITS
    æľī人说
    0.11
    åIJįä¹ī
    0.10
    æİī
    0.10
    æĢİä¹ĪåĬŀ
    0.10
    åıijçİ°äºĨ
    0.09
     bitten
    0.09
    å¸ļ
    0.09
    .asc
    0.09
    åĬłåħ¥äºĨ
    0.09
    转身
    0.08
    Activations Density 0.013%

    No Known Activations