© 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. 63236
    Prev
    Next
    INDEX
    Explanations

    at the beginning 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
    leck
    -0.09
    Reviewed
    -0.09
    olid
    -0.09
    positories
    -0.09
    resents
    -0.09
    顶
    -0.08
    è®
    -0.08
    éķ¿çŁŃ
    -0.08
    occo
    -0.08
    ож
    -0.08
    POSITIVE LOGITS
    ä¸Ģ个æĸ°çļĦ
    0.11
    nings
    0.11
    æĸ°çļĦä¸Ģ
    0.10
    ãģ¯ãģĺ
    0.10
    å¸ĪçĶŁ
    0.10
    çĿĢæīĭ
    0.09
    æĸ°ä¸Ģè½®
    0.09
    å´Ń
    0.09
    æݥ触
    0.09
     adulthood
    0.09
    Activations Density 0.039%

    No Known Activations