© 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. 33105
    Prev
    Next
    INDEX
    Explanations

    references and links

    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
    æĿijå§Ķä¼ļ
    -0.09
     pil
    -0.08
     <![
    -0.08
    æĪijæĬĬ
    -0.08
    åīįéĢĶ
    -0.08
    spir
    -0.08
    èĵĦçĶµæ±ł
    -0.08
    çĮ¿
    -0.08
    å½ĵçĦ¶æĺ¯
    -0.08
    _IMPL
    -0.08
    POSITIVE LOGITS
    OTES
    0.10
    ä¸įçĶĺ
    0.10
    Ñıд
    0.09
    ç»Ĩå¾®
    0.09
    à¸ĺร
    0.09
    otes
    0.09
    ectl
    0.09
    ibs
    0.09
    ä¸įæĪIJ
    0.08
    Lua
    0.08
    Activations Density 1.019%

    No Known Activations