© 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. 11822
    Prev
    Next
    INDEX
    Explanations

    binding actions

    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
    çĦ¶æĺ¯
    -0.10
    rans
    -0.10
    tery
    -0.09
    itative
    -0.09
    ened
    -0.09
    ä½ĵç³»
    -0.09
    arium
    -0.09
    è¿Ļä¸ĢçĤ¹
    -0.09
    æľĽåİ»
    -0.08
     Santana
    -0.08
    POSITIVE LOGITS
    icut
    0.11
    ä½ıäºĨ
    0.10
     lá»Ĺi
    0.10
    ç¼ļ
    0.09
    maids
    0.09
    enu
    0.09
    åľ¨ä¸Ģèµ·
    0.09
    emas
    0.09
    ding
    0.09
    anni
    0.09
    Activations Density 0.061%

    No Known Activations