© 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. 3941
    Prev
    Next
    INDEX
    Explanations

    coenzyme or cofactor

    np_acts-logits-general · gemini-2.5-flash-lite

    biochemical compound names and molecular cofactors involved in metabolic processes.

    oai_token-act-pair · claude-4-5-haikuTriggered by @yooniel31
    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
    anten
    -0.09
    pole
    -0.09
    ä¸ĥå¹´
    -0.09
    lab
    -0.08
    -shot
    -0.08
    ags
    -0.08
    ä¸įåıĬ
    -0.08
    лÑij
    -0.08
    åĪº
    -0.08
    赶ä¸Ĭ
    -0.08
    POSITIVE LOGITS
    æİĸ
    0.10
    ä¸ĢæŶ
    0.10
    inea
    0.10
    åĬłæ²¹ç«Ļ
    0.09
    åĬ©éķ¿
    0.09
    äºĭä¸ļ
    0.09
    åIJ¬èµ·æĿ¥
    0.09
    åĬŀåħ¬
    0.09
    è·ij
    0.09
    åľ¨äº¬
    0.09
    Activations Density 0.197%

    No Known Activations