© 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. 4984
    Prev
    Next
    INDEX
    Explanations

    steam, heating, cooking, flavor

    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
     matchmaking
    -0.10
    .TXT
    -0.10
    详
    -0.10
    _FC
    -0.09
    enas
    -0.08
    èĪįä¸įå¾Ĺ
    -0.08
    误导
    -0.08
    icted
    -0.08
    éļ¾å¾Ĺ
    -0.08
     Powder
    -0.08
    POSITIVE LOGITS
    å¨ĥå¨ĥ
    0.12
    åĩ«
    0.12
    åħ¬ç«ĭ
    0.11
    åĹ·
    0.11
    steam
    0.10
    çĥĿ
    0.09
     steam
    0.09
    åįļæĸĩ
    0.09
    éľİ
    0.09
    è®´
    0.09
    Activations Density 0.034%

    No Known Activations