© 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. 27134
    Prev
    Next
    INDEX
    Explanations

    detective Kim Stone, Corps, or private

    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
    ings
    -0.10
    nings
    -0.09
    lessly
    -0.09
    å¾Ĺèµ·
    -0.09
    åĦ¿
    -0.09
    绫
    -0.09
    nda
    -0.09
    ร
    -0.08
    etto
    -0.08
    çµ
    -0.08
    POSITIVE LOGITS
    奥æŀĹ
    0.09
    ëĭĺ
    0.09
    æݾ
    0.09
    gram
    0.09
    anton
    0.09
    天é¹ħ
    0.09
    åľ¨è¿ĩ渡
    0.09
    å¢ħ
    0.09
    æīĭå¥Ĺ
    0.09
    angelo
    0.09
    Activations Density 0.005%

    No Known Activations