© 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. 30908
    Prev
    Next
    INDEX
    Explanations

    refused to commit

    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
    æł¹æľ¬å°±ä¸į
    -0.09
    rap
    -0.09
    ife
    -0.08
    æľ¬æĿ¥å°±
    -0.08
    åĽĽå¤Ħ
    -0.08
    I
    -0.08
     already
    -0.08
     indeed
    -0.08
    tv
    -0.08
    ä¸įä»ħ
    -0.08
    POSITIVE LOGITS
    åį´æ²¡æľī
    0.13
    å°ļ
    0.12
    ä¹Łä¸įæķ¢
    0.11
     bergen
    0.11
    ä»įå°Ĩ
    0.10
    åľºéĿ¢
    0.10
     henüz
    0.10
    æĺİç¡®
    0.10
    强硬
    0.10
    enton
    0.10
    Activations Density 0.187%

    No Known Activations