© 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. 2550
    Prev
    Next
    INDEX
    Explanations

    right or wrong

    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
    ç»Ļå®Ŀå®Ŀ
    -0.09
    èĤī
    -0.09
    æĸ°ä¸ĸ纪
    -0.09
    anded
    -0.09
    æĺİæľĹ
    -0.08
    åıªè§ģ
    -0.08
    .hist
    -0.08
    inating
    -0.08
    羣æŃ£
    -0.08
    çĻ»ä¸Ĭ
    -0.08
    POSITIVE LOGITS
    åĪ°å¤Ħ
    0.10
     about
    0.09
    thic
    0.09
    ç»ıèIJ¥çIJĨ念
    0.09
    å¤Ħå¤Ħ
    0.09
    åIJĦè¡ĮåIJĦä¸ļ
    0.08
    ofi
    0.08
    duct
    0.08
     consequential
    0.08
    æ²®
    0.08
    Activations Density 0.012%

    No Known Activations