© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 19413
    Prev
    Next
    INDEX
    Explanations

    spite of

    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
    èħ¥
    -0.10
    缨
    -0.10
    éĢĤåIJĪèĩªå·±
    -0.09
    ä¸ĭæīĭ
    -0.09
    èĪĨ论
    -0.09
    æĪijå¦Ī
    -0.09
    hev
    -0.08
    èħĬ
    -0.08
    西红
    -0.08
    å·´æĸ¯
    -0.08
    POSITIVE LOGITS
    ä¸ĩåĪĨ
    0.10
    çľĭçĤ¹
    0.10
    å°Ĩ继ç»Ń
    0.09
    åĤ¨èĵĦ
    0.09
    endra
    0.09
    å°Ĩ以
    0.09
    еÑĩа
    0.08
    éħį
    0.08
    okin
    0.08
    åĵŃ
    0.08
    Activations Density 0.001%

    No Known Activations