© 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. 39016
    Prev
    Next
    INDEX
    Explanations

    plus sign separation

    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
    crete
    -0.10
    åł¤
    -0.08
    è¿Ļæł·è¯´
    -0.08
    çŃ·
    -0.08
    éĹªçĥģ
    -0.08
    found
    -0.08
    å²Į
    -0.08
    éŵ
    -0.08
    reen
    -0.08
    åij¼
    -0.08
    POSITIVE LOGITS
    主ä¹īæĢĿæĥ³
    0.09
    ะ
    0.09
    äºĨ大éĩıçļĦ
    0.08
    eer
    0.08
    anium
    0.08
    絡
    0.08
    Dean
    0.08
     Commission
    0.08
    æīĪ
    0.08
    ãĥ¼ãĤ·ãĥ§ãĥ³
    0.08
    Activations Density 0.050%

    No Known Activations