© 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. 35695
    Prev
    Next
    INDEX
    Explanations

    mitigation and reducing impact

    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
    ularity
    -0.11
     Pruitt
    -0.10
    ä½ŀ
    -0.10
    ihil
    -0.09
    .PERMISSION
    -0.09
    hydro
    -0.09
    Äĥng
    -0.09
    icker
    -0.09
    ANGER
    -0.09
    iness
    -0.09
    POSITIVE LOGITS
    æݪæĸ½
    0.11
    æİī
    0.10
    åĪ©äºļ
    0.09
    /mit
    0.09
     Lag
    0.08
    人éĢł
    0.08
    潮湿
    0.08
    æİīäºĨ
    0.08
    erchant
    0.07
    å¹´ä¹ĭ
    0.07
    Activations Density 0.017%

    No Known Activations