© 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. 50735
    Prev
    Next
    INDEX
    Explanations

    flush operations

    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
    aceous
    -0.10
    ookie
    -0.09
    岸
    -0.09
    äºİæѤ
    -0.09
    rega
    -0.09
    acle
    -0.08
    thers
    -0.08
    think
    -0.08
    äºĽè®¸
    -0.08
    hood
    -0.08
    POSITIVE LOGITS
    inator
    0.10
    tá
    0.10
    æķ´ä¸ªäºº
    0.10
    ergy
    0.10
    buster
    0.10
    aders
    0.10
    engers
    0.09
    ÑĤÑĢи
    0.09
    à¸Ľà¸£à¸°
    0.09
    arez
    0.09
    Activations Density 0.009%

    No Known Activations