© 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. 10401
    Prev
    Next
    INDEX
    Explanations

    house features

    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
    Ñİ
    -0.09
    鬼
    -0.09
    าม
    -0.09
    imon
    -0.09
    饥饿
    -0.09
     конÑĤакÑĤ
    -0.08
    大楼
    -0.08
    裤åŃIJ
    -0.08
    宿èĪį
    -0.08
    交
    -0.08
    POSITIVE LOGITS
    å¥ļ
    0.09
     living
    0.09
    EDIA
    0.09
    è´¯éĢļ
    0.09
    èIJ½
    0.09
    管家
    0.08
    emin
    0.08
     lẽ
    0.08
     Living
    0.08
    åݨ
    0.08
    Activations Density 0.114%

    No Known Activations