© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-4B
    5. 20-TRANSCODER-HP
    7. 129147
    Prev
    Next
    INDEX
    Explanations

    privacy and confidential information

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_20.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.20.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Negative Logits
    isque
    -0.28
    .newBuilder
    -0.28
    ìĬ¨
    -0.25
    æĸľ
    -0.24
    clin
    -0.24
    stands
    -0.24
    Į¨
    -0.24
    代åĬŀ
    -0.23
    失败
    -0.23
     crus
    -0.23
    POSITIVE LOGITS
    ä¿Ŀå¯Ĩ
    0.59
     Confidential
    0.50
    ç§ĺå¯Ĩ
    0.49
     confidential
    0.48
     confidentiality
    0.47
    éļIJç§ģ
    0.46
     secrecy
    0.44
     secret
    0.44
     secrets
    0.43
     privacy
    0.40
    Activations Density 38.977%

    No Known Activations