© 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. 22-TRANSCODER-HP
    7. 131239
    Prev
    Next
    INDEX
    Explanations

    universities

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_22.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.22.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Negative Logits
    sep
    -0.28
    顶端
    -0.26
    ä¸ĩåħĥ以ä¸Ĭ
    -0.25
    çij¾
    -0.24
     sep
    -0.24
    BaseContext
    -0.24
    åĽ¢
    -0.24
     convo
    -0.24
    iges
    -0.23
    åĽĽåij¨
    -0.23
    POSITIVE LOGITS
    èŃĺ
    0.25
     Matter
    0.24
    ä¸Ķ
    0.24
    iliated
    0.23
     traded
    0.23
    /book
    0.23
    ferences
    0.23
     Animated
    0.23
     Classified
    0.23
    rames
    0.23
    Activations Density 0.006%

    No Known Activations