© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-4B
    5. 20-TRANSCODER-HP
    7. 63850
    Prev
    Next
    INDEX
    Explanations

    that

    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
    Not in Any Lists

    No Comments

    Negative Logits
    holm
    -0.24
     parsed
    -0.24
    /topics
    -0.23
     reclaim
    -0.23
    REET
    -0.23
    è¯ķçĤ¹
    -0.23
    kul
    -0.23
    åħ¥å¸Ĥ
    -0.23
     subtitle
    -0.23
     reversal
    -0.23
    POSITIVE LOGITS
    ëijĺ
    0.28
    tracts
    0.27
    ä»ĸ们ä¼ļ
    0.25
    _fg
    0.25
    ùng
    0.24
    olah
    0.24
    èIJ¥éĢłèī¯å¥½
    0.24
    ------------↵
    0.24
    æł¹æľ¬æ²¡æľī
    0.24
    éĥ½ä¸įæĺ¯
    0.24
    Activations Density 1.936%

    No Known Activations