© 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. 63647
    Prev
    Next
    INDEX
    Explanations

    Different languages

    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
    ä¸įæĺ¯ä¸Ģ个
    -0.28
    æĺ¯éŁ©åĽ½
    -0.28
    æľĢ好ä¸įè¦ģ
    -0.25
    ä¸įåıªæĺ¯
    -0.25
    .fi
    -0.24
    Äĩi
    -0.24
    Disposable
    -0.23
    ķĮ
    -0.23
    æĸ¤
    -0.23
    ä¸į足以
    -0.23
    POSITIVE LOGITS
    ispers
    0.28
    heits
    0.26
     (*)
    0.26
    Ĥ¬
    0.25
     pragma
    0.25
    ymbol
    0.25
    åĮºåŁŁæĢ§
    0.24
    icable
    0.24
    nage
    0.24
    çĿ£
    0.24
    Activations Density 14.936%

    No Known Activations