© 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. 64074
    Prev
    Next
    INDEX
    Explanations

    code/web related text

    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
    who
    -0.27
    ä¸įåIJĪ
    -0.27
    éĩĩ
    -0.27
    éĵ¢
    -0.26
    å¼ŁåŃIJ
    -0.25
     wished
    -0.24
    沿éĢĶ
    -0.24
     mun
    -0.24
     kisses
    -0.23
    sen
    -0.23
    POSITIVE LOGITS
    ewis
    0.29
    ég
    0.28
    awe
    0.28
    izing
    0.25
    erset
    0.25
    èĨĺ
    0.25
    è¿ĽäºĨ
    0.25
    ichert
    0.24
    rawer
    0.24
    etak
    0.24
    Activations Density 2.544%

    No Known Activations