© 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. 19-TRANSCODER-HP
    7. 8539
    Prev
    Next
    INDEX
    Explanations

    Code and database entries

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_19.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.19.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     likewise
    -0.29
     Likewise
    -0.28
    ewise
    -0.27
    ijo
    -0.26
     Rel
    -0.26
    pute
    -0.25
    åIJĮæł·æĺ¯
    -0.25
    scal
    -0.24
    Ex
    -0.24
    综åIJĪç´łè´¨
    -0.24
    POSITIVE LOGITS
     campuses
    0.26
     Attached
    0.25
    bourne
    0.24
     attach
    0.24
    rophe
    0.24
     Attach
    0.24
    好èݱåĿŀ
    0.23
    ç§ĭåĨ¬
    0.23
    课ç¨ĭ
    0.23
     attached
    0.23
    Activations Density 0.405%

    No Known Activations