© 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. 9180
    Prev
    Next
    INDEX
    Explanations

    scientific validation and reproducibility

    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
     dem
    -0.29
    ignet
    -0.27
     <",
    -0.26
    ocos
    -0.26
    dem
    -0.26
    olini
    -0.25
    ä¼ī
    -0.25
    æķĻ
    -0.24
     merits
    -0.24
    .Arguments
    -0.24
    POSITIVE LOGITS
     Needed
    0.27
     frozen
    0.27
     scrambled
    0.26
    ä¸Ģ声
    0.26
     altern
    0.25
    ISODE
    0.24
     greatest
    0.24
     Miscellaneous
    0.23
     public
    0.23
     preserved
    0.23
    Activations Density 0.016%

    No Known Activations