© 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. 108337
    Prev
    Next
    INDEX
    Explanations

    Historical events

    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
    榻
    -0.34
    建设åĴĮ
    -0.28
     span
    -0.27
    modelName
    -0.26
    ence
    -0.25
     spans
    -0.25
    optic
    -0.25
    æįĮ
    -0.24
    span
    -0.24
    èµ·åºĬ
    -0.24
    POSITIVE LOGITS
    )<<
    0.31
    .Replace
    0.28
    第äºĮç§į
    0.26
    berman
    0.25
    ä¸įæĪIJ
    0.25
    oni
    0.24
    åľ¨è¿Ļç§į
    0.24
    笫
    0.24
    |R
    0.24
    ide
    0.24
    Activations Density 18.497%

    No Known Activations