© 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. 64224
    Prev
    Next
    INDEX
    Explanations

    Technical/code-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
    .jboss
    -0.28
    _again
    -0.26
    æľĽ
    -0.26
    [L
    -0.25
    closing
    -0.25
    _PTR
    -0.25
     tô
    -0.25
    ä»İ严治
    -0.25
    know
    -0.24
    close
    -0.24
    POSITIVE LOGITS
    uhan
    0.28
     unde
    0.25
    iti
    0.25
    ije
    0.24
    ãģªãĤĵãģ¨
    0.23
    ikes
    0.23
    åIJĪæĪIJ
    0.23
    ä¼ij
    0.23
    .microsoft
    0.23
    abis
    0.23
    Activations Density 7.228%

    No Known Activations