© 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. 129919
    Prev
    Next
    INDEX
    Explanations

    Code and identifiers

    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
    oyal
    -0.27
    appointed
    -0.26
    çļĭ
    -0.26
    wap
    -0.25
    shal
    -0.24
     Ralph
    -0.24
     Cos
    -0.24
    olini
    -0.23
    .ma
    -0.23
     xu
    -0.23
    POSITIVE LOGITS
    ctic
    0.32
    ej
    0.27
    ighth
    0.26
    豺
    0.26
    phe
    0.25
    Canceled
    0.24
    encias
    0.24
    anian
    0.24
     billion
    0.23
    zÄĻ
    0.23
    Activations Density 2.496%

    No Known Activations