© 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. 129866
    Prev
    Next
    INDEX
    Explanations

    parentheses

    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
     than
    -0.29
    .rawValue
    -0.26
    æĭĹ
    -0.26
    /document
    -0.24
     accompanies
    -0.24
    semantic
    -0.24
    åįł
    -0.24
    Rest
    -0.24
    ONUS
    -0.24
     Origins
    -0.24
    POSITIVE LOGITS
    éĶļ
    0.26
    æ´Ľåħĭ
    0.25
    tick
    0.25
    otify
    0.25
    ocking
    0.25
    jong
    0.24
    zar
    0.24
    itters
    0.24
    -neck
    0.24
    ByExample
    0.24
    Activations Density 14.615%

    No Known Activations