© 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. 141706
    Prev
    Next
    INDEX
    Explanations

    Garbled or incomplete sentences

    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
    while
    -0.27
    [Z
    -0.27
     Train
    -0.27
    æľĢå¤ļçļĦ
    -0.26
    [first
    -0.25
    ORIGINAL
    -0.25
    [..
    -0.25
    ä¸ĩåįĥ
    -0.24
    [dir
    -0.24
    çģ«è½¦
    -0.24
    POSITIVE LOGITS
    WithOptions
    0.32
    éĢĤæŶ
    0.28
     static
    0.28
    -static
    0.28
    /pkg
    0.26
    éĿĻæĢģ
    0.25
    static
    0.25
    åİ»åĵªåĦ¿
    0.25
    pkg
    0.24
    说
    0.24
    Activations Density 8.050%

    No Known Activations