© 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. 22-TRANSCODER-HP
    7. 88414
    Prev
    Next
    INDEX
    Explanations

    patents and ISBNs

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_22.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.22.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    inz
    -0.31
     tast
    -0.25
    inis
    -0.25
    ä¾Ľæ±Ĥ
    -0.25
    å·¥ä½ľä¼ļè®®
    -0.25
    .effects
    -0.25
    æĿ¥çľĭ
    -0.24
    elper
    -0.24
    UGE
    -0.24
    _exceptions
    -0.24
    POSITIVE LOGITS
     tow
    0.28
     Beyond
    0.26
    tern
    0.25
    æĽĿ
    0.25
     geme
    0.24
    ition
    0.24
     Lamar
    0.24
    hift
    0.23
    igte
    0.23
     conver
    0.23
    Activations Density 0.156%

    No Known Activations