© 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. 97913
    Prev
    Next
    INDEX
    Explanations

    Code and names

    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
    æīĵåIJ¬
    -0.27
    ç»ijå®ļ
    -0.24
     Bare
    -0.24
    ankan
    -0.24
    顽
    -0.23
    elige
    -0.23
     Kart
    -0.23
     respir
    -0.23
    å¦Ĥä»Ĭ
    -0.23
     Durant
    -0.22
    POSITIVE LOGITS
    çĺ¾
    0.27
    pherd
    0.24
    inosaur
    0.24
    ä¸įåĪ°ä½į
    0.24
    ä¸ĬéĻIJ
    0.23
    rous
    0.23
    çĶ¨æĪ·æıIJä¾Ľ
    0.23
    esseract
    0.23
    istics
    0.23
    æĴº
    0.23
    Activations Density 0.022%

    No Known Activations