© 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. 19-TRANSCODER-HP
    7. 8091
    Prev
    Next
    INDEX
    Explanations

    code/non-natural language

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_19.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.19.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     бил
    -0.27
    è¿Ļä¸Ģå¹´
    -0.25
    ery
    -0.25
    hen
    -0.25
    alu
    -0.24
    è¯¥æł¡
    -0.24
    MES
    -0.23
     epit
    -0.23
    çºł
    -0.23
     Mel
    -0.23
    POSITIVE LOGITS
    figures
    0.26
    lesc
    0.26
    ctions
    0.26
    èĨ¦
    0.25
    .chapter
    0.25
    ä½ľèĢħæľ¬äºº
    0.25
    éĶ¡
    0.25
    cliente
    0.24
     Schro
    0.24
    ction
    0.24
    Activations Density 0.003%

    No Known Activations