© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-4B
    5. 19-TRANSCODER-HP
    7. 7366
    Prev
    Next
    INDEX
    Explanations

    oris

    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
    obra
    -0.29
    аÑĤоÑĢ
    -0.29
    eka
    -0.27
    oha
    -0.26
    iving
    -0.26
    onga
    -0.25
    注åħ¥
    -0.24
    .toInt
    -0.24
    ;-
    -0.24
     ever
    -0.24
    POSITIVE LOGITS
     маÑĤеÑĢиалов
    0.25
     skl
    0.24
    (TM
    0.24
    æµĬ
    0.24
    (Resources
    0.24
    _FS
    0.24
    earer
    0.23
    诰
    0.23
     pers
    0.23
     supports
    0.22
    Activations Density 0.094%

    No Known Activations