© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-4B
    5. 19-TRANSCODER-HP
    7. 5447
    Prev
    Next
    INDEX
    Explanations

    Context, setup, or providing information

    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
     excuse
    -0.28
    åIJ¹
    -0.27
    SB
    -0.27
    okit
    -0.26
    chio
    -0.26
    ä¿ĿèįIJ
    -0.26
     buoy
    -0.26
    滨海
    -0.26
     Blazers
    -0.25
     coinc
    -0.25
    POSITIVE LOGITS
     disposable
    0.28
    .listen
    0.27
    ³
    0.27
    åĽ½æľīèµĦ产
    0.26
    ileged
    0.26
    iero
    0.26
    å°ĨåĨĽ
    0.25
    åĭ¢
    0.25
     neg
    0.24
    沧æ¡ij
    0.24
    Activations Density 0.016%

    No Known Activations