© 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. 30-TRANSCODER-HP
    7. 96549
    Prev
    Next
    INDEX
    Explanations

    0

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_30.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.30.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     wing
    -0.29
    onte
    -0.28
    èµĦæºIJåħ±äº«
    -0.28
     nim
    -0.27
     nø
    -0.25
    á»ķi
    -0.25
     wings
    -0.25
    -hand
    -0.24
    atat
    -0.24
    ãĥĩãĥ³
    -0.24
    POSITIVE LOGITS
     dagger
    0.28
    á¿Ĩ
    0.25
    è¡ĵ
    0.24
    ä¸Ĭ游
    0.24
     short
    0.24
    ubs
    0.23
    ä¸ĢéĿ¢
    0.23
    éĥ¨éķ¿
    0.23
     det
    0.23
    morgan
    0.23
    Activations Density 0.784%

    No Known Activations