© 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. 20-TRANSCODER-HP
    7. 107717
    Prev
    Next
    INDEX
    Explanations

    Pis

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_20.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.20.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    麼
    -0.27
    AFF
    -0.27
    ammer
    -0.26
    å±Ĥåĩº
    -0.26
     strongly
    -0.26
    åķ¦
    -0.26
    à¹Ĩ
    -0.25
     lattice
    -0.24
    ëĿ¼
    -0.24
    haven
    -0.24
    POSITIVE LOGITS
    /sources
    0.30
    åĬ¨äºº
    0.28
    elib
    0.27
     closets
    0.26
     tackled
    0.24
    ä¹Łæĺ¯å¦ĤæѤ
    0.24
    IDL
    0.24
    _sr
    0.24
    subpackage
    0.23
    æIJŃä¹ĺ
    0.23
    Activations Density 8.902%

    No Known Activations