© 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. 63851
    Prev
    Next
    INDEX
    Explanations

    Start of text/code blocks

    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
     shops
    -0.27
    gone
    -0.26
    åIJ¼
    -0.26
    被æīĵ
    -0.25
    å¼ĢåºĹ
    -0.25
    tracking
    -0.25
    OLF
    -0.25
    åķĨæĬ¥
    -0.25
    LEAR
    -0.24
    fern
    -0.24
    POSITIVE LOGITS
    æµĨ
    0.28
    éĢĤ
    0.26
    ava
    0.26
    å¤Ħå¤Ħéķ¿
    0.26
    nia
    0.26
     Platform
    0.26
     tobacco
    0.26
     NIC
    0.25
    /inet
    0.25
    Platform
    0.24
    Activations Density 53.933%

    No Known Activations