© 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. 63933
    Prev
    Next
    INDEX
    Explanations

    Code and technical information

    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
     tender
    -0.28
    çIJ
    -0.28
     tend
    -0.27
    欧ç¾İ
    -0.25
     periods
    -0.25
     harness
    -0.24
    ÙħÙĬر
    -0.24
     Eastern
    -0.24
     processes
    -0.24
    isoft
    -0.24
    POSITIVE LOGITS
    çĶ¸
    0.28
     Obl
    0.28
    æıĨ
    0.27
    quarters
    0.26
    ucks
    0.25
    quirer
    0.25
    æ±Ģ
    0.24
    logan
    0.24
     joe
    0.24
     Tribune
    0.24
    Activations Density 25.786%

    No Known Activations