© 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. 63049
    Prev
    Next
    INDEX
    Explanations
    No Explanations Found
    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
    æĿ¥äºĨ
    -0.28
    æ²Ļ滩
    -0.26
    apur
    -0.25
    æĶ»çķ¥
    -0.25
     '</
    -0.25
    ',(
    -0.24
    èIJ©
    -0.24
    acre
    -0.24
     stayed
    -0.24
    å¥
    -0.24
    POSITIVE LOGITS
     tre
    0.30
     oxidative
    0.26
    åĿĽ
    0.25
     derivative
    0.25
     congestion
    0.25
    (expected
    0.25
    bles
    0.25
     early
    0.25
     prom
    0.25
    -th
    0.25
    Activations Density 25.893%

    No Known Activations

    This feature has no known activations.