© 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. 63542
    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
    rome
    -0.27
    caps
    -0.26
     consenting
    -0.26
    è¿Ļæł·è¯´
    -0.26
     sweetness
    -0.26
    éĤĤ
    -0.25
    款
    -0.25
    temp
    -0.24
     Prem
    -0.24
     Herman
    -0.24
    POSITIVE LOGITS
    chest
    0.28
    æļĩ
    0.26
     Affiliate
    0.26
     back
    0.26
    æĿ¡ä¾ĭ
    0.25
    iless
    0.25
    仡
    0.25
    inous
    0.25
    dez
    0.25
    stab
    0.25
    Activations Density 1.558%

    No Known Activations

    This feature has no known activations.