© 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. 129691
    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
    presence
    -0.28
    ä»ħæľī
    -0.27
     presence
    -0.27
     following
    -0.26
    Presence
    -0.25
     Presence
    -0.25
    á»ĸ
    -0.24
    nage
    -0.24
    _presence
    -0.23
    主ä¹īèĢħ
    -0.23
    POSITIVE LOGITS
    scal
    0.26
    (scale
    0.25
    è¿Ļåľº
    0.25
     scales
    0.25
     leo
    0.25
    ERIC
    0.24
    Ļ¨
    0.24
    favor
    0.23
    rende
    0.23
    ëĮĢë¡ľ
    0.23
    Activations Density 1.076%

    No Known Activations

    This feature has no known activations.