© 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. 107751
    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
    @stop
    -0.27
    ocrates
    -0.26
    etur
    -0.26
    .dir
    -0.25
    urile
    -0.24
    å¤Ķ
    -0.24
     Ethi
    -0.24
    ARCH
    -0.24
    plode
    -0.24
     cruc
    -0.24
    POSITIVE LOGITS
    olen
    0.28
    ÑĢен
    0.27
    侯
    0.27
    ä¸ĢçĽ´åľ¨
    0.26
    pte
    0.26
    serve
    0.26
    sp
    0.26
    ç»ıåħ¸çļĦ
    0.25
    andan
    0.25
    ledo
    0.24
    Activations Density 13.852%

    No Known Activations

    This feature has no known activations.