© 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. 107056
    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
    )":
    -0.27
    endl
    -0.27
    ä¹ĭ称
    -0.26
     historically
    -0.26
    åĮłå¿ĥ
    -0.24
    licts
    -0.24
    ç²¾åįİ
    -0.24
    dings
    -0.24
    acters
    -0.23
    Compile
    -0.23
    POSITIVE LOGITS
    iest
    0.29
    让çĶ¨æĪ·
    0.27
    bolt
    0.26
    å¹²ä»Ģä¹Ī
    0.24
     Origin
    0.24
    çĹĴ
    0.24
    çļĦæĸ¹åIJij
    0.24
    _ORIGIN
    0.24
    宦
    0.23
    iesta
    0.23
    Activations Density 39.695%

    No Known Activations

    This feature has no known activations.