© 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. 107141
    Prev
    Next
    INDEX
    Explanations

    emotions, results

    np_max-act · gemini-2.0-flash
    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
    æŃ»è§Ĵ
    -0.26
    aurus
    -0.26
    åı«
    -0.24
    uggy
    -0.23
    ULE
    -0.23
     Stem
    -0.23
    ervisor
    -0.23
     Tanks
    -0.23
    çŀ»
    -0.23
    \TestCase
    -0.23
    POSITIVE LOGITS
    ãģ½
    0.29
    -Based
    0.28
    åĮĸè¿Ľç¨ĭ
    0.28
    ified
    0.27
    -based
    0.26
    åĤ¬
    0.26
    åĮĸçļĦ
    0.26
     âŁ
    0.26
    èµħ
    0.25
    Clar
    0.24
    Activations Density 24.284%

    No Known Activations