© 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. 141703
    Prev
    Next
    INDEX
    Explanations

    formal documents/academic writings

    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
    çļĦæĪIJåĬŁ
    -0.30
    iT
    -0.30
    ocities
    -0.28
    alice
    -0.28
    èIJ½åIJİ
    -0.25
    uja
    -0.25
    å¾Īæĸ¹ä¾¿
    -0.25
    çļĦç¨ĭ度
    -0.25
    för
    -0.24
    ä¸Ģç«Ļ
    -0.24
    POSITIVE LOGITS
    efa
    0.27
    à¸ĺา
    0.26
    æĸ¯åŁº
    0.26
    appropri
    0.26
    aying
    0.26
    æ¸ħ
    0.24
    (strip
    0.23
    ее
    0.23
    utive
    0.23
     Squ
    0.23
    Activations Density 4.606%

    No Known Activations