© 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. 63813
    Prev
    Next
    INDEX
    Explanations

    see

    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
    é¦ĸåħĪæĺ¯
    -0.31
    ensual
    -0.29
    åĨĮ
    -0.28
    è¹ij
    -0.28
    aml
    -0.26
    Ļ¨
    -0.26
     cubes
    -0.26
    onus
    -0.26
    xit
    -0.25
    onal
    -0.25
    POSITIVE LOGITS
    cca
    0.28
     worse
    0.28
    ppers
    0.26
     Worse
    0.26
    CCA
    0.26
    autop
    0.25
    æİ¥è¿ŀ
    0.25
     engulf
    0.25
     wors
    0.25
     convey
    0.24
    Activations Density 19.938%

    No Known Activations