© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-4B
    5. 20-TRANSCODER-HP
    7. 130906
    Prev
    Next
    INDEX
    Explanations

    Research and signalling

    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
    Negative Logits
    è¾Ļ
    -0.28
    unded
    -0.28
     Sadd
    -0.28
    ode
    -0.27
     Powerful
    -0.27
     UNU
    -0.26
    lements
    -0.26
    oreal
    -0.24
    lish
    -0.24
    æĹłå½¢
    -0.24
    POSITIVE LOGITS
    idata
    0.27
     treat
    0.27
    irs
    0.25
    wij
    0.24
     treated
    0.24
    ijo
    0.24
    åĥı
    0.23
    çªŁ
    0.23
     onData
    0.23
     units
    0.23
    Activations Density 14.655%

    No Known Activations