© 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. 62541
    Prev
    Next
    INDEX
    Explanations

    so

    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
    pository
    -0.27
    å¹´æĬ¥
    -0.27
    stad
    -0.25
     tal
    -0.25
    Sing
    -0.25
     Sing
    -0.25
    æīįçŁ¥éģĵ
    -0.25
    Normally
    -0.25
     dom
    -0.24
     UN
    -0.24
    POSITIVE LOGITS
    -li
    0.27
     xsi
    0.27
    æĹłå½¢
    0.27
    _geom
    0.26
    contra
    0.25
    æŶåĪ»
    0.25
    å¥ł
    0.25
     Lucia
    0.25
    ĥģ
    0.25
     lsp
    0.24
    Activations Density 42.312%

    No Known Activations