© 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. 18-TRANSCODER-HP
    7. 90273
    Prev
    Next
    INDEX
    Explanations

    Formal/technical documents and code

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_18.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.18.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Negative Logits
     reached
    -0.30
     Reached
    -0.27
    onte
    -0.25
     Morg
    -0.25
    åħ¬æĬ¥
    -0.25
    éģĶ
    -0.25
     proclaim
    -0.25
    秤
    -0.24
    opleft
    -0.24
     reach
    -0.24
    POSITIVE LOGITS
    ä¸Ģå®ļ
    0.28
    éĹ¯
    0.27
     Parr
    0.25
     kısı
    0.24
    artz
    0.24
    åĸĤ
    0.24
    äºĶå®ĺ
    0.24
    terdam
    0.24
    assic
    0.24
    pare
    0.23
    Activations Density 3.423%

    No Known Activations