© 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. 90058
    Prev
    Next
    INDEX
    Explanations

    technical or medical writing

    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
    ecd
    -0.33
    åıįæŃ£
    -0.28
     mostly
    -0.26
    naissance
    -0.26
     most
    -0.26
     owner
    -0.26
     radio
    -0.25
    篷
    -0.25
    rend
    -0.24
    atti
    -0.24
    POSITIVE LOGITS
    алÑĮн
    0.28
     fart
    0.27
     personals
    0.25
    书ä¸Ń
    0.25
    æĭĤ
    0.25
     misunder
    0.25
     Sunshine
    0.25
     Colbert
    0.24
    emann
    0.24
    æŁ¥çľĭæĽ´å¤ļ
    0.24
    Activations Density 2.282%

    No Known Activations