© 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. 88093
    Prev
    Next
    INDEX
    Explanations

    Scientific/technical 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
    åľį
    -0.26
    ech
    -0.26
    sworth
    -0.25
     Either
    -0.25
     Ups
    -0.24
    Pipe
    -0.24
    Gro
    -0.24
     navigationController
    -0.24
    æĢ§è´¨
    -0.24
    飫
    -0.23
    POSITIVE LOGITS
    åĨ°
    0.30
    壤
    0.29
    iously
    0.28
     Mass
    0.27
    iot
    0.26
     Fowler
    0.26
    rips
    0.25
    _mass
    0.25
    -te
    0.24
    ç²²
    0.24
    Activations Density 66.931%

    No Known Activations