© 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. 64749
    Prev
    Next
    INDEX
    Explanations

    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_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
     crackers
    -0.26
    ois
    -0.26
     attendance
    -0.26
    èijŃ
    -0.24
     gr
    -0.24
    åįı
    -0.24
     groceries
    -0.24
    oning
    -0.24
     Lat
    -0.24
    ieder
    -0.23
    POSITIVE LOGITS
     reels
    0.26
     needles
    0.26
    ind
    0.26
    è§Ħå¾ĭ
    0.26
    ield
    0.24
    å±Ģéĥ¨
    0.24
    éĴĪ
    0.24
    ilogy
    0.24
    icals
    0.23
    pins
    0.23
    Activations Density 67.905%

    No Known Activations