© 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. 35-TRANSCODER-HP
    7. 40599
    Prev
    Next
    INDEX
    Explanations

    legal documents/records

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_35.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.35.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Negative Logits
    ular
    -0.33
    ekt
    -0.31
    etics
    -0.29
     extractor
    -0.27
    unos
    -0.27
    enko
    -0.27
    istr
    -0.26
    isNew
    -0.26
    as
    -0.26
    ators
    -0.26
    POSITIVE LOGITS
    riage
    0.47
    riages
    0.43
    car
    0.34
    ri
    0.32
    rière
    0.30
    riere
    0.29
    coon
    0.29
    ci
    0.29
    ció
    0.28
    CI
    0.28
    Activations Density 0.450%

    No Known Activations