© 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. 30-TRANSCODER-HP
    7. 96891
    Prev
    Next
    INDEX
    Explanations

    say "se" word

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_30.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.30.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Negative Logits
    omics
    -0.27
    enos
    -0.27
     cop
    -0.26
     Universität
    -0.25
    irc
    -0.25
    骢
    -0.25
    ÃŃcia
    -0.25
     inscription
    -0.25
     railroad
    -0.24
     Insp
    -0.24
    POSITIVE LOGITS
    ometown
    0.28
    åıįæĺł
    0.27
     SceneManager
    0.26
    bolt
    0.25
    纪
    0.24
    prox
    0.24
    Abs
    0.24
     setContent
    0.24
    -validate
    0.24
     absorb
    0.23
    Activations Density 0.782%

    No Known Activations