© 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. 31-TRANSCODER-HP
    7. 17743
    Prev
    Next
    INDEX
    Explanations

    Unrelated articles with delimiters

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_31.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.31.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Negative Logits
    bas
    -0.32
    _form
    -0.28
    ans
    -0.27
    ÙĨÙĩ
    -0.26
    FormData
    -0.26
    ben
    -0.26
    .form
    -0.25
    form
    -0.25
     FormData
    -0.25
     Form
    -0.25
    POSITIVE LOGITS
    Ïĭ
    0.31
    pee
    0.30
    èµī
    0.28
    èĶ¸
    0.28
    issement
    0.26
    heed
    0.26
    梦è§ģ
    0.24
    UIViewController
    0.24
    adro
    0.24
    ouncil
    0.24
    Activations Density 0.127%

    No Known Activations