© 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. 17746
    Prev
    Next
    INDEX
    Explanations

    Se

    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
    说ä»Ģä¹Ī
    -0.27
     Down
    -0.27
    csv
    -0.26
    IT
    -0.26
     uv
    -0.25
    urette
    -0.25
     arms
    -0.25
    rende
    -0.25
    utton
    -0.24
     IT
    -0.24
    POSITIVE LOGITS
    omy
    0.27
    ogy
    0.25
    深交æīĢ
    0.24
    æµ´
    0.24
    .Annotations
    0.24
    ãĤ·ãĥ¼
    0.24
    称
    0.24
    åĪĿä¸ī
    0.24
    ä½ĵ系建设
    0.24
    oom
    0.24
    Activations Density 0.170%

    No Known Activations