© 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. 17898
    Prev
    Next
    INDEX
    Explanations
    No Explanations Found
    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
     inducing
    -0.35
    (ind
    -0.31
     Ind
    -0.31
     induces
    -0.31
    Ind
    -0.29
    _ind
    -0.29
    ind
    -0.29
    inds
    -0.28
     practiced
    -0.28
     ind
    -0.28
    POSITIVE LOGITS
     incomplete
    0.28
    phan
    0.28
     either
    0.28
    鼷æĸ¯
    0.27
    为空
    0.27
    omorphic
    0.27
     dette
    0.26
     completely
    0.26
    ammed
    0.25
    顾
    0.25
    Activations Density 0.078%

    No Known Activations

    This feature has no known activations.