© 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. 19869
    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
    -Man
    -0.29
     dara
    -0.27
    ä¸ĩåħĥ以ä¸Ĭ
    -0.26
     partnering
    -0.25
    度è¿ĩ
    -0.24
    /ca
    -0.24
     infl
    -0.24
    常è¯Ĩ
    -0.24
     pouring
    -0.24
     heter
    -0.24
    POSITIVE LOGITS
    ounce
    0.31
    cratch
    0.31
    è¿Ļ次
    0.28
    ç½ijæł¼
    0.25
    acen
    0.25
    discard
    0.25
    cased
    0.25
     bare
    0.25
     '#
    0.25
    ulas
    0.25
    Activations Density 0.000%

    No Known Activations

    This feature has no known activations.