© 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. 18-TRANSCODER-HP
    7. 88179
    Prev
    Next
    INDEX
    Explanations
    No Explanations Found
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_18.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.18.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Negative Logits
    è¾ĸ
    -0.26
     FALL
    -0.25
    å¸Ľ
    -0.25
    corn
    -0.24
    è°ļ
    -0.24
    Bits
    -0.23
    å¼Ģå¹ķ
    -0.23
     groom
    -0.23
    举åĬŀçļĦ
    -0.23
    elor
    -0.23
    POSITIVE LOGITS
    cee
    0.26
    åĪĨæĺİ
    0.25
    Enumeration
    0.25
     guesses
    0.23
    Ä°YE
    0.23
    EndElement
    0.23
    EqualTo
    0.23
    .googleapis
    0.23
    车éĹ´
    0.23
     IDb
    0.23
    Activations Density 39.618%

    No Known Activations

    This feature has no known activations.