© 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. 90100
    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
     pou
    -0.27
     Downloads
    -0.27
    èIJİ
    -0.26
    edo
    -0.26
     Pou
    -0.26
    ruz
    -0.25
     pressures
    -0.25
    æķĻåŃ¦è´¨éĩı
    -0.24
    AndPassword
    -0.24
    äº®çľ¼
    -0.24
    POSITIVE LOGITS
    缴è¾ĸå¸Ĥ
    0.28
    éĶļ
    0.25
    -uppercase
    0.25
     annihil
    0.24
    æ»ŀåIJİ
    0.24
    åįıè°ĥ
    0.24
    omorphic
    0.23
    ,@
    0.23
    å«Į
    0.23
    tring
    0.23
    Activations Density 36.976%

    No Known Activations

    This feature has no known activations.