© 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. 90342
    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
    OMATIC
    -0.29
    Bullet
    -0.27
    çļĦ女人
    -0.26
    anism
    -0.25
    /lib
    -0.25
     unarmed
    -0.25
    é»ĺ
    -0.25
    illas
    -0.24
     lib
    -0.24
     settled
    -0.24
    POSITIVE LOGITS
    æ§ģ
    0.25
    eko
    0.25
    urable
    0.25
    ç¼Ķ
    0.25
    cuador
    0.25
     Contractor
    0.24
    .gmail
    0.24
    ä¸ĢæŶ
    0.24
     duro
    0.24
    ç³Ł
    0.24
    Activations Density 1.492%

    No Known Activations

    This feature has no known activations.