© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-4B
    5. 18-TRANSCODER-HP
    7. 89699
    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
    Not in Any Lists

    No Comments

    Negative Logits
    frage
    -0.28
    交æµģ
    -0.26
    沸
    -0.26
    asiswa
    -0.25
    ningar
    -0.25
    apan
    -0.24
      ↵  ↵
    -0.24
    ×ķ×¥
    -0.24
    çģ«çĥ§
    -0.24
    çģ«èĬ±
    -0.24
    POSITIVE LOGITS
    lua
    0.28
    icum
    0.28
    为空
    0.27
     görmek
    0.26
    uya
    0.25
    BuilderInterface
    0.24
    æĹ¶æľŁçļĦ
    0.23
    侪
    0.23
    .md
    0.23
    海绵
    0.23
    Activations Density 87.479%

    No Known Activations

    This feature has no known activations.