© 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. 89271
    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
    earn
    -0.26
    åľºåĿĩ
    -0.26
    æĥłåıĬ
    -0.24
     vÄĥn
    -0.24
     biased
    -0.24
    模
    -0.24
    åģ¶
    -0.24
    åħīèį£
    -0.24
    ide
    -0.23
     definite
    -0.23
    POSITIVE LOGITS
    arus
    0.32
     Shel
    0.25
    .reactivex
    0.25
     Holdings
    0.25
    fed
    0.24
    ä¸ļæĢģ
    0.24
    arez
    0.24
    çļĦæĢģ度
    0.23
     bastard
    0.23
    é²ľ
    0.23
    Activations Density 19.111%

    No Known Activations

    This feature has no known activations.