© 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. 88794
    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
    -clock
    -0.27
    ĽĦ
    -0.26
     luck
    -0.26
     generics
    -0.26
     clock
    -0.24
    usi
    -0.24
    æĬ¥åijĬ
    -0.24
    èĢ»
    -0.24
     Luck
    -0.23
    çķļ
    -0.23
    POSITIVE LOGITS
     skipped
    0.28
    æľ¬ç½ij
    0.26
     pÅĤ
    0.26
    温
    0.26
    onas
    0.25
    ǧ
    0.25
    æĹ¶è¡¨ç¤º
    0.25
    åįŁ
    0.25
     prive
    0.25
    çļĦ身影
    0.24
    Activations Density 1.329%

    No Known Activations

    This feature has no known activations.