© 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. 89220
    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
    empt
    -0.26
    æ¥Ŀ
    -0.26
    ones
    -0.26
    ë¶Ī
    -0.25
    |↵↵
    -0.25
    åĪĨ
    -0.25
    æµ·è¿IJ
    -0.24
    .Single
    -0.24
    åĪĨæ³Į
    -0.23
    .special
    -0.23
    POSITIVE LOGITS
    æ¸ħæ¥ļ
    0.27
     appendix
    0.27
    vide
    0.26
    pNet
    0.26
    å¦Ĥå®ŀ
    0.26
    _atts
    0.25
    çĽ¸ä¼´
    0.25
    æĪ¥
    0.24
     filmer
    0.24
     baths
    0.23
    Activations Density 1.041%

    No Known Activations

    This feature has no known activations.