© 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. 20-TRANSCODER-HP
    7. 64760
    Prev
    Next
    INDEX
    Explanations
    No Explanations Found
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_20.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.20.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Negative Logits
    igos
    -0.26
    egin
    -0.26
     assembly
    -0.26
     arms
    -0.25
     ten
    -0.25
    èijĽ
    -0.25
    aye
    -0.25
    ä¹°çļĦ
    -0.25
    vis
    -0.24
    æݩ饰
    -0.24
    POSITIVE LOGITS
    @register
    0.32
    orate
    0.28
     satisfy
    0.28
    åĪİ
    0.28
    ä¾Ľæ±Ĥ
    0.25
    社ä¼ļ稳å®ļ
    0.25
     Beard
    0.24
    äºķ
    0.24
    inator
    0.23
    @admin
    0.23
    Activations Density 1.069%

    No Known Activations

    This feature has no known activations.