© 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. 64727
    Prev
    Next
    INDEX
    Explanations

    participate

    np_max-act · gemini-2.0-flash
    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
     NSRange
    -0.30
    éķľ
    -0.29
    éķľåŃIJ
    -0.28
    æķ£å¸ĥ
    -0.27
    èĥ½è®©
    -0.27
    æı©
    -0.27
    å°ıäºĭ
    -0.26
     parseInt
    -0.26
    eya
    -0.25
     tslint
    -0.25
    POSITIVE LOGITS
    ;++
    0.25
     competed
    0.25
     carved
    0.25
     intermediate
    0.25
     screaming
    0.24
     independent
    0.24
     instituted
    0.24
     social
    0.24
    è··
    0.24
     cool
    0.24
    Activations Density 15.364%

    No Known Activations