© 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. 142575
    Prev
    Next
    INDEX
    Explanations

    General text snippets

    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
    åľ¨è¿ĩ渡
    -0.30
    ä¹Łéļıä¹ĭ
    -0.28
    abhäng
    -0.26
     trách
    -0.25
    ä¸į解
    -0.25
     +++
    -0.25
    å°±å¼Ģå§ĭ
    -0.25
    âľķ
    -0.25
    ksam
    -0.25
    åij¼
    -0.24
    POSITIVE LOGITS
     aw
    0.28
    inish
    0.26
     Aw
    0.24
    .tim
    0.24
    bial
    0.24
    getBlock
    0.24
    illery
    0.24
    è¿Ľä¿®
    0.24
    reno
    0.23
    çĻ¾èĬ±
    0.23
    Activations Density 4.956%

    No Known Activations