© 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. 64839
    Prev
    Next
    INDEX
    Explanations

    Pacific

    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
    apat
    -0.27
     Less
    -0.27
    amen
    -0.25
    enos
    -0.25
    udy
    -0.25
    łéĻ¤
    -0.25
     verg
    -0.25
    .nan
    -0.24
     nas
    -0.24
    lingen
    -0.23
    POSITIVE LOGITS
    éric
    0.25
    .toUpperCase
    0.23
    avity
    0.23
    ÑĤаÑĢ
    0.23
    æĢ»ä½ĵ
    0.23
    HDR
    0.23
    icter
    0.23
    èĩªè±ª
    0.23
    主æµģ
    0.23
     Oaks
    0.23
    Activations Density 13.281%

    No Known Activations