© 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. 140627
    Prev
    Next
    INDEX
    Explanations

    to

    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
    Ä°K
    -0.29
     thereof
    -0.26
    åıį
    -0.26
    cor
    -0.26
    uter
    -0.25
    å°Ħ
    -0.24
    ikel
    -0.23
    icated
    -0.23
    ä¹ĭä¸Ģ
    -0.23
    icol
    -0.23
    POSITIVE LOGITS
    asting
    0.28
    ffee
    0.26
    å²Ľ
    0.25
    çĺģ
    0.25
    resent
    0.24
     MOT
    0.24
    orable
    0.24
     molec
    0.24
    大éĻĨ
    0.24
    èĬĹ
    0.23
    Activations Density 24.904%

    No Known Activations