© 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. 64534
    Prev
    Next
    INDEX
    Explanations

    Punctuation and connector words

    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
    Not in Any Lists

    No Comments

    Negative Logits
    &r
    -0.27
    é¢Ĩ
    -0.26
    车åŃIJ
    -0.26
    .pref
    -0.25
    :right
    -0.25
    ายà¸Ļ
    -0.25
    &A
    -0.24
    ä¸įä»ħ
    -0.24
    ÙģÙĬÙĦ
    -0.24
    tracted
    -0.24
    POSITIVE LOGITS
     disap
    0.31
    quia
    0.28
    otte
    0.26
     Jug
    0.24
     matter
    0.24
    çıª
    0.24
    maid
    0.24
    ä¸ĢåIJĮ
    0.24
    gere
    0.24
     Relation
    0.24
    Activations Density 8.214%

    No Known Activations