© 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. 142693
    Prev
    Next
    INDEX
    Explanations

    punctuation

    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
     Pee
    -0.33
    åľ¨åĽ½åĨħ
    -0.27
    æĺ¯åĽ½åĨħ
    -0.27
    æľŁåĨħ
    -0.27
    ãĥ¼ãĤ¸
    -0.27
    eness
    -0.26
    tron
    -0.26
    ìĤ¬ëĬĶ
    -0.25
    enie
    -0.25
    èµ·çĤ¹
    -0.24
    POSITIVE LOGITS
    fern
    0.27
     cross
    0.26
    yield
    0.26
    ç»Ħ
    0.26
     yield
    0.26
    :index
    0.25
     jack
    0.25
    /month
    0.25
    abouts
    0.25
     geld
    0.24
    Activations Density 17.580%

    No Known Activations