© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Qwen3-4B
    3. 20-TRANSCODER-HP
    4. 63372
    Prev
    Next
    INDEX
    Explanations

    Web-scraped text

    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
    å¯Įæľī
    -0.28
    YW
    -0.28
    宫
    -0.27
    ür
    -0.26
    åħ±
    -0.26
    ixo
    -0.26
    comed
    -0.25
    âĦł
    -0.25
    ä¿®
    -0.25
    mailbox
    -0.24
    POSITIVE LOGITS
    Traits
    0.27
     Fore
    0.26
    è·Łå¥¹
    0.26
    дов
    0.26
     fore
    0.26
     minut
    0.25
     ÐľÐ¸Ð½
    0.25
    Fore
    0.24
     nowhere
    0.24
    Paren
    0.23
    Activations Density 1.641%

    No Known Activations