© 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
    3. Qwen3-4B
    5. 20-TRANSCODER-HP
    7. 108309
    Prev
    Next
    INDEX
    Explanations

    technical jargon

    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
    éĴĬ
    -0.26
    ojis
    -0.26
    åŁ¹åħ»
    -0.26
    éĹ´æİ¥
    -0.26
    çļĦæĥħ
    -0.26
    ordered
    -0.25
    çĭ¬ä¸Ģ
    -0.25
     desi
    -0.25
     double
    -0.25
    èĥİ
    -0.24
    POSITIVE LOGITS
    iken
    0.28
     ken
    0.28
    entials
    0.26
    jal
    0.25
    aset
    0.25
    azel
    0.24
    areth
    0.24
    éķľå¤´
    0.24
    åŁºéĩij份é¢Ŀ
    0.24
    perate
    0.23
    Activations Density 32.724%

    No Known Activations