© 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. 142221
    Prev
    Next
    INDEX
    Explanations

    Alphanumeric characters in code

    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§Ķ
    -0.27
    èŁł
    -0.26
    åį§
    -0.24
    éĵ¸éĢł
    -0.24
    渺
    -0.24
    å¤Ħ
    -0.24
    ä¹Łä¸įæķ¢
    -0.24
    å°ı康社ä¼ļ
    -0.24
    df
    -0.24
     informative
    -0.24
    POSITIVE LOGITS
     contrad
    0.29
    ãģľ
    0.28
    appe
    0.26
    é¹ĥ
    0.26
    %E
    0.26
    人åĿĩ
    0.26
    äng
    0.25
    ClientId
    0.25
     heck
    0.25
    abra
    0.24
    Activations Density 11.609%

    No Known Activations