© 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. 108243
    Prev
    Next
    INDEX
    Explanations

    Code, URLs, and foreign languages

    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
    omid
    -0.25
    ä¸İåħ¶
    -0.24
    ategory
    -0.24
    >F
    -0.23
     rôle
    -0.23
    å¹ħ度
    -0.23
    estr
    -0.23
    壮
    -0.23
    Role
    -0.23
    sid
    -0.23
    POSITIVE LOGITS
    urch
    0.30
    æĢ¥
    0.27
    urge
    0.26
    urgeon
    0.26
    @s
    0.25
     sticks
    0.24
    redd
    0.24
    oup
    0.24
     disemb
    0.24
    lag
    0.23
    Activations Density 3.446%

    No Known Activations