© 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. 129791
    Prev
    Next
    INDEX
    Explanations

    technical

    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
    çħ§
    -0.27
    actly
    -0.27
    ĶåĽŀ
    -0.26
    setName
    -0.26
    IGNAL
    -0.25
    ħ§
    -0.25
    orra
    -0.25
     genuinely
    -0.25
    onde
    -0.24
    heat
    -0.24
    POSITIVE LOGITS
    pps
    0.30
    áĢĬ
    0.29
    èĬ
    0.26
     Tud
    0.26
     imperial
    0.25
    kan
    0.25
    É¡
    0.24
    åIJĮæĹ¶ä¹Łæĺ¯
    0.23
     Ung
    0.23
    天çĶŁ
    0.23
    Activations Density 75.602%

    No Known Activations