© 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. 63557
    Prev
    Next
    INDEX
    Explanations

    computer code segments

    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
    -0.34
    uis
    -0.28
     âĢ¦↵↵
    -0.25
     âĢ¦
    -0.24
    天ä¸ĭ
    -0.24
     tract
    -0.24
     Nas
    -0.23
    ä¸įä¸ĭ
    -0.23
    ci
    -0.23
    -0.23
    POSITIVE LOGITS
    yle
    0.27
    OME
    0.26
    umber
    0.24
     pr
    0.23
    bomb
    0.23
    orman
    0.23
    edBy
    0.23
    dex
    0.23
    edException
    0.23
    StringEncoding
    0.22
    Activations Density 81.021%

    No Known Activations