© 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. 130531
    Prev
    Next
    INDEX
    Explanations

    emphasize

    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
     Tradition
    -0.25
    eturn
    -0.24
    表éĿ¢
    -0.24
    åŁĭ
    -0.24
    ERSIST
    -0.24
    aras
    -0.23
    漫æŃ¥
    -0.23
    quete
    -0.23
    surface
    -0.23
    s
    -0.23
    POSITIVE LOGITS
    亲
    0.30
     Loud
    0.29
    upal
    0.28
    Ñĭва
    0.28
    onor
    0.26
    æŸ
    0.26
    aisy
    0.26
    hid
    0.24
     all
    0.24
     ord
    0.24
    Activations Density 14.906%

    No Known Activations