© 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. 141957
    Prev
    Next
    INDEX
    Explanations

    Random excerpts from various sources

    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
    天ä¸ĭ
    -0.29
    ést
    -0.29
    ually
    -0.28
     scratch
    -0.27
     Hurricane
    -0.26
    èĪªå¤©
    -0.25
    itin
    -0.25
    çļĦä¼Łå¤§
    -0.24
    berger
    -0.24
    utral
    -0.24
    POSITIVE LOGITS
    éĺµèIJ¥
    0.30
    åĿĬ
    0.26
     working
    0.25
     pockets
    0.25
    åįıåĬŀ
    0.24
    CellValue
    0.24
    means
    0.24
    æ¤Ĵ
    0.24
    åĵĩ
    0.23
     variety
    0.23
    Activations Density 9.606%

    No Known Activations