© 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. 63541
    Prev
    Next
    INDEX
    Explanations

    formal/technical documentation excerpts

    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
    ä»ħ代表
    -0.26
    sequ
    -0.26
     makeStyles
    -0.26
    thumbnail
    -0.25
     rude
    -0.24
     bure
    -0.24
     sequ
    -0.24
    uvwxyz
    -0.24
     полÑĮз
    -0.24
    çĬ¹
    -0.24
    POSITIVE LOGITS
    ograd
    0.29
    ORN
    0.29
    Either
    0.28
    è¦ģä¹Ī
    0.26
     Either
    0.26
     Incoming
    0.25
     either
    0.25
    è¿IJ转
    0.25
    ãĤ«ãĥĨãĤ´
    0.25
    ategories
    0.24
    Activations Density 24.267%

    No Known Activations