© 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. 63593
    Prev
    Next
    INDEX
    Explanations

    years

    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
    effect
    -0.28
    interpret
    -0.26
    _gap
    -0.26
    prim
    -0.26
    arbonate
    -0.25
    çļĦæĪIJ绩
    -0.24
    æĸ¥
    -0.24
    è¿Ļéĥ¨åĪĨ
    -0.24
    çĶŁæĪIJ
    -0.24
    estation
    -0.24
    POSITIVE LOGITS
    emat
    0.28
    éķ¿æľŁ
    0.28
    unist
    0.27
    å¼ĺæī¬
    0.26
     '=
    0.26
    éķ·æľŁ
    0.25
    aos
    0.24
    ç½ij讯
    0.24
     hovered
    0.23
    åģ»
    0.23
    Activations Density 10.180%

    No Known Activations