© 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. 108287
    Prev
    Next
    INDEX
    Explanations

    digital technology

    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
    à°³
    -0.33
     credit
    -0.27
    cho
    -0.26
     scrambling
    -0.25
    sie
    -0.25
    åĴĦ
    -0.24
    éħ¥
    -0.24
     thaw
    -0.24
    å¤ĸè´¸
    -0.24
    -ra
    -0.23
    POSITIVE LOGITS
     veg
    0.26
    pres
    0.25
     pres
    0.25
     minut
    0.25
    ((*
    0.25
    _over
    0.25
    躺
    0.25
    :key
    0.24
     kab
    0.24
    ä¸Ģèµ·
    0.24
    Activations Density 10.273%

    No Known Activations