© 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. 141040
    Prev
    Next
    INDEX
    Explanations

    ,

    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
    å®ĺ
    -0.30
    ä¹Łä¸º
    -0.29
    åŃIJ
    -0.26
    inen
    -0.25
    usk
    -0.25
    Disconnect
    -0.24
    uestos
    -0.23
     plist
    -0.23
    UTO
    -0.23
    poster
    -0.23
    POSITIVE LOGITS
    çģ°èī²
    0.30
    åĶ¯ä¸ĢçļĦ
    0.28
    æĽ·
    0.27
    æĥŁ
    0.24
     NRA
    0.24
     Problem
    0.24
    å¼ĢåIJ¯äºĨ
    0.23
    èĨ³
    0.23
     revis
    0.23
     suppressed
    0.23
    Activations Density 6.841%

    No Known Activations