© 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. 64053
    Prev
    Next
    INDEX
    Explanations

    the

    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
    æī¼
    -0.28
    Ắ
    -0.27
    åıĭæĥħ
    -0.26
    tó
    -0.26
    åīĮ
    -0.25
     Directions
    -0.24
    eses
    -0.24
    -bar
    -0.23
     Parallel
    -0.23
     counter
    -0.23
    POSITIVE LOGITS
    å¾Ĺèµ·
    0.26
    çºĮ
    0.26
    è®®
    0.26
    reed
    0.26
    roe
    0.25
     backlog
    0.25
     Appro
    0.25
    åĩºåĬ¨
    0.24
     approves
    0.24
    ãģĦãģ¾
    0.24
    Activations Density 22.731%

    No Known Activations