© 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. 63153
    Prev
    Next
    INDEX
    Explanations
    No Explanations Found
    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
     lived
    -0.30
    Lifecycle
    -0.27
    .live
    -0.27
    еÑĩ
    -0.27
     measured
    -0.26
    Ĥ¤
    -0.26
    -lived
    -0.25
    .Operation
    -0.25
     earth
    -0.25
    ä¸Ģé¡¿
    -0.25
    POSITIVE LOGITS
    sy
    0.30
    ctal
    0.27
    æİĩ
    0.26
    [Test
    0.26
    赢家
    0.26
    æł¡å¤ĸ
    0.25
    åħ·ä½ĵçļĦ
    0.25
    adelphia
    0.25
    ç°§
    0.25
    allel
    0.25
    Activations Density 2.081%

    No Known Activations

    This feature has no known activations.