© 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. 64190
    Prev
    Next
    INDEX
    Explanations

    numbers and percentages

    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
     zero
    -0.38
     negative
    -0.38
    鼶
    -0.36
    ZERO
    -0.34
    _zero
    -0.34
     ZERO
    -0.34
    Zero
    -0.34
     positive
    -0.34
    zero
    -0.33
    negative
    -0.33
    POSITIVE LOGITS
    аÑĤоÑĢ
    0.32
    缸è§ģ
    0.26
    ä¸Ģåı£æ°Ķ
    0.26
    èµ°å¾Ĺ
    0.26
    ä»ħ次äºİ
    0.25
    .encoder
    0.24
    etsk
    0.24
     dispenser
    0.24
     Second
    0.23
    forc
    0.23
    Activations Density 22.716%

    No Known Activations