© 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. 63912
    Prev
    Next
    INDEX
    Explanations

    punctuation

    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¦ĤæĦı
    -0.28
    lio
    -0.27
    č
    -0.27
    åĵī
    -0.25
     themselves
    -0.25
    åı«
    -0.25
    æĪij没æľī
    -0.25
     o
    -0.24
    æĪij没
    -0.24
    pth
    -0.24
    POSITIVE LOGITS
    urally
    0.31
    plate
    0.30
     Able
    0.29
    èĪįå¾Ĺ
    0.26
    Plate
    0.26
     DVR
    0.25
    .pkg
    0.25
     Conserv
    0.25
     fibre
    0.23
    _tolerance
    0.23
    Activations Density 2.835%

    No Known Activations