© 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. 64228
    Prev
    Next
    INDEX
    Explanations

    comma

    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
    iasm
    -0.26
    chal
    -0.25
    è®®æ¡Ī
    -0.24
    ä¸įä½ı
    -0.23
    ä¼ģä¸ļæıIJä¾Ľ
    -0.23
    @qq
    -0.23
     graph
    -0.23
     LSU
    -0.23
    ModelError
    -0.23
    é
    -0.23
    POSITIVE LOGITS
     calves
    0.26
     Fernando
    0.25
    æ´²
    0.25
    hel
    0.25
     podr
    0.24
    é£
    0.24
    ipline
    0.24
    å¹³æĿ¿
    0.24
     Venez
    0.24
     automáticamente
    0.24
    Activations Density 24.482%

    No Known Activations