© 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. 107945
    Prev
    Next
    INDEX
    Explanations

    Version

    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°ıé¢Ŀ
    -0.26
     rat
    -0.26
    romatic
    -0.25
    å°Ĩèĩªå·±çļĦ
    -0.25
     vis
    -0.24
     repe
    -0.24
    anager
    -0.24
    示
    -0.24
    ä¸Ģè¡Į
    -0.24
    itative
    -0.24
    POSITIVE LOGITS
    erde
    0.30
    gger
    0.27
    er
    0.25
    contri
    0.25
    Uses
    0.25
     thẻ
    0.25
    eka
    0.25
     Lives
    0.24
    erge
    0.24
    ä¹ĭå¤ĸ
    0.23
    Activations Density 1.125%

    No Known Activations