© 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. 64463
    Prev
    Next
    INDEX
    Explanations

    technical information

    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
    ç»ĵæŀĦæĢ§
    -0.27
     Hobby
    -0.27
     sä
    -0.26
    racial
    -0.26
    ä¸Ģ款
    -0.26
    ÙĦÙĪÙħ
    -0.26
    篮æĿ¿
    -0.25
    åĩĮ
    -0.24
     inex
    -0.24
    æ¯Ķ
    -0.24
    POSITIVE LOGITS
    -away
    0.28
    æľĢå¤ļçļĦ
    0.26
    SW
    0.25
     approval
    0.25
    åįİ缼
    0.25
    ï¼ħ
    0.24
    åŁºæķ°
    0.24
    -most
    0.24
    udo
    0.24
    ãģĭãģijãģ¦
    0.24
    Activations Density 3.603%

    No Known Activations