© 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. 63494
    Prev
    Next
    INDEX
    Explanations
    No Explanations Found
    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
    elist
    -0.28
    PX
    -0.28
    åİŁ
    -0.24
    ç³»
    -0.24
     bình
    -0.24
    åįģä¸Ģ
    -0.23
     PD
    -0.23
    (pd
    -0.23
     pil
    -0.23
     hom
    -0.23
    POSITIVE LOGITS
     interfer
    0.29
    å¹²æī°
    0.28
    ç¯ģ
    0.27
    容éĩı
    0.26
    åĽ°éļ¾
    0.26
    天èĬ±
    0.26
    attention
    0.25
     interference
    0.25
    åŁİ管
    0.24
    çīĩåŃIJ
    0.24
    Activations Density 38.133%

    No Known Activations

    This feature has no known activations.