© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-4B
    5. 20-TRANSCODER-HP
    7. 140741
    Prev
    Next
    INDEX
    Explanations

    code/technical content

    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
    Negative Logits
    åħ¬å¼Ģ
    -0.27
     arbitrarily
    -0.27
    è¦ģä¹Ī
    -0.26
    soever
    -0.25
    whole
    -0.25
     publi
    -0.25
     pis
    -0.25
    AMES
    -0.24
    åİŁå§ĭ
    -0.24
    åıijå¸ĥ
    -0.24
    POSITIVE LOGITS
    elia
    0.29
    elize
    0.28
    ä¸įåłª
    0.26
    çĻ½é¢Ĩ
    0.24
    Listener
    0.23
     бол
    0.23
    婶
    0.23
    gL
    0.23
    åIJ¬è¯´
    0.23
    转
    0.23
    Activations Density 2.253%

    No Known Activations