© 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. 14-TRANSCODER-HP
    7. 41527
    Prev
    Next
    INDEX
    Explanations

    retro

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_14.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.14.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Negative Logits
    eya
    -0.28
     dealt
    -0.27
    MU
    -0.24
    ROUTE
    -0.24
     mouse
    -0.24
    æĸ°éĹ»è®°èĢħ
    -0.24
     deals
    -0.24
    eph
    -0.24
    åĬŁå¤«
    -0.23
     sag
    -0.23
    POSITIVE LOGITS
    夹
    0.28
    ä¸Ĭåīį
    0.26
    é¦ĸåħĪæĺ¯
    0.26
     acknow
    0.26
    spir
    0.25
    recogn
    0.25
    achu
    0.25
    (not
    0.25
    bell
    0.25
    URRE
    0.25
    Activations Density 0.312%

    No Known Activations