© 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. 129123
    Prev
    Next
    INDEX
    Explanations

    diverse forum posts

    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
    GRP
    -0.31
    GES
    -0.28
    æĺ¯éŁ©åĽ½
    -0.26
    åĭŁ
    -0.25
     Majority
    -0.25
    åĬłçĽŁåºĹ
    -0.25
    QRST
    -0.25
    ç³ħ
    -0.25
     GK
    -0.24
    (cam
    -0.24
    POSITIVE LOGITS
     dere
    0.31
    éĢļ
    0.30
    least
    0.27
    ya
    0.25
    çªģ
    0.25
     unique
    0.25
    éĿŀçī©è´¨
    0.25
    ubby
    0.24
    utf
    0.24
    ç´ł
    0.24
    Activations Density 45.960%

    No Known Activations