© 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. 108589
    Prev
    Next
    INDEX
    Explanations

    Actions (often -ing verbs)

    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
    ä¹Łå°Ĩ
    -0.30
    è¿ĺå°Ĩ
    -0.27
    å°±æĺ¯åĽłä¸º
    -0.25
    _so
    -0.25
    odesk
    -0.24
     suffice
    -0.24
    使åij½æĦŁ
    -0.24
    ÙĨØ©
    -0.24
     fingertips
    -0.24
    abay
    -0.24
    POSITIVE LOGITS
    强迫
    0.25
    çī¹æľĹ
    0.25
    ÑģпекÑĤ
    0.24
    æĶ
    0.23
    achu
    0.23
    è¯ķçĿĢ
    0.23
    æľ«
    0.23
    ."),
    0.23
    ...",
    0.23
    æľĪæľ«
    0.23
    Activations Density 5.821%

    No Known Activations