© 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. 64875
    Prev
    Next
    INDEX
    Explanations

    they

    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
    åIJĪçIJĨæĢ§
    -0.30
     correctness
    -0.30
     orientation
    -0.27
    Norm
    -0.27
     norm
    -0.26
     tie
    -0.26
     Norm
    -0.26
    好äºĨ
    -0.25
    ª
    -0.25
    norm
    -0.24
    POSITIVE LOGITS
    iman
    0.28
    _DM
    0.27
     convertible
    0.27
    eya
    0.26
    illet
    0.26
    plies
    0.25
    Æ°ng
    0.25
    OrFail
    0.25
    arine
    0.25
    ãģ¼
    0.25
    Activations Density 0.496%

    No Known Activations