© 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. 35-TRANSCODER-HP
    7. 38833
    Prev
    Next
    INDEX
    Explanations

    two-letter codes

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_35.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.35.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Negative Logits
    å²±
    -0.30
     hy
    -0.28
     dav
    -0.28
     CRT
    -0.28
     tah
    -0.27
     de
    -0.27
     MST
    -0.27
     qx
    -0.27
     BSP
    -0.27
     a
    -0.27
    POSITIVE LOGITS
    ls
    0.37
    ln
    0.33
    ulin
    0.31
    IB
    0.31
    lb
    0.29
    IV
    0.28
    lm
    0.28
    ian
    0.28
    ID
    0.27
    iaux
    0.27
    Activations Density 42.935%

    No Known Activations