© 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. 64689
    Prev
    Next
    INDEX
    Explanations

    Beginning of texts

    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
    cly
    -0.28
    _FLAG
    -0.27
    èle
    -0.26
    é¼ĵåĬ±
    -0.26
     FLAG
    -0.26
     '::
    -0.26
    éķł
    -0.25
    klass
    -0.25
     trotz
    -0.24
    _Flag
    -0.24
    POSITIVE LOGITS
    京
    0.28
    èĥ¸åīį
    0.28
    ä¸Ģ身
    0.27
    OMEM
    0.26
    è°©
    0.25
    Paste
    0.25
    人身
    0.25
    atto
    0.24
    win
    0.24
     gum
    0.24
    Activations Density 6.848%

    No Known Activations