© 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. 64837
    Prev
    Next
    INDEX
    Explanations

    Article/blog post excerpts

    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
    åħħ满äºĨ
    -0.28
     tensors
    -0.27
     Ranked
    -0.26
    ีย
    -0.26
    ä¸ªåĽ½å®¶
    -0.25
    =cut
    -0.25
    .intent
    -0.25
    æįī
    -0.25
    çĸ«
    -0.25
    ç°¿
    -0.25
    POSITIVE LOGITS
     wont
    0.28
    ãĤªãĥ¼
    0.27
     desn
    0.27
    ousse
    0.26
     common
    0.25
    bsolute
    0.25
     reference
    0.25
    วา
    0.24
    _reference
    0.24
     Aval
    0.24
    Activations Density 49.078%

    No Known Activations