© 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. 62763
    Prev
    Next
    INDEX
    Explanations

    understanding

    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
    æĪIJ为ä¸Ģ个
    -0.29
    pit
    -0.27
     aide
    -0.26
     achie
    -0.25
    mac
    -0.25
    åĪ©
    -0.25
    è¾ħ
    -0.24
    æįİ
    -0.24
     Achievement
    -0.24
     ‘
    -0.24
    POSITIVE LOGITS
    edException
    0.29
     rounds
    0.27
     unaware
    0.26
     Searching
    0.25
    ÑĮÑİ
    0.25
     Rounds
    0.24
    odyn
    0.24
    adol
    0.24
    edImage
    0.24
    weeks
    0.24
    Activations Density 15.599%

    No Known Activations