© 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. 140736
    Prev
    Next
    INDEX
    Explanations

    multiple, different options/quantities

    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
    è´¢åĬ¡ç®¡çIJĨ
    -0.28
    ãĤ«ãĥĨãĤ´ãĥª
    -0.26
    okable
    -0.26
    åħ¨çľģ
    -0.26
    æĸĩä½ĵ
    -0.25
     вок
    -0.25
    ä¸ĢéŨ
    -0.25
    ocity
    -0.25
    anio
    -0.24
    cox
    -0.24
    POSITIVE LOGITS
    éĻIJæľŁ
    0.29
     purs
    0.27
    atters
    0.26
     repe
    0.26
     outlier
    0.26
     Attr
    0.26
     bubble
    0.25
    at
    0.24
    ombre
    0.24
     marginal
    0.24
    Activations Density 77.466%

    No Known Activations