© 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. 18-TRANSCODER-HP
    7. 88206
    Prev
    Next
    INDEX
    Explanations

    Measurements, data, or technical context

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_18.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.18.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Negative Logits
    æݥ纳
    -0.26
    obili
    -0.24
    perimental
    -0.24
    chine
    -0.24
    'il
    -0.24
    blick
    -0.23
    'acc
    -0.23
    çŁ¥
    -0.23
     bq
    -0.23
    æĸ¹å½¢
    -0.23
    POSITIVE LOGITS
    ioned
    0.30
    é¢Ŀ头
    0.29
    å¥ĸ
    0.26
     curtain
    0.25
    åĶĩ
    0.25
    è¿°èģĮ
    0.25
    åĢĻéĢī人
    0.24
    invoke
    0.24
    uple
    0.24
    è¡Ķ
    0.24
    Activations Density 1.866%

    No Known Activations