© 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. 88130
    Prev
    Next
    INDEX
    Explanations

    research publications

    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
    交代
    -0.28
    æ²īæµ¸åľ¨
    -0.27
    iei
    -0.26
    山路
    -0.26
    åijĬè¯ī她
    -0.26
    inea
    -0.25
    对æĪijæĿ¥è¯´
    -0.25
    æ¯ĶæĪij
    -0.25
    ue
    -0.24
    è°ķ
    -0.24
    POSITIVE LOGITS
    anol
    0.28
    ende
    0.27
    æīij
    0.27
    æĵ¦
    0.26
    ç»Ī
    0.26
     Wh
    0.25
     noses
    0.25
    ç®ĢåįķçļĦ
    0.25
    ovsky
    0.25
    aurus
    0.25
    Activations Density 4.598%

    No Known Activations