© 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. 88187
    Prev
    Next
    INDEX
    Explanations
    No Explanations Found
    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
    KeyType
    -0.25
     fout
    -0.24
    åİĭå®ŀ
    -0.23
    æIJľæķij
    -0.23
    ç¾Ĭ
    -0.23
    ouples
    -0.23
    空æ°Ķè´¨éĩı
    -0.23
     Bücher
    -0.23
    åijĬè¯ī大家
    -0.23
    åģ¥åº·ç®¡çIJĨ
    -0.23
    POSITIVE LOGITS
    portunity
    0.27
     (++
    0.26
     Standing
    0.24
    gre
    0.24
    个å¤ļ
    0.24
     distraction
    0.24
     downfall
    0.23
    ãĥ¼ãĤ·ãĥ§ãĥ³
    0.23
    津贴
    0.23
    è¯Ľ
    0.23
    Activations Density 49.440%

    No Known Activations

    This feature has no known activations.