© 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. 22-TRANSCODER-HP
    7. 86898
    Prev
    Next
    INDEX
    Explanations

    Grammatical text/sentences

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_22.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.22.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Negative Logits
    æµij
    -0.27
     chosen
    -0.27
    æĹı
    -0.26
    roach
    -0.26
    æĹħ
    -0.25
    ph
    -0.25
     di
    -0.24
     argue
    -0.24
    |^
    -0.24
    |
    -0.24
    POSITIVE LOGITS
     differed
    0.30
    äºĨèĩªå·±çļĦ
    0.29
    æīĵäºĨ
    0.28
    åĩºæĿ¥äºĨ
    0.28
    äºĨåĩł
    0.27
    åIJ¬äºĨ
    0.27
    åıĤåĬłäºĨ
    0.26
    éĺħ读åħ¨æĸĩ
    0.26
    äºĨä¸Ģåľº
    0.26
    appeared
    0.26
    Activations Density 0.002%

    No Known Activations