© 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. 31-TRANSCODER-HP
    7. 19878
    Prev
    Next
    INDEX
    Explanations

    file

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_31.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.31.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Negative Logits
    å¤©åľ°
    -0.32
    kip
    -0.32
    eturn
    -0.30
    çļĦç¨ĭ度
    -0.30
    缮çļĦ
    -0.27
    нд
    -0.27
    ,âĢ¦↵↵
    -0.26
    ooting
    -0.25
    apia
    -0.25
    vern
    -0.25
    POSITIVE LOGITS
     on
    0.41
     saved
    0.38
     formats
    0.36
     storing
    0.35
     saving
    0.34
     stored
    0.34
     format
    0.34
     saves
    0.33
    iez
    0.32
     holds
    0.31
    Activations Density 1.465%

    No Known Activations