© 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. 17995
    Prev
    Next
    INDEX
    Explanations

    Arch

    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
     jit
    -0.29
    éģ¿
    -0.28
    èĮĦ
    -0.26
    до
    -0.26
    esi
    -0.26
    å¤ļä½į
    -0.26
    agination
    -0.25
     dma
    -0.25
    å¸ĥ
    -0.25
    eof
    -0.25
    POSITIVE LOGITS
     pancre
    0.28
    ppv
    0.25
    å®ģæĦ¿
    0.25
    ationally
    0.24
    elize
    0.23
    ï¿¥
    0.23
    izzlies
    0.23
    enerate
    0.23
    ounce
    0.23
    صرÙģ
    0.23
    Activations Density 0.021%

    No Known Activations