© 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. 17834
    Prev
    Next
    INDEX
    Explanations

    Technical documents

    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
    弩
    -0.30
    arily
    -0.29
    æ¸ĹéĢı
    -0.28
     circ
    -0.27
    Broad
    -0.26
     GENERATED
    -0.26
    oret
    -0.26
     Buck
    -0.26
    ative
    -0.25
     segreg
    -0.25
    POSITIVE LOGITS
    åĺĪ
    0.28
    haus
    0.28
    ëłģ
    0.26
    tom
    0.26
    UME
    0.25
    pany
    0.25
    è´¼
    0.24
    èĪ±
    0.24
     sammen
    0.24
     nghiêm
    0.24
    Activations Density 0.806%

    No Known Activations