© 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. 17998
    Prev
    Next
    INDEX
    Explanations

    websites and code. #This captures the general theme

    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
    esimal
    -0.32
    èµĸ
    -0.28
    oblin
    -0.27
     Dependencies
    -0.26
    dependencies
    -0.26
    .dependencies
    -0.26
    ç¨
    -0.25
    Dependencies
    -0.25
     dependencies
    -0.24
    èıĮ
    -0.24
    POSITIVE LOGITS
     cet
    0.30
     kind
    0.30
    kind
    0.30
    uty
    0.28
    vard
    0.28
    æ¤ĭ
    0.27
    áĢĺ
    0.27
    ibili
    0.26
    osit
    0.26
     Ut
    0.25
    Activations Density 3.636%

    No Known Activations