© 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. 30-TRANSCODER-HP
    7. 95113
    Prev
    Next
    INDEX
    Explanations

    Descartes, Cartesian

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_30.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.30.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Negative Logits
    atten
    -0.30
    è°Ĵ
    -0.28
    ameda
    -0.26
     brag
    -0.25
     UNITY
    -0.24
    丼
    -0.23
    AllowAnonymous
    -0.23
    åħ¥åı£
    -0.23
    abbr
    -0.23
    对å¤ĸå¼ĢæĶ¾
    -0.23
    POSITIVE LOGITS
     Desc
    0.76
     Newton
    0.73
    笼
    0.68
    Newton
    0.67
    Cart
    0.66
     Cartesian
    0.64
    çīĽ
    0.63
     Cart
    0.63
    Desc
    0.56
     cart
    0.54
    Activations Density 0.920%

    No Known Activations