© 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. 96920
    Prev
    Next
    INDEX
    Explanations

    Code, licenses, legal documents, webinar

    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
    åľ°
    -0.28
    CHR
    -0.27
    å¿Ļ
    -0.26
    Up
    -0.25
     bapt
    -0.25
     spirit
    -0.25
     Hier
    -0.25
    äºĭåĬ¡
    -0.25
     spirits
    -0.24
     bang
    -0.24
    POSITIVE LOGITS
    arden
    0.28
    antz
    0.27
    arde
    0.27
    obar
    0.27
    itz
    0.26
    obi
    0.25
    ughty
    0.25
    foot
    0.25
    cope
    0.25
    èį
    0.25
    Activations Density 0.054%

    No Known Activations