© 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. 19977
    Prev
    Next
    INDEX
    Explanations

    dry

    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
    翼
    -0.32
    æŀĦ
    -0.30
    WP
    -0.28
     Sark
    -0.28
    bis
    -0.27
    izador
    -0.27
     ech
    -0.27
    _ENCODING
    -0.27
     obs
    -0.26
    åij¼
    -0.26
    POSITIVE LOGITS
    ery
    0.31
     doGet
    0.29
    饿
    0.27
    á½²
    0.27
    íĭ´
    0.27
    aries
    0.27
    ody
    0.26
    atem
    0.26
    ova
    0.25
    symbols
    0.25
    Activations Density 0.002%

    No Known Activations