© 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. 33-TRANSCODER-HP
    7. 54678
    Prev
    Next
    INDEX
    Explanations

    Pin

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_33.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.33.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Negative Logits
    orses
    -0.29
    erator
    -0.28
    iske
    -0.28
    å±±åİ¿
    -0.27
    Ħ件
    -0.27
     Titles
    -0.26
    ätze
    -0.25
    ificates
    -0.25
    urence
    -0.24
     pumps
    -0.24
    POSITIVE LOGITS
    碰
    0.30
    ä¸įåĪ°
    0.29
    åĬŁ
    0.28
    olith
    0.28
    rans
    0.27
    a
    0.27
    k
    0.27
    åIJ¬äºĨ
    0.26
    li
    0.26
     bullet
    0.26
    Activations Density 0.005%

    No Known Activations