© 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. 96945
    Prev
    Next
    INDEX
    Explanations

    Let

    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
    åĪ
    -0.28
    ç¡®å®ļ
    -0.28
     Giz
    -0.26
    cı
    -0.26
    ç»Ħç»ĩçĶŁæ´»
    -0.26
    åĪĹ
    -0.25
    éĿ´
    -0.25
     구
    -0.25
     geil
    -0.25
    udu
    -0.24
    POSITIVE LOGITS
    OTT
    0.28
     introduced
    0.26
    ä¸įåłª
    0.26
    令
    0.26
    认æ¸ħ
    0.25
    terror
    0.25
    çļ±çº¹
    0.25
    ting
    0.24
    tbl
    0.24
    TableRow
    0.24
    Activations Density 0.907%

    No Known Activations