© 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. 25-TRANSCODER-HP
    7. 109587
    Prev
    Next
    INDEX
    Explanations

    part

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_25.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.25.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Negative Logits
    æѦ
    -0.28
     Collective
    -0.28
     collective
    -0.28
    uft
    -0.26
    åĩ¿
    -0.26
     worse
    -0.26
     civilians
    -0.25
    强èĢħ
    -0.25
     Refuge
    -0.25
    çļĦæĺ¯
    -0.24
    POSITIVE LOGITS
    è¿Ļ款
    0.27
    -con
    0.27
    wise
    0.26
    Å¡ÃŃ
    0.25
    HI
    0.25
    è¿Ļ款车
    0.25
    leur
    0.25
     theirs
    0.24
     ways
    0.24
    utory
    0.24
    Activations Density 0.938%

    No Known Activations