© 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. 107016
    Prev
    Next
    INDEX
    Explanations

    deal

    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
    SOR
    -0.28
    产åĵģç»ıçIJĨ
    -0.26
    ä½ľåĩºäºĨ
    -0.25
    parator
    -0.25
     accuses
    -0.24
    æĶ¿åįı
    -0.24
    æľīäºĨ
    -0.24
    spar
    -0.24
    URY
    -0.23
    hue
    -0.23
    POSITIVE LOGITS
    PEND
    0.26
    cca
    0.26
    åıĸåĨ³
    0.25
    isia
    0.24
     Xin
    0.23
    -headed
    0.23
    nga
    0.23
     Qgs
    0.23
    å°ıåŃIJ
    0.23
    ugin
    0.22
    Activations Density 0.224%

    No Known Activations