© 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. 106991
    Prev
    Next
    INDEX
    Explanations

    Mixed code and text

    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
    lags
    -0.27
    æ¹¾åĮº
    -0.25
    å½ĴæĿ¥
    -0.25
    离
    -0.24
    ocity
    -0.24
    å²Ľ
    -0.24
    èĵĿ
    -0.24
    éĿŀåĩ¡
    -0.23
     mümkün
    -0.23
    åĪ°åľº
    -0.23
    POSITIVE LOGITS
     ref
    0.27
     bat
    0.27
     conting
    0.26
    订éĺħ
    0.25
    iger
    0.25
    ird
    0.25
     infinitely
    0.25
    irst
    0.24
    IRD
    0.24
     digitally
    0.24
    Activations Density 0.895%

    No Known Activations