© 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. 35-TRANSCODER-HP
    7. 40724
    Prev
    Next
    INDEX
    Explanations

    Code/file paths

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_35.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.35.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Negative Logits
     associated
    -0.27
    便æĺ¯
    -0.27
    åıĤä¸İåĪ°
    -0.26
     dedicated
    -0.25
     corresponding
    -0.25
    perienced
    -0.25
     mixed
    -0.24
    家åĽŃ
    -0.24
     resh
    -0.23
    zier
    -0.23
    POSITIVE LOGITS
    Ñģли
    0.33
    if
    0.30
    èĭ¥
    0.28
    åĩ»
    0.28
    ;if
    0.27
    ifr
    0.26
    _if
    0.25
    enza
    0.25
    iling
    0.25
    ifice
    0.25
    Activations Density 29.388%

    No Known Activations