© 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. 109299
    Prev
    Next
    INDEX
    Explanations

    say punctuation after parenthesis

    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
     openings
    -0.25
     rehe
    -0.25
     reopen
    -0.25
     дал
    -0.25
    ÃŁen
    -0.25
     Babe
    -0.24
     Recommend
    -0.23
     reopening
    -0.23
    _FUN
    -0.23
    çIJĨæĥ³
    -0.23
    POSITIVE LOGITS
    à¹ĥà¸Ī
    0.29
    æ°ijæĦı
    0.26
     gest
    0.25
    æĸĭ
    0.25
    éĺĹ
    0.25
    eworld
    0.25
    romise
    0.25
    ä¸ĸ代
    0.24
    èİ·
    0.24
    urd
    0.24
    Activations Density 0.960%

    No Known Activations