© 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. 107158
    Prev
    Next
    INDEX
    Explanations

    School grades and classes

    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
     administrations
    -0.26
     Organizations
    -0.26
    ssl
    -0.26
     halves
    -0.25
    laps
    -0.25
     whe
    -0.24
    åѦåijĺ
    -0.24
    /plain
    -0.24
    езн
    -0.24
    æĪĸå¤ļæĪĸå°ij
    -0.24
    POSITIVE LOGITS
    pdev
    0.28
    evil
    0.26
    oola
    0.25
    æ¶ĪéĻ¤
    0.24
    åij½
    0.23
    tek
    0.23
    ött
    0.23
    .Mutable
    0.23
     Woody
    0.23
    erate
    0.23
    Activations Density 0.904%

    No Known Activations