© 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. 107105
    Prev
    Next
    INDEX
    Explanations

    code and comments

    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
    宿
    -0.25
    persona
    -0.25
    æĬĹåĩ»çĸ«æĥħ
    -0.25
    (seed
    -0.24
    ngine
    -0.24
    rena
    -0.24
    èĸ°
    -0.23
     defence
    -0.23
    æĸ°æĶ¿
    -0.23
    .Environment
    -0.23
    POSITIVE LOGITS
     priorities
    0.34
     labels
    0.34
    :id
    0.32
     positions
    0.32
    NewItem
    0.30
    ä¼ĺåħĪ
    0.30
    å²Ĺä½į
    0.30
    éĩįçĤ¹é¡¹çĽ®
    0.30
    inals
    0.29
     sinks
    0.29
    Activations Density 1.589%

    No Known Activations