© 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. 107074
    Prev
    Next
    INDEX
    Explanations

    Greetings

    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
     tracks
    -0.29
     tracked
    -0.29
     track
    -0.29
    ellar
    -0.28
    iciel
    -0.27
    IClient
    -0.25
    track
    -0.25
    åĪ©çĶ¨æĤ¨çļĦ
    -0.25
     gap
    -0.24
    éŃ
    -0.24
    POSITIVE LOGITS
    亲çĪ±çļĦ
    0.30
    æĺ¥é£İ
    0.29
     dear
    0.28
    æĻ¯
    0.27
    亲çĪ±
    0.27
    åIJĮå¿Ĺ们
    0.25
    åĬłå¿«å»ºè®¾
    0.25
     downtown
    0.25
     Agenda
    0.25
     Dear
    0.25
    Activations Density 1.183%

    No Known Activations