© 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. 31-TRANSCODER-HP
    7. 19875
    Prev
    Next
    INDEX
    Explanations

    Code or instructions

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_31.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.31.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Negative Logits
    дÑĥмал
    -0.29
    Õ§
    -0.26
    овал
    -0.26
    çł´åĿı
    -0.26
    çĶ¨æĪ·æıIJä¾Ľ
    -0.25
     was
    -0.25
    -provider
    -0.25
    æĿ¥äºĨ
    -0.25
    åIJĵ
    -0.25
    èĢĮæĿ¥
    -0.24
    POSITIVE LOGITS
     keeps
    0.33
     kept
    0.32
     keeping
    0.31
    Acts
    0.29
     Acts
    0.29
     acts
    0.29
    keep
    0.28
     keep
    0.28
    å·¥ä½ľçļĦ
    0.27
     Ton
    0.27
    Activations Density 1.370%

    No Known Activations