© 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. 20-TRANSCODER-HP
    7. 108617
    Prev
    Next
    INDEX
    Explanations

    General descriptions and information

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_20.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.20.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Negative Logits
    ä¸ĢåĪĩéĥ½
    -0.39
    ä¸ĢåĪĩ
    -0.39
     everything
    -0.36
     Everything
    -0.35
    Everything
    -0.34
     tudo
    -0.32
    è¿ĻäºĽéĹ®é¢ĺ
    -0.31
     various
    -0.29
    everything
    -0.29
     Various
    -0.29
    POSITIVE LOGITS
    inc
    0.26
    @store
    0.25
    ç
    0.25
    bió
    0.25
    .charCodeAt
    0.24
    asticsearch
    0.24
     Smarty
    0.24
    ï¼Į请
    0.24
    ["$
    0.23
    FromArray
    0.23
    Activations Density 32.756%

    No Known Activations