© 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. 19-TRANSCODER-HP
    7. 9507
    Prev
    Next
    INDEX
    Explanations

    like as

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_19.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.19.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Negative Logits
     optionally
    -0.28
    é«ĺä»·
    -0.27
    æĹ¢æĺ¯
    -0.27
    esub
    -0.25
    çªģçĦ¶
    -0.24
    tele
    -0.24
    stinence
    -0.24
    é«ĺé¢Ŀ
    -0.23
    cou
    -0.23
    åłªç§°
    -0.23
    POSITIVE LOGITS
    å¨ĥå¨ĥ
    0.26
    RARY
    0.26
    ëĭĺ
    0.26
     Bicycle
    0.25
    ILITY
    0.25
    pled
    0.25
     Narrative
    0.24
    Become
    0.24
    ä¸Ģ樣
    0.24
    çļĦèīºæľ¯
    0.24
    Activations Density 0.011%

    No Known Activations