© 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. 9351
    Prev
    Next
    INDEX
    Explanations

    references, citations

    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
    Not in Any Lists

    No Comments

    Negative Logits
    ç½ij讯
    -0.29
    èģĶç½ij
    -0.29
    Extent
    -0.26
    von
    -0.25
    alone
    -0.24
    -alone
    -0.24
     dared
    -0.24
    ikal
    -0.24
     Koch
    -0.23
    +","+
    -0.23
    POSITIVE LOGITS
    æŶæĬ¥
    0.31
    è¿Ļä¸ĢçĤ¹
    0.28
    lok
    0.28
    #{
    0.27
     feedback
    0.27
    å±Ģéķ¿
    0.27
    åıįé¦Ī
    0.24
    jure
    0.24
    æѤæ¡Ī
    0.24
    split
    0.24
    Activations Density 0.029%

    No Known Activations