© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-4B
    5. 19-TRANSCODER-HP
    7. 7248
    Prev
    Next
    INDEX
    Explanations

    programming and multiple languages

    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
    .side
    -0.29
    gent
    -0.28
    åı³è¾¹
    -0.26
    fällig
    -0.26
    éĿ©æĸ°
    -0.26
    侧
    -0.26
    gam
    -0.26
    gradation
    -0.25
    å©¿
    -0.25
    ges
    -0.24
    POSITIVE LOGITS
     ausp
    0.28
    ersed
    0.27
    appy
    0.26
    LICENSE
    0.26
     seeded
    0.25
    ali
    0.25
    ussed
    0.25
     Necessary
    0.24
    cassert
    0.24
    å¿«ä¸ī
    0.24
    Activations Density 0.013%

    No Known Activations