© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-4B
    5. 19-TRANSCODER-HP
    7. 8755
    Prev
    Next
    INDEX
    Explanations

    capitalized letters inside words

    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
    åįıä½ľ
    -0.28
    æī¶æĮģ
    -0.26
    e
    -0.26
    çļĦä¸ĢåĪĩ
    -0.26
    iT
    -0.24
    ut
    -0.24
    uron
    -0.24
    ur
    -0.24
     sidelines
    -0.24
    IT
    -0.24
    POSITIVE LOGITS
    åĪ©çĽĬ
    0.27
    wash
    0.27
    porter
    0.25
    nings
    0.24
    uggle
    0.24
    inals
    0.24
    ANDING
    0.24
     WithEvents
    0.24
    åľ¨åĵª
    0.23
    åľ°éģĵ
    0.23
    Activations Density 0.094%

    No Known Activations