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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-4B
    5. 19-TRANSCODER-HP
    7. 6383
    Prev
    Next
    INDEX
    Explanations

    country

    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
    çļĦæĪIJæľ¬
    -0.25
    è´µ
    -0.25
     Career
    -0.24
     OCC
    -0.24
    ве
    -0.24
     Patt
    -0.23
    è²´
    -0.23
     báºŃc
    -0.23
    pressor
    -0.23
    贪婪
    -0.23
    POSITIVE LOGITS
    仪
    0.25
    æľŁ
    0.25
    伪è£ħ
    0.24
     Hayden
    0.24
     conception
    0.24
    èĽĭ
    0.24
    Mer
    0.23
    sym
    0.23
    >Edit
    0.23
    èģĶè°Ĭ
    0.23
    Activations Density 0.020%

    No Known Activations