© 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. 7001
    Prev
    Next
    INDEX
    Explanations

    Code and blog content

    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
     works
    -0.32
    ally
    -0.31
     treat
    -0.30
    ently
    -0.28
     reason
    -0.28
    æĶ¯æı´
    -0.28
     instead
    -0.27
     statistics
    -0.27
     related
    -0.27
     [
    -0.27
    POSITIVE LOGITS
     dicks
    0.30
    -men
    0.27
    imitive
    0.27
    ardy
    0.26
    åıįä¹ĭ
    0.26
     millennials
    0.25
    uÃŁ
    0.25
    uien
    0.24
    Wow
    0.24
     LIMIT
    0.24
    Activations Density 0.010%

    No Known Activations