© 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. 18-TRANSCODER-HP
    7. 90161
    Prev
    Next
    INDEX
    Explanations

    class

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_18.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.18.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Negative Logits
     currently
    -0.28
    icer
    -0.26
    ops
    -0.26
     '['
    -0.26
    Or
    -0.26
    Ok
    -0.26
    Op
    -0.25
    ëĵ¬
    -0.25
     pretend
    -0.25
    lit
    -0.25
    POSITIVE LOGITS
    @↵↵
    0.30
    .RequestParam
    0.27
    âĢ¦”↵↵
    0.27
    ipping
    0.26
    .separator
    0.26
    .setter
    0.26
    REEN
    0.25
    .’”↵↵
    0.24
     favour
    0.24
    -loader
    0.24
    Activations Density 5.196%

    No Known Activations