© 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. 18-TRANSCODER-HP
    7. 89676
    Prev
    Next
    INDEX
    Explanations

    Proper nouns, legal documents

    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
    Not in Any Lists

    No Comments

    Negative Logits
    æľ¯è¯Ń
    -0.28
    term
    -0.27
     presentation
    -0.26
    gap
    -0.26
    mental
    -0.26
    ä¼ļéķ¿
    -0.25
    åľŁè±ª
    -0.25
    -cover
    -0.25
     polar
    -0.24
    ÑĤол
    -0.24
    POSITIVE LOGITS
     reels
    0.30
    unos
    0.27
     argued
    0.27
     compounded
    0.27
    交
    0.25
     UserType
    0.25
     translated
    0.24
    -US
    0.24
    æľīæĽ´å¤ļçļĦ
    0.23
    казан
    0.23
    Activations Density 11.702%

    No Known Activations