© 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. 22-TRANSCODER-HP
    7. 98893
    Prev
    Next
    INDEX
    Explanations

    excerpts from mixed sources

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_22.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.22.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     inventions
    -0.30
    èĥĮ
    -0.30
     invent
    -0.29
     projects
    -0.28
     configurations
    -0.27
     obscure
    -0.27
     rounds
    -0.27
     assignments
    -0.26
     baths
    -0.26
     combinations
    -0.25
    POSITIVE LOGITS
    ernes
    0.33
    æľīåĩłä¸ª
    0.32
    åħ©åĢĭ
    0.27
    两个
    0.27
    reeNode
    0.26
    ruta
    0.26
    è¿Ļ对
    0.26
    erne
    0.26
    两åıª
    0.26
    è¿Ļåĩłä¸ª
    0.25
    Activations Density 0.052%

    No Known Activations