© 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. 112799
    Prev
    Next
    INDEX
    Explanations

    Knowledge and awareness

    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
    rieve
    -0.29
    олÑĮ
    -0.28
    inders
    -0.27
     Intl
    -0.26
    ahn
    -0.26
    è¿·
    -0.26
     ignorance
    -0.25
     æĢ
    -0.25
    忱
    -0.25
    éĨĩ
    -0.24
    POSITIVE LOGITS
    çļĦ人æĿ¥è¯´
    0.30
    irez
    0.25
    pine
    0.24
    èIJ©
    0.24
    issy
    0.23
    istra
    0.23
    iado
    0.23
    iros
    0.23
    ä»ĸå·²ç»ı
    0.23
    æľĢ主è¦ģ
    0.23
    Activations Density 0.013%

    No Known Activations