© 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. 23-TRANSCODER-HP
    7. 25610
    Prev
    Next
    INDEX
    Explanations

    historical figures/rulers

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_23.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.23.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Negative Logits
    çļĦåĬŁèĥ½
    -0.27
     DST
    -0.27
    è¦ģåİ»
    -0.24
    èĩªæĿ¥
    -0.24
    ãĤŃãĥ¼
    -0.24
    æľŁæľĽ
    -0.24
    estro
    -0.24
    èħĪ
    -0.24
    æī©å®¹
    -0.24
    #error
    -0.24
    POSITIVE LOGITS
     royal
    0.29
    ports
    0.26
    Placeholder
    0.25
    çĻ¾å§ĵ
    0.25
    ÃŃn
    0.24
     nearest
    0.24
     snaps
    0.24
     biggest
    0.23
    herit
    0.23
     smallest
    0.23
    Activations Density 0.003%

    No Known Activations