© 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. 23606
    Prev
    Next
    INDEX
    Explanations

    Common connecting words

    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
    enden
    -0.28
    icos
    -0.27
    encia
    -0.27
    æľ¬ç½ijç«Ļ
    -0.26
    otti
    -0.26
    weeney
    -0.25
    rom
    -0.25
    enary
    -0.25
    essional
    -0.24
    ROM
    -0.24
    POSITIVE LOGITS
    çľĭçľĭ
    0.69
     compare
    0.62
    æĿ¥çľĭçľĭ
    0.60
     Compare
    0.59
    Compare
    0.56
    compare
    0.56
     comparison
    0.52
    è¯ķè¯ķ
    0.52
     comparing
    0.52
     compares
    0.51
    Activations Density 0.004%

    No Known Activations