© 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. 131120
    Prev
    Next
    INDEX
    Explanations

    Code-related text

    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
    Negative Logits
    olen
    -0.31
    òn
    -0.28
     DISPATCH
    -0.28
    anuts
    -0.27
    åįĥ
    -0.25
     upside
    -0.25
     }}↵↵
    -0.24
    aney
    -0.24
    phalt
    -0.24
    èĥ½åĬĽå¼º
    -0.24
    POSITIVE LOGITS
    ç°ª
    0.26
     goal
    0.25
    -looking
    0.25
    goal
    0.25
     observ
    0.24
    -facing
    0.24
     basal
    0.24
     hist
    0.24
    -formed
    0.23
    Even
    0.23
    Activations Density 0.008%

    No Known Activations