© 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. 86795
    Prev
    Next
    INDEX
    Explanations

    Code/technical writing snippets

    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
    åĴĮå¹³
    -0.28
    ä¸İæŶ
    -0.27
    olicy
    -0.26
    etur
    -0.26
    å½ĵä¹ĭ
    -0.26
    ucker
    -0.25
     flats
    -0.25
    otton
    -0.25
     Koch
    -0.24
    Fitness
    -0.24
    POSITIVE LOGITS
    sume
    0.27
     smoke
    0.25
    .Encode
    0.24
    ive
    0.24
     mart
    0.24
    æľºæŀĦ
    0.23
    MA
    0.23
    duk
    0.23
    #endregion
    0.23
    _GE
    0.23
    Activations Density 0.017%

    No Known Activations