© 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. 32-TRANSCODER-HP
    7. 44256
    Prev
    Next
    INDEX
    Explanations

    High or low values

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_32.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.32.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Negative Logits
    lla
    -0.26
    caff
    -0.26
    åŀĽ
    -0.26
    -sales
    -0.26
    eyJ
    -0.26
    èĥ´
    -0.26
    ophile
    -0.24
    yas
    -0.23
    queeze
    -0.23
     McCabe
    -0.23
    POSITIVE LOGITS
    ä½İ
    0.35
    é«ĺ
    0.32
     low
    0.32
    low
    0.30
    enne
    0.29
     Low
    0.29
     lower
    0.28
     high
    0.28
    Low
    0.28
    é«ĺ强度
    0.28
    Activations Density 0.001%

    No Known Activations