© 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. 44010
    Prev
    Next
    INDEX
    Explanations

    code/website text

    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
    chie
    -0.30
     acknowledge
    -0.25
     effect
    -0.24
    åľ°åĿª
    -0.24
    ellar
    -0.24
     civil
    -0.24
    mun
    -0.23
    formik
    -0.23
    OVE
    -0.23
     sole
    -0.23
    POSITIVE LOGITS
    å¸Ń
    0.27
    eny
    0.25
    arton
    0.25
    MATRIX
    0.24
    å±Ģ
    0.24
    íĮħ
    0.24
    /board
    0.23
    ilon
    0.23
    åĪ©äºļ
    0.23
    phan
    0.23
    Activations Density 0.001%

    No Known Activations