© 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. 20-TRANSCODER-HP
    7. 64693
    Prev
    Next
    INDEX
    Explanations

    code

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_20.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.20.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Negative Logits
    eft
    -0.27
    ieves
    -0.27
    ors
    -0.26
    æĪij们ä¸Ģèµ·
    -0.26
    ews
    -0.25
    泺
    -0.24
    tag
    -0.24
     sinks
    -0.23
     Uph
    -0.23
    println
    -0.23
    POSITIVE LOGITS
    å¦Ĭ
    0.29
    SSERT
    0.29
    æĻºåķĨ
    0.26
    rlen
    0.26
    åį®
    0.26
    -muted
    0.25
    çļĦå®ŀéĻħ
    0.24
     '/'↵
    0.24
    inus
    0.23
    å¾Īå¤ļæŶåĢĻ
    0.23
    Activations Density 18.648%

    No Known Activations