© 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. 12-TRANSCODER-HP
    7. 24251
    Prev
    Next
    INDEX
    Explanations

    antithesis

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_12.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.12.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Negative Logits
    -inverse
    -0.24
     iÅŁle
    -0.24
     ssh
    -0.24
    Congratulations
    -0.24
    exampleModal
    -0.23
     inve
    -0.23
    news
    -0.23
     war
    -0.23
    åĽĽå¹´
    -0.23
    QRS
    -0.23
    POSITIVE LOGITS
    -hot
    0.28
    åĮº
    0.26
    ption
    0.26
     Stark
    0.26
     ster
    0.26
    æļĤ
    0.25
    æĺİæ¸ħ
    0.25
    .receiver
    0.25
    ger
    0.25
     household
    0.25
    Activations Density 0.000%

    No Known Activations