© 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. 31-TRANSCODER-HP
    7. 17976
    Prev
    Next
    INDEX
    Explanations

    Beginnings of sentences

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_31.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.31.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Negative Logits
    gard
    -0.30
    autoload
    -0.29
     DataLoader
    -0.28
    idge
    -0.28
    IMIZE
    -0.27
    omba
    -0.26
    æĦıè§ģåıįé¦Ī
    -0.25
     OTA
    -0.25
    andra
    -0.25
    gart
    -0.25
    POSITIVE LOGITS
    Republic
    0.26
    è§Ĩ
    0.25
    åħ¥åı£
    0.25
     Rousse
    0.24
    ures
    0.24
    źni
    0.24
    cie
    0.24
    uation
    0.24
    ophone
    0.24
    Ben
    0.24
    Activations Density 0.002%

    No Known Activations