© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-4B
    5. 14-TRANSCODER-HP
    7. 40262
    Prev
    Next
    INDEX
    Explanations

    broad, informative, contextual descriptions

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_14.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.14.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Sox
    -0.26
    hus
    -0.24
     вообÑīе
    -0.24
     Incredible
    -0.24
    odia
    -0.24
     [].
    -0.23
    æŁĦ
    -0.23
     buff
    -0.23
     incom
    -0.23
     incent
    -0.23
    POSITIVE LOGITS
    Åijs
    0.27
    istik
    0.26
    éĥ½æľī
    0.26
    游æĪıä¸Ń
    0.25
     vão
    0.25
     digitally
    0.24
    éĥ½ä¼ļæľī
    0.24
    aggregate
    0.24
    enheim
    0.24
    主ä¹ī
    0.23
    Activations Density 1.807%

    No Known Activations