© 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. 31-TRANSCODER-HP
    7. 18791
    Prev
    Next
    INDEX
    Explanations

    Names and publications

    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
    Not in Any Lists

    No Comments

    Negative Logits
    дов
    -0.29
    .va
    -0.27
    æݧèĤ¡
    -0.26
     trunk
    -0.26
    éĩįéĩı
    -0.26
    Ju
    -0.26
    é¾Ļ头
    -0.26
    满满
    -0.26
    æĸ¯é¡¿
    -0.25
     pudd
    -0.25
    POSITIVE LOGITS
    put
    0.28
    Put
    0.26
    orer
    0.26
    ates
    0.26
    çıŃ
    0.25
    legt
    0.25
    eyJ
    0.25
    踺
    0.25
    .Put
    0.25
     put
    0.25
    Activations Density 0.001%

    No Known Activations