© 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. 19265
    Prev
    Next
    INDEX
    Explanations

    Code/technical documentation

    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
    æ·±
    -0.33
    æĽ´æ·±
    -0.32
    æ·±åħ¥
    -0.32
    å¾Īæ·±
    -0.32
    eced
    -0.32
     deeper
    -0.32
    深度
    -0.31
    åIJİ
    -0.30
     previous
    -0.30
     deep
    -0.30
    POSITIVE LOGITS
    ãģ¤ãĤĤãĤĬ
    0.31
    HUD
    0.28
    ä¹Į
    0.27
    imony
    0.26
     Bones
    0.25
    /loader
    0.25
    æ´»
    0.25
    atische
    0.24
    ulation
    0.24
     loaders
    0.24
    Activations Density 2.825%

    No Known Activations