© 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. 33-TRANSCODER-HP
    7. 53521
    Prev
    Next
    INDEX
    Explanations

    as

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_33.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.33.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     comm
    -0.32
    hog
    -0.28
     prop
    -0.27
     ag
    -0.26
     ba
    -0.26
    年人
    -0.25
    åłĤ
    -0.25
    list
    -0.25
     kn
    -0.25
    æ²ī
    -0.24
    POSITIVE LOGITS
    devil
    0.29
     setattr
    0.27
     Nationals
    0.27
    æµģæ°ĵ
    0.26
    fbe
    0.26
    unread
    0.25
     Attribution
    0.24
    æĶ¹éĿ©åĪĽæĸ°
    0.24
    ody
    0.24
    篦
    0.24
    Activations Density 0.002%

    No Known Activations