© 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. 19431
    Prev
    Next
    INDEX
    Explanations

    political figures

    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
     protester
    -0.31
    æĶ¹éĿ©åĪĽæĸ°
    -0.30
    çĶ¨æĪ·æıIJä¾Ľ
    -0.30
    ket
    -0.26
    æ·±åĮĸæĶ¹éĿ©
    -0.26
    å¿Ĺ强
    -0.24
     Oaks
    -0.24
    ãģĦãģŁãģłãģijãĤĭ
    -0.24
     tiled
    -0.24
    æģ©
    -0.23
    POSITIVE LOGITS
    å¯Ĩ
    0.32
    åĪ¤
    0.29
    isu
    0.28
    ISC
    0.28
    AdminController
    0.27
    isphere
    0.26
     reg
    0.26
     rap
    0.26
    å¯Ħ
    0.25
    rap
    0.24
    Activations Density 0.957%

    No Known Activations