© 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. 30-TRANSCODER-HP
    7. 96452
    Prev
    Next
    INDEX
    Explanations

    equality and fairness

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_30.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.30.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     decks
    -0.28
     stand
    -0.28
     Backbone
    -0.28
    带
    -0.28
     Pure
    -0.27
    纯
    -0.26
     deck
    -0.26
    çĥĪ
    -0.26
    ç¬Ķ
    -0.24
    atter
    -0.24
    POSITIVE LOGITS
     equality
    0.39
     promoting
    0.39
     fairness
    0.38
     promotion
    0.37
     empowerment
    0.34
     dignity
    0.33
     respect
    0.33
    Prom
    0.33
     collaboration
    0.33
     tolerance
    0.33
    Activations Density 0.887%

    No Known Activations