© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    5. GPT2-Small
    7. Residual Stream
    9. 0-RES-JB
    11. 13536
    Prev
    Next
    INDEX
    Explanations

    royalty-related terms, particularly focusing on queens

    oai_token-act-pair · gpt-3.5-turbo

    references to royalty and specifically the term "queen."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 0-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.hook_resid_pre
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    24,576
    Data Type
    torch.float32
    Hook Point
    blocks.0.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Trent
    -0.75
     Firearms
    -0.74
     Explos
    -0.72
     License
    -0.70
     Brady
    -0.70
     Technical
    -0.68
    stown
    -0.68
     Component
    -0.68
     Phil
    -0.68
     Distribut
    -0.67
    POSITIVE LOGITS
     queen
    3.74
     queens
    3.43
     princess
    1.54
    Queen
    1.53
     hive
    1.39
     brood
    1.34
     Queen
    1.28
     bride
    1.26
     Duchess
    1.13
     lady
    1.11
    Activations Density 0.039%

    No Known Activations