© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    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. 18268
    Prev
    Next
    INDEX
    Explanations

    instances of the word "King" with different contexts and scenarios

    oai_token-act-pair · gpt-3.5-turbo

    the repeated mention of "King" as a prominent reference in the text

    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
    recy
    -0.73
     vec
    -0.72
    edly
    -0.71
     recy
    -0.67
    iOS
    -0.65
     contam
    -0.64
     ticks
    -0.64
     notch
    -0.62
    dB
    -0.62
    omething
    -0.61
    POSITIVE LOGITS
     King
    3.64
    King
    2.95
     king
    2.30
     Queen
    2.10
     KING
    2.08
     Kings
    1.95
    Queen
    1.75
     kings
    1.72
     Prince
    1.64
    Kings
    1.62
    Activations Density 0.012%

    No Known Activations