© 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. 13311
    Prev
    Next
    INDEX
    Explanations

    references to a specific location, the "Oval Office."

    oai_token-act-pair · gpt-3.5-turbo

    references to the Oval Office

    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
     recognized
    -0.61
    elin
    -0.60
    dat
    -0.60
     bandits
    -0.59
     burd
    -0.59
     merchant
    -0.59
     ev
    -0.58
    ascal
    -0.58
     merchants
    -0.58
    less
    -0.58
    POSITIVE LOGITS
     Oval
    4.31
     oval
    1.93
     White
    1.24
     Cabinet
    1.18
    White
    1.08
     presidency
    0.98
     Presidency
    0.97
    OTUS
    0.94
     Medal
    0.92
     WHITE
    0.92
    Activations Density 0.023%

    No Known Activations