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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 11-RES-JB
    11. 18782
    Prev
    Next
    INDEX
    Explanations

    mentions of government officials and bureaucratic titles, particularly in the context of healthcare and law enforcement

    oai_token-act-pair · gpt-3.5-turbo

    references to government officials and health-related entities

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 11-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    11
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     mold
    -0.86
     colorful
    -0.85
     coloring
    -0.83
     favored
    -0.82
     rumors
    -0.80
     molded
    -0.78
     bicycl
    -0.77
     favorable
    -0.76
     rumor
    -0.75
     Sina
    -0.75
    POSITIVE LOGITS
     Scotland
    1.33
    Scotland
    1.28
     NHS
    1.17
     organisation
    1.10
    aviour
    1.09
     Wales
    1.08
    England
    1.07
     Yorkshire
    1.07
     Organisation
    1.06
     Parliamentary
    1.04
    Activations Density 0.384%

    No Known Activations