© 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. 1-RES-JB
    11. 2249
    Prev
    Next
    INDEX
    Explanations

    instances of "Environmental Protection" agencies or acts

    oai_token-act-pair · gpt-3.5-turbo

    references to the concept of "protection" in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 1-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    1
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    plays
    -0.76
    hig
    -0.74
    chrom
    -0.74
    iewicz
    -0.71
    Leary
    -0.70
    bor
    -0.70
    ched
    -0.69
    ones
    -0.68
    ching
    -0.67
    ãĥ£
    -0.67
    POSITIVE LOGITS
     Protection
    1.41
     Enforcement
    0.91
     Protective
    0.90
     Policies
    0.88
     protection
    0.85
     Agency
    0.82
    osate
    0.81
     Enhancement
    0.81
     Protect
    0.80
     Assistance
    0.80
    Activations Density 0.009%

    No Known Activations