© 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. 7089
    Prev
    Next
    INDEX
    Explanations

    countries along with descriptions about their economic, demographic, or political status

    oai_token-act-pair · gpt-3.5-turbo
    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
    reads
    -0.91
     presentations
    -0.79
    frames
    -0.79
    abilities
    -0.79
     arrays
    -0.79
    mails
    -0.77
    bats
    -0.74
    inventoryQuantity
    -0.73
     suites
    -0.72
    bots
    -0.72
    POSITIVE LOGITS
     country
    1.82
     nation
    1.72
     superpower
    1.53
     continent
    1.41
     republic
    1.38
     province
    1.28
     countries
    1.28
     democracy
    1.27
     kingdom
    1.26
    country
    1.25
    Activations Density 0.236%

    No Known Activations