© 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. 10-RES-JB
    11. 3157
    Prev
    Next
    INDEX
    Explanations

    occurrences of the word "this" followed by descriptions of various things related to a country

    oai_token-act-pair · gpt-3.5-turbo

    references to the current state or conditions in the country

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 10-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    10
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    Examples
    -0.82
    icons
    -0.79
     Pwr
    -0.78
    nergy
    -0.74
    ickets
    -0.72
    acers
    -0.72
    apses
    -0.70
    rings
    -0.69
    phies
    -0.68
    Effects
    -0.68
    POSITIVE LOGITS
     country
    1.15
     hemisphere
    1.10
     century
    1.05
     regard
    0.96
     continent
    0.94
     particular
    0.93
     region
    0.93
     year
    0.91
     venerable
    0.89
     magnificent
    0.88
    Activations Density 0.135%

    No Known Activations