© 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. OpenAI · Sparse Autoencoder for GPT2-Small - v5 2024
    5. GPT2-Small
    7. Residual Post - 32k
    9. 10-RES_POST_32K-OAI
    11. 29762
    Prev
    Next
    INDEX
    Explanations

    indicators related to economic performance and trends

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 10-res_post_32k-oai
    Configuration
    jbloom/GPT2-Small-OAI-v5-32k-resid-post-SAEs/v5_32k_layer_10.pt
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    32,768
    Data Type
    torch.float32
    Hook Name
    blocks.10.hook_resid_post
    Hook Layer
    10
    Architecture
    standard
    Context Size
    64
    Dataset
    Skylion007/openwebtext
    Activation Function
    topk
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    54
    +0.10
    0.5%
    684
    +0.10
    0.5%
    698
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    54
    +0.10
    0.06
    616
    +0.10
    0.06
    611
    +0.10
    0.06
    Negative Logits
    obos
    -0.71
    sama
    -0.70
    ModLoader
    -0.67
    agy
    -0.65
    bia
    -0.65
     Yugoslavia
    -0.64
    devices
    -0.62
    errors
    -0.61
    lihood
    -0.61
    icity
    -0.61
    POSITIVE LOGITS
     quarter
    1.07
     quarters
    0.99
     consecutive
    0.94
     calendar
    0.91
     period
    0.90
    financial
    0.87
     fiscal
    0.84
     financial
    0.83
     downturn
    0.81
     trough
    0.81
    Activations Density 0.694%

    No Known Activations