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

    Neuronpedia

    Jacobian LensNEW
    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. 29017
    Prev
    Next
    INDEX
    Explanations

    mentions of the Affordable Care Act and its components, particularly related to Obamacare

    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
    0.4%
    762
    +0.10
    0.4%
    15
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    288
    +0.10
    0.03
    343
    +0.10
    0.02
    53
    +0.10
    0.02
    Negative Logits
    DragonMagazine
    -0.86
    athi
    -0.85
    tops
    -0.78
     CLSID
    -0.77
     Obi
    -0.76
    worldly
    -0.72
    angular
    -0.70
     Manit
    -0.69
     Conan
    -0.69
    ials
    -0.68
    POSITIVE LOGITS
     repeal
    1.16
     rollout
    0.91
    Care
    0.89
    amacare
    0.85
     regulations
    0.81
     repealed
    0.78
     enrollment
    0.78
     implementation
    0.77
     Amendments
    0.77
     enforcement
    0.75
    Activations Density 0.126%

    No Known Activations