© 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. 30797
    Prev
    Next
    INDEX
    Explanations

    mentions of government-related concepts and terms associated with subsidies

    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
    102
    +0.13
    0.6%
    314
    +0.13
    0.6%
    265
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    152
    +0.13
    0.08
    314
    +0.13
    0.01
    102
    +0.11
    0.06
    Negative Logits
     Deity
    -0.70
     jaws
    -0.66
     losers
    -0.65
     Survive
    -0.64
     disappearance
    -0.63
    ogenesis
    -0.63
    itiveness
    -0.62
     Graveyard
    -0.62
     Parenthood
    -0.61
     Rouge
    -0.60
    POSITIVE LOGITS
    issued
    1.57
    made
    1.55
    designed
    1.52
    approved
    1.48
    built
    1.44
    produced
    1.40
    provided
    1.39
    created
    1.37
    installed
    1.32
    licensed
    1.32
    Activations Density 0.279%

    No Known Activations