© 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
    2. OpenAI · Sparse Autoencoder for GPT2-Small - v5 2024
    3. GPT2-Small
    4. Residual Post - 32k
    5. 8-RES_POST_32K-OAI
    6. 5294
    Prev
    Next
    INDEX
    Explanations

    phrases emphasizing political positioning and the role of the United States in democracy and global affairs

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_post_32k-oai
    Configuration
    jbloom/GPT2-Small-OAI-v5-32k-resid-post-SAEs/v5_32k_layer_8.pt
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    32,768
    Data Type
    torch.float32
    Hook Name
    blocks.8.hook_resid_post
    Hook Layer
    8
    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
    534
    +0.11
    0.5%
    406
    +0.11
    0.5%
    416
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    278
    +0.11
    0.07
    522
    +0.11
    0.07
    534
    +0.10
    0.05
    Negative Logits
     flares
    -0.71
     Marcos
    -0.70
     Denis
    -0.68
     Shad
    -0.68
     pseudonym
    -0.68
     routine
    -0.63
     Pengu
    -0.63
     Ezek
    -0.63
     Sergio
    -0.63
    çͰ
    -0.62
    POSITIVE LOGITS
     tolerant
    0.91
     democracies
    0.91
    atories
    0.89
     laughing
    0.89
     battleground
    0.88
     nation
    0.85
     bast
    0.85
     havens
    0.85
    ctuary
    0.83
     exporting
    0.81
    Activations Density 2.013%

    No Known Activations