© 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
    3. OpenAI · Sparse Autoencoder for GPT2-Small - v5 2024
    5. GPT2-Small
    7. Residual Post - 32k
    9. 8-RES_POST_32K-OAI
    11. 5531
    Prev
    Next
    INDEX
    Explanations

    references to prime ministers and government leadership

    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
    373
    +0.12
    0.6%
    203
    +0.11
    0.5%
    764
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    203
    +0.12
    0.03
    47
    +0.11
    0.04
    667
    +0.11
    0.03
    Negative Logits
    bleacher
    -0.67
     ASC
    -0.62
     Sigma
    -0.62
     brackets
    -0.61
     zip
    -0.60
     Delta
    -0.58
     gig
    -0.57
     caching
    -0.56
     wells
    -0.56
     lumber
    -0.56
    POSITIVE LOGITS
     Minister
    1.08
     secretary
    0.90
     minister
    0.90
     Secretary
    0.85
     Ministers
    0.83
     Cabinet
    0.79
     ministerial
    0.79
    bernatorial
    0.77
    Secretary
    0.76
    inator
    0.74
    Activations Density 0.127%

    No Known Activations