© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW 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. 26436
    Prev
    Next
    INDEX
    Explanations

    mentions of the term "Shabbat" or related references to the Jewish Sabbath

    oai_token-act-pair · gpt-4o-mini
    Triggered 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
    Neuron Alignment
    Index
    Value
    % of L₁
    465
    +0.15
    0.7%
    299
    +0.10
    0.4%
    145
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    103
    +0.15
    0.02
    299
    +0.10
    0.02
    166
    +0.09
    0.02
    Negative Logits
    govtrack
    -0.70
     Lum
    -0.67
     Cheong
    -0.67
    stanbul
    -0.64
     Meridian
    -0.62
     Gutenberg
    -0.62
    oshenko
    -0.61
     territ
    -0.61
    lasses
    -0.60
    nces
    -0.59
    POSITIVE LOGITS
    atical
    1.08
    itt
    1.06
    inical
    1.04
    its
    1.02
    ucket
    1.01
    inic
    1.01
    alist
    0.94
    atar
    0.93
    ats
    0.91
    iah
    0.90
    Activations Density 0.256%

    No Known Activations