© 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. 26389
    Prev
    Next
    INDEX
    Explanations

    references to specific historical events or notable public figures

    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
    362
    +0.10
    0.5%
    457
    +0.10
    0.4%
    631
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    362
    +0.10
    0.17
    608
    +0.10
    0.13
    385
    +0.09
    0.11
    Negative Logits
    eers
    -0.76
    reprene
    -0.74
     thresholds
    -0.71
     policymakers
    -0.69
     probabilities
    -0.68
     hurdle
    -0.67
    oother
    -0.65
     parity
    -0.64
     disadvantage
    -0.64
     economists
    -0.64
    POSITIVE LOGITS
     Trayvon
    0.99
     Watergate
    0.94
     Sandy
    0.92
     massacre
    0.91
     Benghazi
    0.91
     Sept
    0.91
     Rodney
    0.91
     Orlando
    0.90
     Ferguson
    0.90
    Sept
    0.89
    Activations Density 0.987%

    No Known Activations