© 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. 8-RES_POST_32K-OAI
    11. 9012
    Prev
    Next
    INDEX
    Explanations

    instances of the word "the" and other related determiners or articles

    oai_token-act-pair · gpt-4o-mini
    Triggered 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
    Neuron Alignment
    Index
    Value
    % of L₁
    258
    +0.10
    0.5%
    23
    +0.10
    0.5%
    608
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    362
    +0.10
    0.17
    608
    +0.10
    0.16
    393
    +0.09
    0.11
    Negative Logits
    agus
    -0.77
    arest
    -0.68
     TABLE
    -0.64
    agues
    -0.64
    isode
    -0.64
    Contents
    -0.63
    iar
    -0.62
    wrong
    -0.61
    arten
    -0.60
    arter
    -0.60
    POSITIVE LOGITS
     fact
    1.50
     cumbers
    1.09
     sheer
    1.07
     unbeliev
    1.04
     lack
    0.98
     seeming
    0.96
     similarities
    0.94
     reluctance
    0.94
     insistence
    0.93
     tremendous
    0.92
    Activations Density 0.971%

    No Known Activations