© 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. 8946
    Prev
    Next
    INDEX
    Explanations

    punctuation and sentence-ending markers indicating the conclusion of thoughts or statements

    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
    745
    +0.12
    0.6%
    161
    +0.10
    0.5%
    63
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    6
    +0.12
    0.14
    430
    +0.10
    0.12
    440
    +0.10
    0.09
    Negative Logits
    etry
    -0.64
     clitor
    -0.63
    iqueness
    -0.61
     pronouns
    -0.61
     leveling
    -0.59
    worms
    -0.57
    ledge
    -0.57
    nesia
    -0.57
    aterasu
    -0.56
     princip
    -0.56
    POSITIVE LOGITS
     Photo
    1.61
     PHOTO
    1.46
     REUTERS
    1.43
     Courtesy
    1.42
     Photograph
    1.39
     Image
    1.35
     Credit
    1.29
     ©
    1.27
     Picture
    1.26
     Photographer
    1.26
    Activations Density 0.849%

    No Known Activations