© 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. 9076
    Prev
    Next
    INDEX
    Explanations

    references to specific types of data and their quantities within scientific contexts

    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
    354
    +0.10
    0.5%
    341
    +0.10
    0.4%
    158
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    288
    +0.10
    0.13
    698
    +0.10
    0.06
    714
    +0.09
    0.05
    Negative Logits
    assetsadobe
    -0.77
     Sabha
    -0.77
    sis
    -0.76
    peer
    -0.75
    imeo
    -0.73
    IPS
    -0.70
    profits
    -0.69
     Availability
    -0.68
    oday
    -0.67
    availability
    -0.66
    POSITIVE LOGITS
     spoilers
    0.81
     elements
    0.79
     filler
    0.79
     remnants
    0.78
     surprises
    0.74
     goodies
    0.74
     guts
    0.71
     essentials
    0.68
     essential
    0.66
     redundant
    0.65
    Activations Density 2.176%

    No Known Activations