© 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. 8949
    Prev
    Next
    INDEX
    Explanations

    the word "few" and its variants, indicating a focus on quantity or scarcity

    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
    625
    +0.11
    0.5%
    71
    +0.10
    0.5%
    393
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    625
    +0.11
    0.04
    313
    +0.10
    0.04
    366
    +0.10
    0.04
    Negative Logits
    acus
    -0.81
    ESE
    -0.77
    yrinth
    -0.73
    webkit
    -0.72
    igation
    -0.72
    eus
    -0.71
    earch
    -0.70
    heed
    -0.70
    venture
    -0.69
    urden
    -0.68
    POSITIVE LOGITS
     dozen
    1.24
     hundred
    1.07
     weeks
    1.01
     thousand
    0.95
     minutes
    0.91
     bucks
    0.89
     days
    0.87
     paragraphs
    0.86
     moments
    0.86
     nights
    0.86
    Activations Density 0.087%

    No Known Activations