© 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. 26254
    Prev
    Next
    INDEX
    Explanations

    relational phrases indicating family connections

    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
    659
    +0.12
    0.5%
    121
    +0.11
    0.5%
    313
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    121
    +0.12
    0.08
    597
    +0.11
    0.05
    321
    +0.10
    0.04
    Negative Logits
    èĢħ
    -0.86
    £ı
    -0.84
    wcs
    -0.83
    soever
    -0.81
     guiActiveUn
    -0.78
    代
    -0.77
    ãĥ¥
    -0.75
    ancial
    -0.74
    è£ħ
    -0.73
    Shape
    -0.72
    POSITIVE LOGITS
     cousins
    1.25
     grandparents
    1.08
     sister
    1.06
     brother
    1.06
     sons
    1.05
     siblings
    1.03
     sisters
    1.03
     grandchildren
    1.02
     daughters
    1.02
     cousin
    1.02
    Activations Density 0.545%

    No Known Activations