© 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. 8883
    Prev
    Next
    INDEX
    Explanations

    numerical data and statistics related to age and populations

    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
    409
    +0.12
    0.5%
    200
    +0.10
    0.4%
    296
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    223
    +0.12
    0.08
    314
    +0.10
    0.01
    693
    +0.10
    0.08
    Negative Logits
     spectators
    -0.75
     rods
    -0.71
     sparks
    -0.70
     spores
    -0.70
     hugs
    -0.70
     photoc
    -0.67
    çĶ°
    -0.66
     photons
    -0.66
    phot
    -0.65
     laughs
    -0.64
    POSITIVE LOGITS
     industrialized
    0.88
     Countries
    0.84
     regimes
    0.82
     largest
    0.81
    olicy
    0.80
     era
    0.79
     fiscal
    0.78
     countries
    0.78
     jurisdictions
    0.77
     flagship
    0.76
    Activations Density 0.783%

    No Known Activations