© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI 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. 27944
    Prev
    Next
    INDEX
    Explanations

    phrases that emphasize independence from external influences or sponsorship

    oai_token-act-pair · gpt-4o-miniTriggered 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
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    21
    +0.13
    0.6%
    655
    +0.10
    0.5%
    464
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    21
    +0.13
    0.05
    579
    +0.10
    0.05
    31
    +0.10
    0.04
    Negative Logits
    noticed
    -0.77
    nen
    -0.68
    ÄŁ
    -0.68
    adj
    -0.66
    few
    -0.64
     yawn
    -0.63
    very
    -0.62
    favorite
    -0.61
    gap
    -0.61
     puzzling
    -0.61
    POSITIVE LOGITS
     endorsement
    1.05
     endorse
    1.05
     endorsed
    1.00
     affiliated
    0.94
     warranty
    0.93
     condone
    0.93
     endorsements
    0.91
     guarantee
    0.89
     warranties
    0.87
     affiliation
    0.82
    Activations Density 0.950%

    No Known Activations