© 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. 30820
    Prev
    Next
    INDEX
    Explanations

    the word "who" and variations of it, indicating a focus on identifying individuals or experts

    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
    129
    +0.13
    0.6%
    700
    +0.13
    0.6%
    502
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    139
    +0.13
    0.10
    224
    +0.13
    0.07
    454
    +0.10
    0.06
    Negative Logits
    SPONSORED
    -0.69
    INA
    -0.64
     expiration
    -0.63
    Spread
    -0.62
     laughter
    -0.62
     mockery
    -0.61
     ANGEL
    -0.61
     Mechdragon
    -0.60
    Myth
    -0.60
     Clock
    -0.60
    POSITIVE LOGITS
     oversaw
    1.05
     oversees
    1.05
     supervised
    0.96
     heads
    0.93
     directs
    0.91
     helped
    0.87
     worked
    0.86
     chaired
    0.85
     represents
    0.85
     superv
    0.85
    Activations Density 0.546%

    No Known Activations