© 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. 32497
    Prev
    Next
    INDEX
    Explanations

    occurrences of mathematical or structured expressions

    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
    720
    +0.11
    0.5%
    314
    +0.10
    0.5%
    681
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    445
    +0.11
    0.01
    678
    +0.10
    0.01
    520
    +0.10
    0.01
    Negative Logits
    enegger
    -0.79
    cule
    -0.70
    onica
    -0.66
     balloons
    -0.65
     curated
    -0.63
     Orche
    -0.63
     Jobs
    -0.62
     balloon
    -0.61
    ierre
    -0.61
    hips
    -0.61
    POSITIVE LOGITS
    åħī
    0.71
    Wan
    0.70
    rendered
    0.67
    mint
    0.67
    AppData
    0.64
    speaking
    0.63
    Sales
    0.62
    import
    0.61
    practice
    0.60
    ×ŀ
    0.59
    Activations Density 0.079%

    No Known Activations