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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 29679
    Prev
    Next
    INDEX
    Explanations

    phrases that indicate relationships and connections, particularly with "of" and "for" preceding other phrases

    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
    669
    +0.12
    0.5%
    608
    +0.10
    0.5%
    349
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    608
    +0.12
    0.14
    669
    +0.10
    0.13
    362
    +0.10
    0.12
    Negative Logits
    achy
    -0.77
    soever
    -0.75
    taboola
    -0.71
    asive
    -0.67
    hunt
    -0.66
    yout
    -0.66
    Translation
    -0.65
    etooth
    -0.65
    NVIDIA
    -0.64
    ebook
    -0.64
    POSITIVE LOGITS
     past
    1.34
     previous
    1.28
     earlier
    1.16
     old
    1.14
     predecessors
    1.13
     decades
    1.03
     eras
    1.01
     childhood
    1.00
    gone
    1.00
     predecessor
    0.99
    Activations Density 1.187%

    No Known Activations