© 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. 29441
    Prev
    Next
    INDEX
    Explanations

    comparisons and similarities between different subjects or entities

    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
    172
    +0.13
    0.6%
    296
    +0.11
    0.5%
    364
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    296
    +0.13
    0.04
    172
    +0.11
    0.04
    41
    +0.10
    0.04
    Negative Logits
    erity
    -0.88
    phant
    -0.69
     nonetheless
    -0.66
    advant
    -0.62
    é¾įåĸļ士
    -0.61
    Perhaps
    -0.61
    possibly
    -0.61
    few
    -0.60
    ãĥĥãĥī
    -0.60
    chance
    -0.59
    POSITIVE LOGITS
     above
    1.09
     previous
    0.87
     before
    0.85
     preceding
    0.80
     except
    0.79
     Above
    0.77
    above
    0.75
     described
    0.74
     below
    0.72
     pictured
    0.72
    Activations Density 0.528%

    No Known Activations