© 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. 8-RES_POST_32K-OAI
    11. 6036
    Prev
    Next
    INDEX
    Explanations

    numerical statistics and scores related to sports performances

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_post_32k-oai
    Configuration
    jbloom/GPT2-Small-OAI-v5-32k-resid-post-SAEs/v5_32k_layer_8.pt
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    32,768
    Data Type
    torch.float32
    Hook Name
    blocks.8.hook_resid_post
    Hook Layer
    8
    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
    373
    +0.14
    0.6%
    71
    +0.11
    0.5%
    365
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    693
    +0.14
    0.04
    260
    +0.11
    0.04
    238
    +0.10
    0.03
    Negative Logits
    enegger
    -0.71
     ens
    -0.70
     spotlight
    -0.63
     nomine
    -0.62
    lessly
    -0.62
     likeness
    -0.62
     skyrocket
    -0.61
     tease
    -0.59
     parted
    -0.59
     headquartered
    -0.58
    POSITIVE LOGITS
     occasions
    1.01
    slaught
    0.86
    anism
    0.82
     occasion
    0.81
    ct
    0.77
    ancock
    0.77
    perties
    0.69
    demand
    0.69
    arity
    0.68
    yx
    0.67
    Activations Density 0.171%

    No Known Activations