© 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. 29269
    Prev
    Next
    INDEX
    Explanations

    references to iconic video games and their elements

    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
    354
    +0.10
    0.5%
    683
    +0.09
    0.4%
    68
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    232
    +0.10
    0.08
    314
    +0.09
    -0.01
    659
    +0.09
    0.05
    Negative Logits
    */(
    -0.85
     Cosponsors
    -0.80
    umblr
    -0.77
     cataly
    -0.74
    bably
    -0.74
     fals
    -0.73
     porous
    -0.73
    ItemTracker
    -0.72
     improvised
    -0.71
     testim
    -0.71
    POSITIVE LOGITS
     Empires
    1.09
     Racer
    1.08
     Pinball
    1.05
     Skies
    1.05
     Simulator
    1.04
    ombat
    1.00
     Chronicles
    0.99
     Assault
    0.99
     Siege
    0.98
    Storm
    0.97
    Activations Density 0.744%

    No Known Activations