© 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 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. 28651
    Prev
    Next
    INDEX
    Explanations

    references to astronomical objects and phenomena

    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
    586
    +0.12
    0.5%
    378
    +0.10
    0.4%
    203
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    314
    +0.12
    -0.00
    232
    +0.10
    0.09
    288
    +0.10
    0.10
    Negative Logits
     sts
    -0.72
     Hardcore
    -0.69
    autions
    -0.67
     Assassins
    -0.66
     defic
    -0.66
     Kens
    -0.66
    iquette
    -0.66
     lax
    -0.64
     LLP
    -0.64
     Lans
    -0.64
    POSITIVE LOGITS
    oplan
    1.10
     asteroid
    0.98
    osphere
    0.96
     Observatory
    0.96
     telescope
    0.96
     asteroids
    0.95
    Mars
    0.94
    auri
    0.94
     Nebula
    0.94
     Telescope
    0.94
    Activations Density 1.367%

    No Known Activations