© 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 EvalsExportsinterp-engineNEWAPI 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. 31898
    Prev
    Next
    INDEX
    Explanations

    references to research materials and methodologies

    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
    373
    +0.14
    0.7%
    516
    +0.11
    0.5%
    118
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    516
    +0.14
    0.03
    435
    +0.11
    0.02
    450
    +0.11
    0.02
    Negative Logits
    partisan
    -0.72
    ardless
    -0.65
    govtrack
    -0.64
     Provincial
    -0.64
    ifax
    -0.64
     Rebell
    -0.62
     Mondays
    -0.62
     referendum
    -0.62
     Premier
    -0.61
    incial
    -0.60
    POSITIVE LOGITS
     embodiments
    1.24
     invention
    1.10
     FIG
    1.07
    ENTION
    1.03
     embodiment
    1.00
     drawings
    0.98
     herein
    0.96
    FIG
    0.96
     contemplated
    0.95
     preferred
    0.92
    Activations Density 0.307%

    No Known Activations