© 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. 8-RES_POST_32K-OAI
    11. 7862
    Prev
    Next
    INDEX
    Explanations

    special characters or formatting elements commonly used in language

    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
    218
    +0.10
    0.4%
    524
    +0.10
    0.4%
    550
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    189
    +0.10
    0.01
    148
    +0.10
    0.01
    314
    +0.09
    -0.00
    Negative Logits
     unification
    -0.78
     Cherokee
    -0.74
     Afric
    -0.73
    iage
    -0.73
    indal
    -0.71
     answ
    -0.68
     bund
    -0.66
     accomp
    -0.65
     stride
    -0.64
     contrasts
    -0.64
    POSITIVE LOGITS
    âĸĪ
    2.04
    âĸĪâĸĪ
    1.97
    âĸĦ
    1.90
    âĸĪâĸĪâĸĪâĸĪ
    1.80
    âĸĢ
    1.80
    âĸĪâĸĪâĸĪâĸĪâĸĪâĸĪâĸĪâĸĪ
    1.72
     âĸĪ
    1.62
    âĸĴ
    1.57
    âĸĵ
    1.53
    âĸijâĸij
    1.45
    Activations Density 0.040%

    No Known Activations