© 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. 4479
    Prev
    Next
    INDEX
    Explanations

    punctuation marks and their associated formatting in text

    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
    464
    +0.12
    0.6%
    615
    +0.09
    0.4%
    496
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    640
    +0.12
    0.20
    6
    +0.09
    0.19
    107
    +0.09
    0.13
    Negative Logits
    olulu
    -0.96
    bleacher
    -0.77
    monton
    -0.75
    rael
    -0.73
    trak
    -0.69
    ideo
    -0.69
    omer
    -0.68
    htaking
    -0.67
    renheit
    -0.67
    awks
    -0.67
    POSITIVE LOGITS
    Thank
    1.01
     THANK
    0.98
     Please
    0.96
    FAQ
    0.96
    please
    0.95
     PLEASE
    0.95
    Requirements
    0.93
    IRC
    0.93
     Thank
    0.92
    PLE
    0.90
    Activations Density 1.248%

    No Known Activations