© 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. 30576
    Prev
    Next
    INDEX
    Explanations

    patterns and special characters related to file paths or configurations

    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
    +0.12
    0.6%
    63
    +0.09
    0.4%
    458
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    345
    +0.12
    0.01
    314
    +0.09
    -0.01
    326
    +0.09
    0.03
    Negative Logits
    Lens
    -0.82
    CSS
    -0.76
    Ear
    -0.75
    enegger
    -0.75
     Noon
    -0.74
    andowski
    -0.73
    Square
    -0.72
     Fargo
    -0.72
    CLASSIFIED
    -0.71
    Tweet
    -0.71
    POSITIVE LOGITS
    supp
    0.88
    primary
    0.79
    asa
    0.74
    source
    0.69
    enium
    0.67
    prov
    0.62
    license
    0.61
    internal
    0.61
    server
    0.59
    usha
    0.59
    Activations Density 0.106%

    No Known Activations