© 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. 30591
    Prev
    Next
    INDEX
    Explanations

    variations of the word "bro" and other casual or slang references

    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
    541
    +0.11
    0.5%
    516
    +0.10
    0.5%
    286
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    232
    +0.11
    0.03
    314
    +0.10
    0.00
    0
    +0.10
    0.02
    Negative Logits
    arily
    -0.71
     1886
    -0.68
    atorium
    -0.66
    heny
    -0.66
    places
    -0.62
     Luther
    -0.62
    ations
    -0.61
    hiba
    -0.61
     Wales
    -0.60
    nce
    -0.58
    POSITIVE LOGITS
    elled
    0.76
    xus
    0.72
    oding
    0.68
    aca
    0.68
    housing
    0.67
    ok
    0.67
    eller
    0.66
    thel
    0.66
    mares
    0.66
    ilit
    0.66
    Activations Density 0.147%

    No Known Activations