© 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. 28864
    Prev
    Next
    INDEX
    Explanations

    mentions of social events or gatherings

    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
    2
    +0.13
    0.6%
    541
    +0.09
    0.4%
    100
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    720
    +0.13
    0.10
    485
    +0.09
    0.07
    297
    +0.09
    0.06
    Negative Logits
    wcs
    -0.82
     largeDownload
    -0.81
    ibrary
    -0.80
    %:
    -0.78
    ahime
    -0.76
    Ô
    -0.76
    ãĤ¨ãĥ«
    -0.75
    \\\\\\\\
    -0.71
    aturday
    -0.71
    rencies
    -0.71
    POSITIVE LOGITS
     Mister
    1.21
     sir
    1.14
     comrade
    1.04
     Lieutenant
    1.04
     Doctor
    1.03
     darling
    1.03
     gentlemen
    1.02
     dear
    0.99
     Miss
    0.96
     Mr
    0.96
    Activations Density 0.331%

    No Known Activations